Report Template - SME E

Transcription

E-COMmerce Proficient Analytics in Security and Sales for SMEs
D2.1 – REQUIREMENTS ANALYSIS
Contractual Delivery Date: M5 – May 2014
Actual Delivery Date: June 2014
Nature: Report
Version: 1.0
Public Deliverable
Abstract
This report reflects the collection activities and analysis conducted by SME-AGs and RTD performers
for identifying and elaborating on the user and data requirements of the SME E-COMPASS Project.
Within a series of interviews with SME-AGs members and e-commerce experts, dedicated focus
groups organisation and by implementing an extended electronic SME survey in four European
regions, the consortium validated and analysed trends, practices and requirements that SME-AGs
members active in on-line commerce face up. Through the accomplishment of this deliverable, SME
E-COMPASS Project has achieved to validate the hypotheses of the user requirements for micro,
small and medium enterprises active in e-commerce through dedicated and well-structured
activities that raised the awareness of the project to SMEs and highlighted the benefits that they
can gain via the evaluation, participation in pilot activities and continuous use of the project’s
applications after its completion.
Copyright by the SME E-COMPASS consortium, 2014-2015
SME E-COMPASS is a project co-funded by the European Commission within the 7th Framework Programme.
For more information on SME E-COMPASS, please visit http://www.sme-ecompass.eu/
DISCLAIMER
This document contains material, which is the copyright of the SME E-COMPASS consortium members and the European
Commission, and may not be reproduced or copied without permission, except as mandated by the European Commission
Grant Agreement no 315637 for reviewing and dissemination purposes.
The information contained in this document is provided by the copyright holders "as is" and any express or implied
warranties, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose are
disclaimed. In no event shall the members of the SME E-COMPASS collaboration, including the copyright holders, or the
European Commission be liable for any direct, indirect, incidental, special, exemplary, or consequential damages (including,
but not limited to, procurement of substitute goods or services; loss of use, data, or profits; or business interruption)
however caused and on any theory of liability, whether in contract, strict liability, or tort (including negligence or otherwise)
arising in any way out of the use of the information contained in this document, even if advised of the possibility of such
damage.

E-COMPASS
D2.1 –Requirements Analysis
Executive Summary
Abstract
The aim of this project is to develop two “software-as-a-service” web applications that provide European
SMEs active in e-commerce with the technological tools for strengthening their sustainability, increase their
customers’ trust in secure card not present transaction (CNP), position their e-shops in competitive
environments, and expand over new cross-border markets in Europe.
The integration task in the project is carried out by means of an RDF repository, which covers all required
data from multiple heterogeneous data sources. An OWL ontology will be used as mediated schema to
explicitly describe the data source semantics, providing developed services with a shared vocabulary for
their own specification, implementation and deployment.
In this regard, the main objective of this work package (WP2) “User and Data Requirements” is to garner
the expertise and knowledge of e-business specialists and entrepreneurs with respect to the challenges and
hazards in 24/7 transactions. A series of interviews with SME-AGs members and e-commerce experts, focus
groups organisation and implementation of an extended electronic survey for SMEs in four European
regions, have been vital in summarizing and analyzing valuable experience on e-fraud management and
identifying everyday practices for sales operations and digital marketing. These activities have allowed us to
shed light on the kind of data and tools European SMEs coming from different online markets use currently
and to determine the data parameters and the user requirements which are essential for the efficient
design and operation of the project’s “software-as-a-service” web applications.
From a technical point of view, this work package (WP2) aims additionally to produce a semantic model of
the domain represented by an OWL ontology. This ontology will be related with schemas of existing data
sets by means of mappings, enabling their integration in a common data model.
About this deliverable
This report summarizes the methodology followed in order to collect the user profile, current practices,
trends, challenges and SME requirements. Furthermore presents the results obtained in each SME-AG of
after analysing and processing them, and demonstrates region-specific and consolidated conclusions.
Furthermore, some official statistics on e-commerce economic activity in the project’s countries are
presented as well as specific figures regarding the regions where the project’s SME-AGs are located. Finally,
an initial version of the project’s semantic model is provided by offering an introductory description of the
first version of the SME E-COMPASS OWL ontology. The first integrated semantic model produced as an
OWL ontology is in progress and will be delivered as planned in D2.2 (Month 9). Specifically, this deliverable
is structured under the following sections:
1) Introduction
2) E-commerce economic activity in the countries involved in the project
3) Methodology for collecting user requirements
4) Results and analysis of the user data collection process and activities per SME-AG
5) User requirements and implications for the project’s applications
6) Study developed over existing data and the data that will be produced during the project
Grant Agreement 315637

PUBLIC
Page 4 of 255
E-COMPASS
Broadly, this report reflects the work accomplished in the first six months of the project in two of the four
tasks of WP2, namely: T2.1: User Requirement Analysis and T2.2: Model Design.
Achievements:
By the completion of the current deliverables the following sub-tasks and activities of the project have
been fulfilled:
1) Specific statistics and analyses for each SME-AG regarding e-commerce members profiling, user and
data requirements for online anti-fraud and data mining for e-sales applications.
2) Common statistical results and analyses for all SME-AGs regarding e-commerce members profiling,
user and data requirements for online anti-fraud and data mining for e-sales applications.
3) Functional/Non-functional requirements extraction, Data Base common features collection and
specification.
4) Semantic data model: Initial version of the OWL Ontology.
5) Consolidated list of interested SMEs to participate as pilot users, aiming to collaborate with the
project by providing data and validating the developed prototypes.
6) Supporting material for information/data collection such as online questionnaires, slides and guides
for focus groups and interviews, project’s communication material and dissemination of the WP2
activities.
Relation with other deliverable(s)
D1.1 - SME-E-COMPASS Methodological Framework-v1.0 (necessary/required reading)

PUBLIC
Page 5 of 255
E-COMPASS
Table of Contents
EXECUTIVE SUMMARY
TABLE OF CONTENTS
TABLE OF FIGURES
TABLE OF TABLES

1
INTRODUCTION

2
E-COMMERCE ECONOMIC ACTIVITY IN THE PROJECT COUNTRIES
2.1
Greece .................................................................................................................................. 15
2.1.1 Statistics on e-commerce activities in Greece
2.1.2 Statistics for e-fraud and data mining tools in Greece
2.1.3 Corinth Prefecture (Greece) and Corinthia Chamber of Commerce 
2.2
United Kingdom (UK) ............................................................................................................. 21
2.2.1 Statistics on e-commerce activities
2.2.2 Specific figures statistics for e-fraud and data mining tools
2.2.3 Halton Chamber of Commerce and Enterprise 
2.3
Spain (Basque Country and Valencia) ..................................................................................... 31
2.3.1 Statistics on e-commerce activities in Spain
2.3.2 GAIA & ATEVAL SME-AGs

3
METHODOLOGY FOR COLLECTING USER REQUIREMENT

4
RESULTS OF THE USER DATA COLLECTION PROCESS
4.1
Results and Analyses from Greek Chamber (EPK) ................................................................... 42
4.1.1 Organisation of SME E-COMPASS INFODAY in Corinth region by EPK, EXUS & ETR
4.1.2 Organisation of e-Survey in Corinth prefecture by EPK
4.1.3 Focus group and interviews organised in EPK
4.2
Results and Analyses from British Chamber (HALTON) ............................................................ 60
4.2.1 Organisation of Infoday and Focus Groups by HALTON and FRA 
4.2.2 Organisation of e-Survey in Halton region by HALTON and FRA
4.2.3 Results of Focus Groups and Interviews by HALTON and FRA
4.3
Results and Analyses from Spanish SME Association GAIA ...................................................... 86
4.3.1 Organisation of e-Survey and Focus Groups by GAIA and CIC
4.3.2 Results of e-Survey by GAIA and CIC
4.3.3 Results of Focus Groups and Interviews by GAIA and CIC 
4.4
Results and Analyses from Spanish Chamber (ATEVAL) ......................................................... 101
4.4.1 Organisation of e-Survey and Focus Groups by ATEVAL and UMA 
4.4.2 Results of e-Survey by ATEVAL and UMA

PUBLIC
Page 6 of 255
E-COMPASS
Results of Focus Groups and Interviews by ATEVAL and UMA
4.4.3
4.5
Common Results and Analyses............................................................................................. 116

5
USER REQUIREMENTS IMPLICATIONS TO THE PROJECT APPLICATIONS
5.1
5.2
Online fraud detection ........................................................................................................ 122
Data mining for e-sales operations....................................................................................... 124

6
EXISTING DATA USED BY PROJECTS PILOT SMES AND THEIR COLLECTION FOR FUTURE ANALYSIS
6.1
Existing Data Used by SME................................................................................................... 128
6.1.1 Anti-fraud
6.1.2 Data mining
6.2
Companies providing data and Pilot Users ........................................................................... 133
6.3
Data that will be produced in the Project ............................................................................. 139
6.3.1 Anti-fraud
6.3.2 Data mining
6.4
Semantic Data Model Initial Proposal .................................................................................. 141
EPILOGUE ....................................................................................................................................... 148

7
REFERENCES
8
APPENDIX
8.1
Supporting Material ............................................................................................................ 151
8.1.1 Online questionnaire for Greek e-commerce SMEs
D2.1 SME E-COMPASS. QUESTIONNAIRE FOR GREEK E-COMMERCE SMES V1.3
8.1.2
Online questionnaire for Spanish e-commerce SMEs
D2.1 E-COMPASS CUESTIONARIO. SPANISH V 1.3
8.1.3
Online questionnaire for British e-commerce SMEs
D2.1 E-COMPASS. EXTENDED SURVEY QUESTIONNAIRE – ENGLISH V 3.3
8.1.4
Focus Group questionnaire guide for fraud in e-commerce
8.1.5
Presentation slides: Corinth Infoday Project Presentation
8.1.6
Presentation slides: Corinth Infoday Press Release 
8.1.7
Presentation slides: Corinth Infoday Express of Interest Form for SMEs 
8.1.8
Presentation slides: Corinth Infoday Anti-Fraud Technologies Presentation
8.1.9
Presentation slides: Corinth Infoday Press Release 
8.1.10
Presentation slides: Halton Infoday/Interviews Online Data Mining Info Collection 
8.1.11
Extended questionnaire: GAIA - CIC Infoday/Interviews Info Collection
8.1.12
Data Base scheme: Core version of products diagram (CIC) 
8.1.13
Presentation slides: ATEVAL Infoday/Interviews Online Data Mining Info Collection
8.1.14
Presentation slides: ATEVAL Infoday/Interviews Anti-fraud Application Info Collection

PUBLIC
Page 7 of 255
E-COMPASS
Table of Figures
Figure 1. Total e-commerce sales .................................................................................................................... 15
Figure 2. Total e-commerce growth rate ........................................................................................................ 15
Figure 3. Agree or disagree with the statement: “You feel confident purchasing goods or services via the
Internet from national retailers / providers .................................................................................................... 19
Figure 4. Agree or disagree with the statement: “You feel confident purchasing goods or services via the
Internet from retailers / providers from another EU country” ....................................................................... 19
Figure 5. Percentage of Internet Users (ONS 2014) ........................................................................................ 21
Figure 6. Total e-commerce sales (ONS 2012) ................................................................................................ 22
Figure 7. Types of Goods and Services bought online in the UK in % of Individuals, 2011 (yStats.com 2012)24
Figure 8. Purchases of certain types of products made online by age group (2013) (ONS 2013) .................. 24
Figure 9. Value of e-commerce sales over a website by size of business (# of employees), 2012 (ONS 2012)
......................................................................................................................................................................... 25
Figure 10. Size of Business by e-Commerce Revenue (Khan 2013) ................................................................. 25
Figure 11. UK Age Structure of Mobile Shoppers, 2012, Percentage of total Population (Ecommerce Europe
2013) ................................................................................................................................................................ 26
Figure 12. Payment Methods (Payvision 2012) ............................................................................................... 26
Figure 13. Time taken to review orders (Khan 2013) ...................................................................................... 27
Figure 14. Proportion of businesses using social media, 2012 (ONS 2012) .................................................... 29
Figure 15. Quantity of B2C E-Commerce (€ million per year) ......................................................................... 32
Figure 16. People who bought online in the last 12 months in 2013 (% of people aged 16 to 74 years)....... 33
Figure 17. Most purchased products over the internet by Valencians ........................................................... 34
Figure 18. Preferred online channels by Valencians ....................................................................................... 34
Figure 19. Infoday Panel: Mrs. Elena Spyropoulou (FP7 SME National Representative) on the podium, panel
members from the left: Mrs Vasilis Nanopoulos (President of EPK), Mr. Orestis Papadopoulos (ETR), Dr
Nikolaos Thomaidis (EXUS) .............................................................................................................................. 43
Figure 20. Corinth Chamber of Commerce SME members attending SME E-COMPASS Infoday .................. 44
Figure 21. Corinth Chamber of Commerce SME members attending SME E-COMPASS Infoday ................... 44
Figure 22. Interviews from local TV channels were taken to (from left) i. Mrs Elena Avatangelou, Project
Coordinator (EXUS), ii. Mr Orestis Papadopoulos, Fraud Specialist (ETR), iii. Mr Panagiotis Gezerlis General
Director of GRECA (invited speaker) ............................................................................................................... 44
Figure 23. The cover pages of the folding invitation ....................................................................................... 45
Figure 24. The inner pages of the folding invitation ....................................................................................... 46
Figure 25. Position of the Responders in the SME in Corinth ......................................................................... 47
Figure 26. Types of Products and Services offered online in Corinth ............................................................. 48
Figure 27. e-Commerce sectors in Corinth ...................................................................................................... 48
Figure 28. e-Commerce personnel in Corinth ................................................................................................. 49
Figure 29. Annual online revenue in Corinth................................................................................................... 49
Figure 30. Annual volume of orders in Corinth ............................................................................................... 50
Figure 31. Years of online business ................................................................................................................. 50
Figure 32. Sources of price comparison .......................................................................................................... 51
Figure 33. Frequency of price comparison among Corinth e-shops ............................................................... 51

PUBLIC
Page 8 of 255
E-COMPASS
Figure 34. Pricing adjustment in Corinth e-shops ........................................................................................... 52
Figure 35. Interest of online SMEs in Corinth for pricing optimisation application ........................................ 52
Figure 36. Use of customer behaviour analysis, churn prediction and/or prevention tool(s) in Corinth ....... 53
Figure 37. Interest of online SMEs in Corinth for customer behaviour analysis application .......................... 53
Figure 38. Interest of online SMEs in Corinth for an application that analyses the behaviour of groups of
customers, discovers habits, and detects new e-shopping tendencies .......................................................... 54
Figure 39. Which online actions fraud threat can possibly force Corinth SMEs to abandon? ........................ 54
Figure 40. Which online activities can fraud threat force Corinth SMEs to abandon .................................... 55
Figure 41. Actions taken from online SMEs for their fraud protection ........................................................... 55
Figure 42. Parameters of a transaction that SMEs (manually or with an assessment tool) take into
consideration or collect for the validation of an e-payment .......................................................................... 56
Figure 43. E-shops supported languages in Corinth ........................................................................................ 56
Figure 45. Time line schedule for collection information from Halton companies......................................... 61
Figure 46. Focus groups info days in Halton.................................................................................................... 62
Figure 47. Question 1) in online questionnaire. English version for HALTON ................................................. 64
Figure 54. Question 10) in online questionnaire. English version for HALTON ............................................... 68
Figure 60. Question 13a) in online questionnaire. English version for HALTON ............................................. 71
Figure 61. Question 13b) in online questionnaire. English version for HALTON............................................. 72
Figure 62. Question 13c) in online questionnaire. English version for HALTON ............................................. 72
Figure 63. Question 13d) in online questionnaire. English version for HALTON............................................. 73
Figure 64. Question 13e) in online questionnaire. English version for HALTON ............................................. 73

PUBLIC
Page 9 of 255
E-COMPASS
Figure 77. Question 1) in online questionnaire. Spanish version for GAIA ..................................................... 87
Figure 86. Question 10) in online questionnaire. Spanish version for GAIA ................................................... 90
Figure 90.Question 14) in online questionnaire. Spanish version for GAIA .................................................... 92
Figure 95. Focus group informative sessions and interviews in ATEVAL ...................................................... 101
Figure 96. Info Flyer for real data collection from e-shop’s owners (Google Analytics and Piwik reports, etc.)
....................................................................................................................................................................... 102
Figure 97. Question 1) in online questionnaire. Spanish version for ATEVAL ............................................... 103
Figure 100. Question 4) in online questionnaire. Spanish version for ATEVAL ............................................. 104
Figure 106. Question 10) in online questionnaire. Spanish version for ATEVAL ........................................... 106
Figure 115. Annual revenue from online sales in 2013 for all the studied SMEs .......................................... 116
Figure 116. Question 8), general results from all questionnaires ................................................................. 117
Figure 117. Price adjustment method of all SMEs ........................................................................................ 118
Figure 118. Customer behaviour analysis for all SMEs.................................................................................. 119
Figure 119. How do you deal with online payment fraud? ........................................................................... 120

PUBLIC
Page 10 of 255
E-COMPASS
Figure 120. General scheme of the complete ontology ................................................................................ 142
Figure 121. Classes, object properties and attributes for modelling e-shops and e-shop owners ............... 143
Figure 122. Classes, object properties and attributes for modelling visitors ................................................ 144
Figure 123. Address subclasses contemplated by the ontology ................................................................... 145
Figure 124. Class modeling a location ........................................................................................................... 145
Figure 125. Class Device in the ontology ....................................................................................................... 146
Figure 126. Visit modeling in the ontology.................................................................................................... 146
Figure 127. Transaction class in the ontology ............................................................................................... 147
Figure 128. Class Page in the ontology .......................................................................................................... 148
Figure 129. Data Base gathering of responses from online questionnaire, Greek version........................... 158
Figure 130. Data Base gathering of responses from online questionnaire, Spanish version ........................ 163
Figure 131. Data Base gathering of responses from online questionnaire, extended English version ......... 177
Table of Tables
Table 1. Top 10 product categories in Valencia e commerce ......................................................................... 34
Table 2. SME e-compass survey questionnaire: Common questions to all versions ...................................... 40
Table 3. SME e-compass Focus Groups question categories: Anti-fraud ........................................................ 41
Table 4. SME e-compass Focus Groups question categories: Data Mining for e-sales ................................... 41
Table 5. Responses from EPK concerning current state in web analytics: visitor behaviour .......................... 57
Table 6. Responses from EPK concerning current state in web analytics: competitor’s analysis ................... 58
Table 7. Reponses from EPK concerning new requirements for web analytics: visitor behaviour ................. 59
Table 8. Responses from EPK concerning new requirements for web analytics: competitor’s analysis ........ 59
Table 9. Pilot users willingness in EPK ............................................................................................................. 59
Table 5. Responses from HALTON concerning current state in web analytics: visitor behaviour .................. 82
Table 6. Responses from HALTON concerning current state in web analytics: competitor’s analysis ........... 82
Table 7. Reponses from HALTON concerning new requirements for web analytics: visitor behaviour ......... 83
Table 8. Responses from HALTON concerning new requirements for web analytics: competitor’s analysis . 83
Table 9. Pilot users in HALTON ........................................................................................................................ 83
Table 10. Traceability matrix (classified information and related variables) .................................................. 86
Table 11. Responses from GAIA concerning current state in web analytics: visitor behaviour ...................... 95
Table 12. Responses from GAIA concerning current state in web analytics: competitor’s analysis ............... 95
Table 13. Reponses from GAIA concerning new requirements for web analytics: visitor behaviour ............. 96
Table 14. Responses from GAIA concerning new requirements for web analytics: competitor’s analysis .... 96
Table 15. Pilot users in GAIA............................................................................................................................ 96
Table 16. Responses from GAIA concerning the current state in fraud detection ......................................... 97
Table 17. Reponses from GAIA concerning the current state in fraud detection ........................................... 97
Table 18. Reponses from GAIA concerning the efficiency of the overall fraud management process ........... 98
Table 19. Responses from GAIA concerning new requirements for Anti-fraud .............................................. 98
Table 20. Responses from ATEVAL concerning the current state in web analytics: visitor behaviour ......... 111
Table 21. Responses from ATEVAL concerning the current state in web analytics: competitor’s analysis .. 111

PUBLIC
Page 11 of 255
E-COMPASS
Table 22. Responses from ATEVAL concerning new requirements for web analytics: visitor behaviour ..... 112
Table 23. Responses from ATEVAL concerning new requirements for web analytics: competitor’s analysis
....................................................................................................................................................................... 112
Table 24. Responses from ATEVAL concerning requirements from automatic actions ................................ 113
Table 25. ATEVAL pilot users ......................................................................................................................... 113
Table 26. Functional requirements for data mining (web analytics) applications ........................................ 124
Table 27. Dimensions for data mining e-sales operations ............................................................................ 125
Table 28. Non-Functional requirements for data mining (web analytics) applications ................................ 126
Table 29. Data parameters and qualitative characteristics used by selected SMEs ..................................... 128
Table 30. Existing data used for web analytics .............................................................................................. 131
Table 31. Pilot e-shop's technical features .................................................................................................... 133
Table 32. Pilot e-shop's descriptions I ........................................................................................................... 134
Table 33. Pilot e-shop's descriptions II .......................................................................................................... 135
Table 34. Pilot e-shop's descriptions III ......................................................................................................... 136
Table 35. Pilot e-shop's descriptions IV ......................................................................................................... 137
Table 36. Pilot e-shop's descriptions V .......................................................................................................... 138
Table 37. Pilot e-shop's descriptions VI ......................................................................................................... 139
Table 34. Competitors’ analysis: data have to be automatically generated ................................................. 140

PUBLIC
Page 12 of 255
E-COMPASS
1
Introduction
The main objective of the second Work Package (WP2) “User and Data Requirements” is to gather
requirements and operational specifications from e-business specialists and SME entrepreneurs with
respect to the challenges and hazards in 24/7 transactions. In a series of activities such as interviews with
online SMEs, organisation of dedicated focus groups and electronic survey, was attempted to collect
valuable experiences from past fraudulent transactions, current anti-fraud practices and to identify
everyday operations and tasks through which SMEs manage their sales and digital activities. From a
technical point of view, this Work Package aims also to develop a semantic model of the domain as an OWL
ontology. This ontology will be related to schemas of existing data sets by means of mappings, enabling
their integration in a common data model.
About this deliverable
This report presents the methodology followed in order to collect the user and data requirements, the
quantitative and qualitative results and needs obtained in each SME-AG of the consortium, and concludes
with region-specific and consolidated analysis of the user and data requirements. Furthermore, selected
official statistics on e-commerce economic activity in the project countries are presented in order to
capture the macro environment of online SMEs in each region of the project.
This report reflects the work accomplished in the first six months of the project in two out of the four tasks
of WP2, namely:
T2.1: User Requirement Analysis
In this task, RDT partners and SME-AGs define the methodology for collecting user and data requirements.
SME-AGs with RTD support contact their associate companies in order to compile their needs and specific
challenges in web-transactions, security and sales operations. This process has been developed in four
steps: company profiling and clustering, basic information collection, technical information collection and
obtaining a commitment from companies to collaborate and participate in the project. Furthermore, RDT
partners and SME-AGs managers interpret the results and draw conclusions for the
T2.2: Model Design
Based on the analysis developed in T2.1, a very initial version of the OWL ontology that can cope with the
data representation requirements of the project is being defined. RTD partners have met over a two day
period, in an ontology workshop, in order to agree on the basic structure of the project ontology. The
complete OWL ontology and the corresponding database mappings will be reported in D2.2 at September
2014.
Document structure
This report follows a structure based on the work effort performed according to the aforementioned tasks.
Section 2 “E-commerce economic activity in the Project Countries” presents selected official statistics about
e-commerce economic activity in the project’s countries and regions. The section includes statistics on e-

PUBLIC
Page 13 of 255
E-COMPASS
commerce activities per region and per country, specific figures on e-fraud and data mining tools per
country and the profiling and motivations of each SME association participating in the project.
Section 3 “Methodology for collecting user requirement” reports the general methodology for collecting
user requirements. The different methods defined for collecting user requirements are presented, namely
questionnaires, info days, focus groups and interviews. Details on the questions asked in each method are
also provided.
Section 4 “Results of the user data collection process” provides the collected results of the region-specific
methodology for collecting user requirements and data. The results obtained in each project’s region are
also presented following the same structure, namely description of results, time lines of interviews, existing
and new data gathering, statistical processing, graphs generation, and analysis and interpretation of results.
Section 5 “User requirements implications to the project applications” recapitulates the main remarks of
the previous section and provides conclusions, including graphs comparing results from different regions.
Furthermore, based on these conclusions, user requirements for both project applications are presented.
Section 6 “Study developed over the existing data and the data that will be produced in the project”. This is
the final section of the report and describes the current data sets being collected by selected SMEs in their
day-to-day web-operations. It also examines the data that will be required for the applications that will be
developed in the project. Furthermore, a description of the companies providing data and requirements to
the project and potentially assessing the project applications is provided. Finally, a very initial description of
the project semantic data model is presented.
The Annex of the report demonstrates the material drafter for the data and requirements collection
activities such as online questionnaires in various languages, slides and guides for focus groups and
interviews, project’s communication material and dissemination of the WP2 activities.

PUBLIC
Page 14 of 255
E-COMPASS
2
E-commerce economic activity in the Project Countries
2.1
Greece
2.1.1 Statistics on e-commerce activities in Greece
The Greek online environment is a strongly developing market. Just over 1.9 million Greek consumers
bought goods and services online totalling € 2.56 billion in 2012. This is an increase of 42.2% compared to
2011. The average Greek online customer annually places 20 online orders and spends a yearly amount of €
1.347, up from € 1.200 in 2011. E-commerce is projected to grow to almost € 3.2 billion in 2013. The
following figures depict the total online sales and B2C e-commerce growth rate from 2009 to 2013 for
Greece1.
Figure 1. Total e-commerce sales
Figure 2. Total e-commerce growth rate


The figures are reprinted from “Southern Europe B2C E-commerce Report 2013” from www.ecommerce-europe.eu

PUBLIC
Page 15 of 255
E-COMPASS
Based on the recently completed annual survey2 of B2C e-commerce statistics and trends in Greece for
2013, its main findings related to the SME E-COMPASS objectives and technologies are briefly presented in
the following paragraphs.
The 35% of Internet users in Greece (about 2.2 millions) purchased at least one product/service on-line.
Although the on-line market was developed with 25 % growth rate compared to 2012, still remains small,
since the corresponding European market will reach € 350 billion with 70% of Internet users buying on-line.
This highlights the prospect of the Greek market which size is around € 3.2 billion and under certain
conditions may reach the coming years the € 6 billion2.
The main characteristics of Greek on-line consumers2 and markets are summarised as follows:
 Greek online consumers have an experience of 5.5 years on average in on-line shopping. 18% began
on-line purchases in 2013, a figure that justifies largely the market growth in 2013 compared to 2012.
 The average value of on-line transactions stood at € 1.500 with the lion's share to be for the online
market of services such as travel services (e-ticketing), booking of travel accommodation, car
insurance, telecommunication services and tickets for athletic and cultural events.
 40% of the online consumers will increase their on-line purchases in 2014, while due to the economic
crisis 20% of the on-line consumers will reduce them.
 The importance of web market in Greece is strengthened by the fact that for on-line buyers the 40%
of purchases in physical stores is taking place after market research and comparison of prices done
over the internet.
 60-65% of the total on-line shopping is directed to Greek e-shops. This demonstrates the prospect of
the Greek digital business in the future, since the corresponding figure for other European markets is
close to 90%.
 The electronic markets that had the highest growth in 2013 were booking of accommodation, tickets
for events, telecommunications services, insurance and pharmaceutical products.
 Although for physical products the orders were numerically much more than the purchased services,
the latter had much greater monetary value.
 The three main criteria for the Greek on-line consumers to trust and select an online store for their
buys are:
i.
provision of a secure environment for electronic transactions,
ii.
the e-shop to be certified from a well-known independent institution or technology (i.e. 3D
secure),
iii.
the e-shop to have a clear and well defined purchasing policy.
All three top criteria are highly linked with the suspiciousness of the online consumer regarding the security
of the transactions and the reliability of the e-merchant.


Organized annually by the E-Business Research Center (ELTRUN) of Athens University of Economics and Business:
www.eltrun.gr/?lang=en 

PUBLIC
Page 16 of 255
E-COMPASS
2.1.2 Statistics for e-fraud and data mining tools in Greece
In Greece in-depth surveys and continuous monitoring and regulation of the e-commerce market are not
yet implemented and matured. Only after the end of 2012 the most active e-commerce companies in
Greece initialized the formation of a dedicated business association in e-commerce, named after GRECA
(Greek e-Commerce Association3). Its main purpose is to preserve and promote the common interests of its
members and of electronic commerce community in general. GRECE aims to play a leading role in
developing and shaping the industry and the protection and representation of its members and consumers.
GRECA is lately organising networking events, seminars and training workshops for its members as well as
meeting with public executives and policy makers for safeguarding the rights of the industry. However,
GRECA or any public policy maker or regulating authority mechanism have not published any industrial
survey detailing the profiling of e-commerce in Greece, thus any available surveys in Greece on ecommerce are exclusively coming from academic institutions or from relevant foreign bodies (i.e. eCommerce Europe). Therefore, specific data and trends for Greek e-shops regarding the use of data mining
applications, anti-fraud programmes or relevant software are rare to be found.
The most recent survey of ELTRUN (R&D Laboratory of Athens University of Economics and Business) held
for 2013 and focused on e-payments and current trends for various sectors of the national online market.
According to this survey, in 2013 among a sample of 2.300 Greek e-shops, a vast majority (81%) is offering
to its customers online payment capabilities via credit cards and 72% accepts payments with debit cards.
The respective figure for 2012 was 52% for credit cards, while for debit the statistics were not captured in
2012. PayPal as an option for e-payment is feasible in 63% of the e-shops compared to 29% for the year
2012. On the consumer’s side, “cash on delivery” and “online payment via credit card” are the most
popular payment methods followed by “debit card” and “PayPal”.
“Cash on delivery” is the most popular method for the Greek online consumers and this fact is explained by
the high sensitivity of the Greek on-line buyers for secure transactions associated with the lack of online
shopping experience and also due to the suspiciousness and disbelief that generally characterizes the Greek
consumer behaviour (this pattern behaviour is strengthened also due to the on-going economic and social
crisis since 2009).
Any other relevant statistics regarding electronic fraud in Greece (i.e. identity theft) from the market or
from national law enforcement organisations are unavailable. Unfortunately the same lack of sources exists
with respect to reports and statistics for the use of data mining methods or relevant software programmes
for consumer behaviour and e-commerce data analysis.
In order to overcome this obstacle, we present figures coming from the European Commission frequent
Eurobarometer studies, that monitor the behaviour and perceptions of the Greek online consumers, as an
effort to identify the trends and concepts that prevail in the Greek e-commerce market from the demand
side. The following figures are based on Flash Eurobarometer report 358 “Consumer attitudes towards
cross-border trade and consumer protection4” as published in June 2013 (the survey was carried in
September 2012) with a comparative manner with the European facts and figures. The Greek as well as the
European figures, pinpoint the importance of the project’s objectives for the European SMEs and
consumers.

3
See http://www.greekecommerce.gr/.
http://ec.europa.eu/public_opinion/flash/fl_358_en.pdf 


PUBLIC
Page 17 of 255
E-COMPASS

10% of internet users across the EU have experienced online fraud, and 6% have experienced
identity theft. 12% have not been able to access online services because of cyber-attacks, and 12%
have had a social media or email account hacked. 7% have been the victim of credit card or banking
fraud online.

Around half of internet users in the EU are concerned about experiencing identity theft (52%) and
about being the victim of online banking fraud (49%). Just under half of internet users are
concerned about: having their social media or email account hacked (45%) and online fraud (42%).

There is considerable variation by country in the proportion of customers that do online banking,
buy/sell goods or services online and watch TV online. Internet users were asked about the various
activities they do online. The vast majority of internet users across the EU use email (84%) and most
respondents say that they read news online (60%). In addition, around half of internet users say
they use social networking sites (53%), buy goods or services (50%), or do online banking (48%) and
18% sell goods or services.

Significant country-wise variation is also observed in the level of confidence that respondents have
in using the internet for online banking or purchases. These variations tend to reflect the levels of
actual use of the internet for these activities. Respondents in Denmark (91%), The Netherlands
(88%), Sweden (88%) and Finland (86%) are most likely to say that they are confident doing online
banking or buying things online. Customers in Denmark and Sweden seem equally confident (63%
and 56% respectively). The lowest levels of confidence are seen in Greece (42%), Hungary (43%)
and Portugal (43%).

Concerns about security of online payments have decreased in Netherlands (down 9 percentage
points compared to the 2012 survey) and UK (8 point down), but have increased in Greece (up to
13 points).

On average across the EU, 6% of internet users say they have experienced or been a victim of
identity theft. This figure is similar in most EU countries, although respondents in Malta, Ireland
and UK (11% in each country) are more likely than the average participant to have victims of
identity theft. Across the EU as a whole, 52% of internet users say they are very or fairly concerned
about identity theft. The proportion of internet users that have experienced online fraud (10% on
average across the EU) is uniformly distributed among EU countries. The highest figures are for
Malta (16%) and UK (16%), while respondents in Greece (2%), Slovenia (4%) and Bulgaria (4%) are
the least likely to have experienced online fraud.

As far as the confidence level in e-commerce transactions is concerned, the following pie illustrates
the trust of the Greek vs. the European online consumer. Greek online consumers are 30% more
distrustful to local e-shops than the European average.

PUBLIC
Page 18 of 255
E-COMPASS
Figure 3. Agree or disagree with the statement: “You feel confident purchasing goods or services via the Internet from
national retailers / providers


The findings of the confidence level of online consumers regarding cross-border online shops are
not varying between Greece and EU27 average. In both cases near half of the European and Greek
consumers are not comfortable with acquiring a product or service from a non-domestic but EU
based e-shop.
Figure 4. Agree or disagree with the statement: “You feel confident purchasing goods or services via the Internet from
retailers / providers from another EU country”
To sum up, in Greece the confidence level of the online consumers is highly related to their experience in ecommerce as well as the level of service, purchasing practices and technologies that local e-shops apply.
Both consumers and e-merchants are in a “developing” state in order to converge with the average
European trends. The project’s objectives and technological outcomes are working towards this direction
by providing both parties a more confident environment of transactions and service for increasing the

PUBLIC
Page 19 of 255
E-COMPASS
national confidence level in e-commerce. In Europe, the cross-border distrust from 1 out of 2 Europeans is
mainly based on the different national legislations regarding consumer protection and refund policies.
However, more secure transactions and enhanced level of service with the use of advanced techniques
such as data mining and computational intelligence will undoubtedly support the raise of confidence for
cross-border e-commerce.
2.1.3
Corinth Prefecture (Greece) and Corinthia Chamber of Commerce
The prefecture of Corinth, whose capital is the city of Corinth, covers a geographical area of 2,290 km2 and
its population is approximately 160,000 citizens. It is composed of 15 municipalities the largest being the
municipality of the city of Corinth with 40,000 inhabitants. The largest percentage of the total population is
mainly located in the coastal area, with largest settlements the towns of Loutraki, Kiato, Agioi Theodori and
Xilokastro. According to the latest national demographic survey (2006) the prefecture produces around
1.5% of the Greek GDP while its per capita GDP is 7.5% higher than the national average.
Corinth is a one of the biggest industrial hubs of the country. Copper cables, petroleum products, leather,
medical equipment, marble, gypsum, ceramic tiles, salt, mineral water and beverages, meat products, and
gums are produced nearby. As of 2005, a period of deindustrialization has commenced as a large pipe work
complex, a textile factory and a meat packing facility disrupted their operations. Furthermore, it is a key
transportation hub for communication with the entire Peloponnese, the Western-Central Greece and the
Ionian islands. The port of Corinth, being at the north of the city centre and close to the north-western
entrance of the Corinth Canal, serves the local needs of industry and agriculture. It is mainly a cargo
exporting facility. Sea traffic is limited to trade in the export of local produce, mainly citrus fruits, grapes,
marble, aggregates and some domestic imports. The port operates as a contingency facility for general
cargo ships, bulk carriers and ferry lines (ROROs), in case of strikes at Piraeus port. There is a RORO
connecting Corinth to Italy.
The morphological distinction between the two geographical areas of Corinth prefecture is affecting their
economic development. In the narrow coastal belt the agricultural production is highly developed and
efficient, where also nearby are located industrial units and tourism facilities. The opposite is the situation
in the mountainous region of the prefecture. Overall, Corinth cultivates about 34% of its land, from which
arable crops cover 17% of the cultivated area, 36% is dedicated to arboriculture, viticulture covers the 22%
and horticulture land-use only 3%. Of the total acreages, approximately 32% is irrigated. Main products are
raisins, currants and sultanas, grapes, apricots, wine, citrus, cereals etc. Also local farmers produce
vegetables, apples, pears, peaches, tobacco, pulses, etc. Collection and processing of pine’s resin also takes
place here. Livestock production is developed mainly in the southern part of the prefecture (goats, seeps
and poultry), while the last years beekeeping and fish farming are becoming popular. The favourable
geographical position of Corinth and the vast agricultural production are some of the key factors
contributing to the installation of a large number of industrial units in the area. Thanks to the same factors,
but also to the thermal baths of Loutraki and Corinthian Coast, the prefecture is one of the most developed
touristic areas in Greece.
The Chamber of Corinthia (EPK) was founded in 1935. It is a public organisation which apart from providing
general services and training to private companies also plays an important role as a regional policy maker.
The chamber’s objective is to promote the developmental activity in the region, as well as the growth of
industry, manufacture, trade and services in the frames of interests of the National Economy. Members of
Chamber are obligatorily all the individual and legal entities that are based in the Prefecture of Corinthia

PUBLIC
Page 20 of 255
E-COMPASS
and practice business activity. The Chamber members are 9,577. In the force of its members are also
included 117 export enterprises. E-commerce in the Corinth prefecture is being gradually developed in the
last five years. Main online markets are traditional retailers that have recently started selling products over
the internet. Mainly the retail sector offers online clothes, shoes, jewellery, watches, para-pharmaceutical
products and cosmetics, hand-made crafts, gifts and decoration objects, specific local agricultural products
and by-products. Services provided on-line are mostly related to tourism such as online travel-agencies,
booking of cars and accommodation. All of these online stores are SMEs and usually during their first steps
of operations they outsource the development and the technical maintenance of the e-shop to external
parties in Corinth or in Athens. Few of them that have been experiencing intense growth through the online
channel tend to hire in-house developer(s) and start to practice cross-border e-commerce. Most of the eshops have active presence in social media and perform e-marketing campaigns for attracting consumers.
There is intense local interest in e-commerce and the Chamber is organising seminars and training courses
for the interested members.
2.2
United Kingdom (UK)
2.2.1 Statistics on e-commerce activities
In Q1 2014 in the UK 44.6 million adults (87%) used the Internet. That’s an increase of almost 3% (1.1
million) since Q1 2013. Differentiated by age, nearly all 16 to 24 year old people (99%) had used the
Internet, compared with 37% of adults aged 75 years and over. The London region had the highest
proportion of Internet users (90%), whereas Northern Ireland had the lowest (79%) (ONS 2014).
Figure 5. Percentage of Internet Users (ONS 2014)
The Eurostat report ‘Internet use in households and by individuals in 2012’ (Seybert 2012) reported that the
UK had the highest online purchasing rate across the EU with 82% of Internet users buying online, followed
by Norway (80%), which is closely followed by Denmark and Sweden (both 79%). In 2013, 72% of all adults

PUBLIC
Page 21 of 255
E-COMPASS
bought goods or services online, up from 53% in 2008 (ONS 2013). The fruits of this high online purchasing
rate can be seen in the figures of the total annual e-commerce value. With an estimated amount of £492
billion, e-commerce sales made up 18% of business turnover in 2012. Compared with £335 billion in 2008,
that is an average annual growth of 10% and a total growth of 47%. (ONS 2012) However B2C e-commerce
sales in the UK are expected to experience declining growth rates from 2013 on (yStats.com 2012).
£600 b
£500 b
+ 16%
+ 2%
2011
2012
+ 12%
+ 12%
£400 b
£300 b
£200 b
£100 b
£0 b
2008
2009
2010
Figure 6. Total e-commerce sales (ONS 2012)
E-Commerce consumers – most of middle income families
In the UK, out of 63.7 million inhabitants 61% are e-shoppers (39.0 million), living in 23.9 “e-households”.
The average annual spend per e-shopper in 2012 was €2,466 (Ecommerce Europe 2013).
Income and age make the difference
In 2013 adults aged 25 to 34, more than any other age group, used the Internet to purchase goods or
services online (92%). Additionally there has been significant growth in the rate of online purchasing by
those aged over 65: over a third of these bought online (36%) in 2013, more than double the 2008 estimate
of 16% (ONS 2013).
The bulk of 2011’s online shoppers were members of middle income families who use the internet to buy
everything from their weekly grocery shop to impulse purchases of the latest fashions, along with less
frequent but higher value purchases such as the annual summer holiday and renewing car insurance via a
price comparison site. Besides this there is a group of people (14%), who are highly skewed towards 56+
and those on lower incomes, who hadn’t purchased online at all or only very rarely. A proportion of this
group is open for online shopping. However, they claim that the flexibility of delivery options and easier
returns are the most important influencers when purchasing online (Zablan, Oates, Jenkings, Bennett, and
Goad 2011).

PUBLIC
Page 22 of 255
E-COMPASS
Reasons for shopping online
In 2013 in the UK 71% of all individuals ordered/bought goods or services for private use over the Internet
at least once in a three-month period. That’s first place compared to all other EU countries and nearly
double the EU average of just 38%. Hence the UK is leading the way in EU e-commerce (Eurostat 2014).
As the UK e-commerce market grew much faster than in the rest of the EU, it may be very interesting to
understand the reasons and factors which might account for this enormous increase:
A good precondition is the widespread availability of broadband Internet access and the high personal and
business computer use. The widespread ownership and use of payment cards provided an already accepted
way for online payments. The shared language with the US helped and successful US e-Shops soon came to
the UK (Bamfield 2013).
The key reasons for shopping online from a consumer’s perspective are lower prices as mentioned by 60%
of households surveyed by the ONS (Kalapesi, Willersdorf, & Zwillenberg 2010). Further important reasons
are convenience and access to a much greater product range than shopping in the high street. Additionally,
the decision to purchase can be influenced by targeted email offers from retailers, voucher codes and deals
from sites such as Groupon (Zablan, Oates, Jenkings, Bennett, and Goad 2011).
In the UK, it appears that saving money is a major factor as to why consumers go online for grocery
shopping – almost half (48%) look for deals, 30% go to coupon websites and 25% compare prices. Among
those looking for grocery coupons, more than a quarter (26%) do so on a daily basis. Britons are more likely
to use the Internet for saving money on groceries than Europeans as a whole; 43% of Europeans look for
deals online, whilst 22% look for coupons. Over the last year, the rise in food prices has been the biggest
factor determining what grocery brands and products Britons have purchased. This is followed by increased
transportation costs (27%), health reasons and retailer loyalty programs (both 21%). The availability of selfservice checkouts has had a major impact on the grocery choices of 18% of Britons online (Nielsen 2012).
E-Commerce products – fashion is leading
“Clothes and Sports Goods” was the leading online product category, bought by 41% of all internet users in
the UK in 2011, followed by “Travel and Holiday Accommodation” (38%) and “Household Goods” (33%)
(yStats.com 2012). In 2010 every second travel was already booked online. The most popular product
categories by gender are clothing and sporting-goods for women and film and music for men (Kalapesi et al.
2010).

PUBLIC
Page 23 of 255
E-COMPASS
Figure 7. Types of Goods and Services bought online in the UK in % of Individuals, 2011 (yStats.com 2012)
There are noticeable differences in the type of goods bought online, when viewed by gender or by age. In
2013, half of all women (50%) bought clothes online, compared with 45% of men (ONS 2013).
Figure 8. Purchases of certain types of products made online by age group (2013) (ONS 2013)

PUBLIC
Page 24 of 255
E-COMPASS
E-Commerce merchants and their shops – the largest are dominating
The comparatively few largest businesses with over 1,000 employees in the UK, dominate e-commerce
sales as they made up over half (51%) of all e-commerce sales in 2012 (£84,9b). Together with all medium
large businesses (more than 250 employees) they made up over three-fourths of all e-commerce sales. In
contrast, the large amount of businesses with less than 250 employees made less than one-fourth (23%).
Figure 9. Value of e-commerce sales over a website by size of business (# of employees), 2012 (ONS 2012)

Based on the annual e-commerce revenue in 2013 only 14% of the merchants stayed below £0.5m,
whereas 30% reached £25m and more (Khan 2013).
Figure 10. Size of Business by e-Commerce Revenue (Khan 2013)
In 2008, 51% of online purchases were from pure-play online retailers. From 2011 that figure has fallen to
41%, while the proportion of online sales to multi-channel retailers has risen from 49% to 59% (Zablan,
Oates, Jenkings, Bennett, and Goad 2011). According to the Royal Mail in 2012, there were about 14,400
online-only retailers (Royal Mail 2013).

PUBLIC
Page 25 of 255
E-COMPASS
Mobile – a new trend
In 2012, the mobile e-commerce, also called m-commerce, in the UK was worth £16.4b. Within a short
period of time from Q4 2012 to Q1 2013 it grew 15.4% so that 20% of all e-commerce consisted of mobile
commerce. The expected growth for 2013 was 71.8%. In addition to mobile shopping, 40% of UK smart
phone users use their phone for mobile e-banking (Ecommerce Europe 2013).
The most popular mobile device used for shopping is the iPad, representing 82% of all mobile shopping in
the UK in 2012. Therefore, e-Shop owners should optimize their websites for mobile devices, especially
considering the iPad, to make mobile shopping an increasingly enjoyable experience even on smaller and
very small screens. Mobile shopping is most common among 15-34 years. 57% of m-shoppers who regularly
shop through a mobile device are in this age group (Ecommerce Europe 2013).
Figure 11. UK Age Structure of Mobile Shoppers, 2012, Percentage of total Population (Ecommerce Europe 2013)
British consumers like to purchase media entertainment goods (e.g. music & films, software programs,
toys, PC and DVD games) through a mobile device - 46% have done so in 2012. Clothing is also becoming
increasingly popular among mobile users (34%) (Ecommerce Europe 2013).
Payment – Cards and PayPal dominate
Almost all UK e-shoppers (96%) use one of three payment methods: credit/debit cards or PayPal. The
preferred card schemes are MasterCard, Amex, Diners Club and VISA. Only 4% make use of alternative
payment methods like Ukash or ClickandBuy (The Paypers 2013, Payvision 2012).
Figure 12. Payment Methods (Payvision 2012)

PUBLIC
Page 26 of 255
E-COMPASS
2.2.2 Specific figures statistics for e-fraud and data mining tools
According to the “9th annual UK e-Commerce Fraud Report” (Khan 2013), in 2012 1.65% of all e-Commerce
revenues were lost due to fraud. Thereby the loss rate highly correlates to market sectors: digital goods
have the highest loss rates, the travel and service sectors have the lowest rates. Based upon the number of
total orders the mean fraud rate is 1.26%.
When asked about the top e-Commerce fraud challenges more than every second fraud manager (51%)
mentioned that they are losing business by turning away too many good customers when trying to detect
fraud. The second and third most common challenges were the high cost for manually reviewing too many
orders (43%) and that the used fraud detection tools are unable to detect the latest fraud threats (39%).
For those reasons, 58% of merchants manually review suspicious orders and even 7% analyse every order
(mainly small e-Shops), whereas 48% of very large merchants (with more than £25m annual revenue) have
completely eliminated manual review processes.
Figure 13. Time taken to review orders (Khan 2013)
More sophisticated fraud detection
In 2012 e-merchants used 5 fraud detection tools on average, ranging from three to four tools used by the
smallest merchants to six or more tools used by the largest ones. The most popular tools were verification
services like Card Number Verification, used by 70%, 3D Secure (61%) and Address Verification Services
(56%). For 2013 a lot of improvements were planned: first of all 18% of e-merchants plan to use “Customer
website behaviour/pattern analysis” for fraud detection because fraudsters often take a very direct,
identifiable route to the checkout, creating recognizable patterns. Behavioural analysis combined with a
rule based system can help to identify these patterns and allow merchants to take appropriate actions.
Furthermore 14% of e-merchants plan to consider the order history of their customers and 13% want to
measure the purchase velocity to improve fraud detection. Again, both are most powerful in combination
with a rule based system (Khan 2013).
Trends which influence the adoption of data mining in e-commerce

PUBLIC
Page 27 of 255
E-COMPASS
A major trend is the identification of large data volumes which have been gathered in the past (e.g. web
analytics metrics, transaction information, etc.) or which will be collected in the near future (e.g. social
media information) and their real-time analysis in order to create a business value. Some of the trends
which foster the application of data mining techniques are introduced in the following sections.
Mobile and Localization
Mobile Internet usage and mobile commerce are the most rapidly growing future trends, driven by broad
distribution of smartphones and tablets. In 2013 the mobile penetration within the UK reached 132%. As a
result “non-smart” mobile phones were increasingly replaced by smartphones, reaching 58% penetration.
Additionally more and more people are using a tablet (penetration 19%) (The Paypers 2013). In 2011
Vodafone reported (Vodafone Quarterly results) that 90% of its new UK contract connections are sold with
a smartphone. Access to the Internet using a mobile phone more than doubled between 2010 and 2013,
from 24% to 53% (ONS 2013).
In the next few years, the following transformations can be expected, making new mobile e-commerce
applications possible. First the coverage of high-speed mobile networks will enhance and at the same time
existing mobile networks are getting faster and faster reaching a new level of speed with 4G (downlink
peak rates of 300 Mbit/s), comparable or even faster than today’s fixed line broadband connections like
DSL. Second the smartphone penetration will rise beyond 100% - nearly everybody will have one or more
smartphones. At the same time smartphones will improve in terms of their processing power, memory and
screen resolution. Thus new and enhanced capabilities can be implemented and used e. g. more precise
GPS and/or WLAN-localization, voice and optical character recognition, image processing and so on.
Through the combination of these changes more and more e-shoppers will use their smartphone to shop or
to assist shopping by comparing prices while they are already inside a shop. Additionally new Apps can
improve the customers shopping experience. For example, a product description or a product video can be
displayed when the corresponding product is recorded by the smartphones camera. This may also happen
in an augmented realty fashion, where the currently watched/recorded product can be annotated with
useful information. For example the app can show the technical data of the product. In addition the app
can make recommendations for related products and guide the customer straight to the right rack, aided
by new indoor-positioning systems. Also a possibility to rate or evaluate a product could be offered, thus
generating a lot of user behavioural data.
Another trend is the use of GPS and WLAN-localization which can be combined with compass and
gyroscope measurements to achieve not only outdoor but also indoor positioning with higher accuracy.
Based on this data an app could list the best/nearest/cheapest shop for a given product.
Social Media
The UK Office for National Statistics states that in 2012, just under half (43%) of all businesses reported that
they made use of social media. However, there is a strong connection between the size of a business and
the likelihood of interacting on social networks: with 79% of the largest businesses almost twice as much as
the smallest ones (40%) used social networks (ONS 2012).

PUBLIC
Page 28 of 255
E-COMPASS
Figure 14. Proportion of businesses using social media, 2012 (ONS 2012)
With the fast and wide distribution of social media, ignoring it more and more becomes a real disadvantage
in e-commerce. By using social media, e-merchants can get to know their customers and their needs better.
In addition, they can enrich web analytics data with detailed data about the social background of the
shoppers. Finally they can significantly improve their ability to customize the shopping experience for each
customer.
Personalization (User-centred and responsive/mobile design)
In pre-e-commerce times some years ago customers would go into a shop where a merchant was waiting
behind a sales counter. The merchant took care of every single shopper, giving each one his or her full
attention. The needs of the customers took centre stage. Nowadays all e-shoppers get the same shop with
the same structure, the same products and the same prices, although the desire for one-to-one assistance
still exists. Additionally through social media sites like Facebook, where everybody gets their own “Wall”
with messages, users are accustomed to personalization and more and more expect websites that perfectly
match their personal needs. By offering a personal and more human interaction, retailers can create an
experience online, or on mobile, that is more akin to the customer service they would deliver in-store,
allowing retailers to attract and retain more loyal customers through their online platforms (Abensur 2013).
The chain of department stores “John Lewis” states that their new personalized recommendation tool on
their website was a key factor in driving a 27.9% increase in sales over Christmas 2011. Their goal was to
provide any shopper coming to their website with the same personalized customer service as if the shopper
was visiting one of their local shops (Moth 2012).
Because consumers not only want unique e-shops, but unique products, in 2014 more businesses will
empower their customers with the ability to personalize, modify, or design the products that they want to
purchase. New technologies like 3D printing will help merchants to offer or to improve customization (Xu
2014).

PUBLIC
Page 29 of 255
E-COMPASS
Big data
The above mentioned trends will generate a lot of new data about individual customers and their shopping
habits as well as their social environment (from social media) and of course their position and movement.
As organisations have already collected vast amounts of personal data, the high degree of mobile usage
generates such an amount of new data about potential customers that it cannot be processed by
traditional methods. So, many companies have been collecting data for years but are not necessarily
putting it to good use. As “big data” becomes a mature technology, nowadays more and more merchants
apply it. Additionally the big players employ data analytics specialists to make smart commercial decisions.
As small businesses cannot afford to do this, they need the help of customizable web services providing
possibilities to evaluate all their collected customer data and drawing useful conclusion for them. Thus they
would get insights, previously only reserved for larger retailers. The next step is to enrich the existing
analytics data with new data sources. For example, Tesco uses weather records to help predict demand for
certain products based on weather forecasts (Hesse 2013). Other useful data sources are the prices and
shipping fees of competitors. A web service using this data could help to find optimal prices regarding high
sales figures and high margins, depending on the pricing policy of the competitors.
2.2.3 Halton Chamber of Commerce and Enterprise
Halton Chamber of Commerce & Enterprise is a not for profit, independent business membership
organisation, serving the Local Authority area of Halton. Working with other Chambers of Commerce, Local
Authorities and the Local Enterprise Partnership, it links into the wider Liverpool City Region.
Originally established in 1995 as Halton Chamber of Commerce & Industry, its legal status is a company
limited by guarantee. Its mission is to promote and support the interests of its member companies and the
wider business community, and to further the prosperity and economic wellbeing of the borough of Halton
in particular and the wider city region in general.
Halton Chamber currently has 320 business members across a wide range of sectors (Manufacturing,
Engineering, Construction, Freight, Legal, Banking, Accountancy, Training, ICT and Technology) which
represents 10% of the Halton business base. Its membership ranges in size from large multi-national
corporations like Ineos, Mexichem, Telefonika, Yokogawa and ABB, to small and medium enterprises,
owner/managed businesses, and start-up companies, who collectively employ in the region of 28,000
people.
Its services include certification and international trade guidance and advice, business training, mentoring,
specialist consultancy along with business networking, information and general business support. It has
delivered a number of ESF and ERDF funded projects and is currently delivering a large proportion of Halton
Council’s ERDF 4.2 business support programme in a range of areas including Business Diagnostics,
Strategic Business Planning, Financial management, HR, Environmental, and Manufacturing Process and
Efficiency. It works closely with other providers on modules covering E-Commerce and ICT.
The Chamber’s mission is to enable its local businesses to better compete in domestic and overseas
markets and a significant proportion of its work is focused on providing information and access to

PUBLIC
Page 30 of 255
E-COMPASS
technological advances and innovation that will help small to medium companies become more
competitive in today’s global markets and technology driven world.
2.3
Spain (Basque Country and Valencia)
2.3.1 Statistics on e-commerce activities in Spain
According to the study of the National Observatory for Telecommunications and the Information Society
(ONTSI) the B2C e-commerce in Spain grew by 13.4% reaching € 12.383M of turnover. This is less than the
increase experienced in 2011, but is especially relevant given the socioeconomic context of crisis in which
we have been living for the last few years.
The main cause of the growth of e-commerce is the increase of Internet purchasers in 2012 which was of
15% (€ 15.2M ). The continued growth of the Internet user population representing 69.9% of the Spanish
population over 15 years has also influenced this increase.
The main buyers are 25 to 49 years old, living in urban habitats of more than 100,000 inhabitants. Men
living in populations between 10,000 and 50,000 inhabitants are attending their first online purchase.
However, young people aged 15 to 24 years and those over 65 are becoming more intensive users.
Between old and new buyers, there are also differences as the first ones have cut their spending while the
new ones have increased it.
The leading products are ‘airline transport’ (47.2%), ‘lodging reservations’ (41.9%) and ‘show tickets’
(32.9%) however these are the sectors experiencing a slower growth.
In 2012 it was also increased the average spending on clothing, accessories and sports equipment, books,
magazines and newspapers, toys, board games and games, appliances, home and garden, and jewellery.
Transport tickets and gambling are the only two product categories with percentages in decline.
Meanwhile, the average spending per buyer decreased by 1.4% (from 828€ in 2011 to 816€ in 2012).
The home is preferred by a 93.5% to make online purchases, followed by the workplace. Up to 16.8% of
people purchased online at least once a month.
Particularly relevant is the fact that for the first time, websites that sell primarily online were consolidated
as the main purchase channel (48.7%), followed by the manufacturer websites (44.4%) and companies on
buying and off (36.1%).
Credit and debit cards are still the favourite mode of payment for almost 63% of respondents. However, its
use declined compared to 2011 by 3.3%, which increased the percentage of exclusively electronic payment
platforms. Price (71.5%) and convenience (62.8%) are the main reasons for purchase.
The e-commerce also scored positive figures in 2012. 2.1M people used mobile or tablet to buy, which is an
increase of 15.1% over 2011. Digital content is the flagship product since it accounts for about 48%. As for
social networks, although 65.9% stated themselves as users, 74.3% declared not having used them in the
purchase process and only 1/3 declares being fan or follower. The report drives to these points of
improvement: delivery defective, shipping delays, discrepancy between the products offered on the web
and received.

PUBLIC
Page 31 of 255
E-COMPASS
Besides, a more approximated data in time (2013) and origin (region level) could be obtained through the
INE (National Statistics Institute of Spain). According to this statistics institute, in 2013 e-Commerce in Spain
has increased, being already about 11 million people who have made any purchase through the web in the
last 12 months. This represents 31.5% of the total population. Among the reasons why consumers prefer to
buy online, 78.0% argued the comfort of this service as a main reason for preferring this way of purchase,
73.2% argues the possibility of finding deals and articles at a better price, and 65.5% reported that it saves
time, not implying to move physically.
Figure 15. Quantity of B2C E-Commerce (€ million per year)
Moreover, the region of Spain that has used more this type of trade is the Basque Country, representing a
41.1%, followed closely by the Community of Madrid with 40.2%. Those who use it less are Canarias
(20.7%) and Extremadura (24.1%).
Basque Country
Putting the focus on the case of the Basque Country, Eustat (Statistics institute of Basque Country) provides
some information about ecommerce through data-bank and annual reports (under the information society
topic). In this report there is an interesting section for the e-commerce.
The 2013 report by Eustat shows at the begging of the 1st chapter the results of a survey gathering the
shopping types carried out by customers during the last 3 months. In general, the 38.9% of the users that
have been connected recently they have effectuated some purchase on the internet.
Among them, a 35.8% mainly acquires goods related to sports equipment and clothing, travel and
accommodation (26.5%), other products or services (20.0%), household products (15, 1%), event tickets
(10.7%), electronic products (14.4%), books and magazines (11.8%), computer equipment (9.8%), software
and video games (4.9 %), cars, bikes and accessories (3.7%), music and musical instruments (3.4%), videos
and movies (1.7%), financial products, investments and insurance (0.7%), lotteries or gambling (0.4%), news
and information (0.2%).
From this section it can be noted that the profile of the online shopper is that of a young man who lives in a
family with children, holds a higher education degree and is currently employed. By gender, women
generally buy home-related products, travels and holiday accommodation. However, the purchase of music
and musical instruments, cars, motorbikes and accessories, and video game software are mostly carried out
by men.

PUBLIC
Page 32 of 255
E-COMPASS
Figure 16. People who bought online in the last 12 months in 2013 (% of people aged 16 to 74 years)
Apart from the information mentioned above, there are other aspects that may also be showcased. These
are deeply related to the amount of money spent on Internet shopping, security to pay online giving
account number or credit card, the profile of the Internet shopper, and so on.
The data analysis conducted in the last decade reveals a stable preference of the Basque Country
population to buy on traditional commerce, with percentages slightly over 50% in recent years. On the
other hand, the data shows that Basque society is losing the fear to make a purchase online due to security
or privacy reasons, of which the percentages are considerably descending in both cases. In fact, between
2003 and 2013 it has continued to increase the percentage of those who believe that paying over the
Internet offers a lot of security (from 13.2% to 20.0%) and significantly (from 42.5% to 53.9%). In addition,
those who see little security (30.5% to 18.7%) and none (13.6% to 7.4%) are reduced. Internet seems to
offer and transmit increasing sense of security.
Valencia
In case of Valencia, the commerce office presented a report about the distribution of retail trade in Rovira
et al. (2013). According to this report, around 26.18% of Valencians between the ages of 16 and 74 have
bought something online in the last year. This is 10.3% more than in 2004. The incorporation of new
technologies and the use of mobile devices is changing consumer behaviour; it is becoming multichannel.
This demonstrates new behaviour where there is no difference between online and offline, they are two
complementary channels for purchasing goods.
The types of product most purchased over the Internet by Valencians (see Figure 17) were those related to
tourism, accommodation rentals (50.2%) followed by the travel-related services (airplane public transport,
car rental).
From 2007 to 2014 online shopping habits have changed. The numbers of products which are purchased
online have significantly increased. The most popular are sports equipment and clothes (increase of 17.7%),
home furnishings and toys (increase of 9%) and electronic equipment (increase of 9.1%).
The main reasons given by the Valencians for preferring the online channel (see Figure 18) are saving time
and convenience; price promotions and offers; and the ease with which they can compare offers and
product information.

PUBLIC
Page 33 of 255
E-COMPASS
Figure 17. Most purchased products over the internet by Valencians
Figure 18. Preferred online channels by Valencians
Finally, Table 1 shows the top 10 products categories which are more demanded by Valencians through ecommerce.
Table 1. Top 10 product categories in Valencia e commerce
Top 10 product categories
Column name
01 - Mobile Phones
02 - Computers and tablets
03 - Home, Garden and furnishings
04 - Cars, motorbikes and accessories

PUBLIC
Page 34 of 255
E-COMPASS
05 - Toys
06 - Clothes, shoes and accessories
07 - Consoles and video games
08 - Sports
09 - Watches and jewellery
10 - Beauty and wellness
The Spanish Association of Digital Economy (adigital) presented the "Report on Payment Systems and
Electronic Commerce Fraud in 2012" which examines the experiences of 317 companies from various fields
of activity.
According to this document, 75% of companies using e-commerce as a sales channel do not have a fraud
management system, mainly due to a still lowερ incidence of it among SMEs or because they transfer the
risk to a third party.
Regarding the presence of fraud activity, 82% of businesses say that this problem represents less than 0.5%
of their turnover. On the other extreme, 2.6% said that it exceeds 5% of its turnover. As for investment in
fraud management, 2.1% of the companies surveyed apply between 50,000 and 100,000 per year; 9.3%
between 5,000 and 50,000, and the vast majority, 86.5% say they spend less than 5,000 euro.
Regarding the most common means of payment 78% accepts credit cards; 76% offer the possibility of bank
transfer; 58% have Pay Pal; 38.8% offer choice cash on delivery and 14.5% said other options, including the
direct debit.
Concerning the presence of mobile devices in the online purchase, 39.4% of the companies surveyed said
that their weight in the total turnover of activity represents less than 1%; while 31.5% state that is between
1% and 5%.
Google Analytics is the most widely used web analytics tool. More than a half of the companies in Spain
already opt for it. It seems that companies increasingly value the need to know, in as much detail as
possible, the behaviour of visitors on their websites. This has led to 56% of companies to use Google
Analytics in order to study, in depth, the behaviour of users on their websites. Moreover, its popularity is
increasing. According to data from E-Consultancy, it has grown by 9% since last year.
According to the E-Consultancy report “Internet Statistics Compendium” in Econsultancy.com. Ltd (2013),
companies particularly value the ability of Google Analytics to measure website traffic and conversions
(86%). Meanwhile, 75% highlighted the ability to track their online activity. With regard to what the tool is
used for, 60% of users analyse the behaviour of visitors on their page, the number of page views or time
spent; while 40% mention the relevance of its content. However, not all companies use Google Analytics.
The report indicates that 35% of companies do not use Google Analytics because they consider its internal
processes are not complex enough; so they think they not need it. 10% of companies does not know the
tool.

PUBLIC
Page 35 of 255
E-COMPASS
2.3.2 GAIA & ATEVAL SME-AGs
GAIA
GAIA is the Association of Electronic and Information Technologies in the Basque Country, a private and
professional non-profit organisation, established in 1983, currently made up of 260 industrial members
(80% SMEs, with a total of 11,000 employees) that offer products and services in the field of electronics,
information technology and telecommunications. GAIA has 12 staff headcount. GAIA has vast experience in
the development of EU funded projects related to the development of new technologies for different fields,
such as environment, health, transport, green technologies, and since almost 10 years one section of the
organization has been working in the identification of knowledge, competences and skills needed for new
jobs in the area of ICT, together with several organization around Europe. Main support actions for
associated SMEs are: strategic projects identification, actions between large companies and SMEs in order
to foster direct subcontracting, competitive networking, and actions between research entities, universities
and SMEs, to have these last benefited from state-of-the-art RTD, identification of strategic technologies
for SMEs in the TPs SRAs.
The Electronics, IT and Telecommunications sector in the Basque Country is one of the most important
concentrations of industrial developments in this sector in Spain.
The tradition of manufacturers and entrepreneurs in the region, the excellent training and research
infrastructures, the high sensitivity and commitment of public administrations and the existence of GAIA as
an association that boosts and coordinates joint, technological and commercial activities are all at the core
of our ongoing growth in our sector and its positioning as an outstanding European reference with a
vocation.
GAIA provides a wide range of services and programmes to member companies in fields such as
technology, management improvement, training, and marketing and internationalisation. It also provides
other general services (representation, coordination of committees and business groups, consultancy,
promotion of business collaboration between member companies and with third parties, etc.), typical of an
Industrial Association.
Its mission is:
 To promote all the aspects of development and growth related to the Electronics, IT and
Telecommunications.
 To defend the legitimate interests of member companies.
 To favour the assimilation and efficient usage of advanced technologies by the Basque Country as a
region, with the aim of collaborating with the development of an Information and Knowledge
Society.
 To be recognised as the most committed private and independent institution to the development of
the electronics and ICTs that it represents with a rational and efficient usage of products and
services based on those technologies, in the Basque Country.
The main points defining the GAIA vision are:
 To be the most important reference as an association, which integrates efforts and skills in the
referred technologies in the Basque Country.

PUBLIC
Page 36 of 255
E-COMPASS
 To be the association that represents the largest number of companies in Spain, in the European
Union and in the global market, through its own actions and possible collaborations as equals with
other institutions.
 To be an active and dynamic agent that reflects the image of the Basque Country as a technologically
advanced society, commercially flexible and committed to society, and to be a key element to the
internationalisation of its member companies.
ATEVAL
ATEVAL is a private industrial association which was founded in 1977, with more than 450 associated textile
companies, and representing almost 11,000 employees. ATEVAL, has as its main objective the
internationalization of the intersectoral R&D&I, encouraging and accompanying textile companies, with the
aim of increasing their competitiveness by providing higher value-added and differentiating their products.
The association aims to organize, promote and develop the Valencian textile cluster, as well as its
cooperation with other Spanish or foreign companies. The main goals of ATEVAL consist of: innovation in
products, processes or materials; participation in international, national and regional processes of
development and knowledge management; the revitalization of the sector by promoting training,
innovation, internationalization and cooperation; and finally, the promotion, innovation and adaptation to
the changes in the textile sector.
The mission of ATEVAL comprises the following activities:

public-private cooperation for the benefit of the industry,

implementation and dissemination of studies, research and reports about business management
and the industry,

canalization of industry proposals obtained through surveys, panels, exploratory sessions and other
participatory means,

promotion of activities in the sector, canalization of public and private aids,

promotion of plans to improve competitiveness, internationalization and cooperation between
companies from the sector,

development of training activities, outreach and support for the benefit of the sector ‘s companies.
These activities are divided into several departments: industry and environment, internationalization,
innovation and others, which aim for continuous contact and collaboration with other institutions and
organisms of different areas, such as: the Institute of Exterior Trade (ICEX), the Valencia Institute of the
Export (IVEX), different Chambers of commerce of the Comunitat Valenciana, and the Technological Textile
Institute (AITEX). In European sphere, ATEVAL takes part in the Employer Textile European (EURATEX) and is
a member of the manufacturers' International Federation of Tapestry.
The association includes companies of all the subsectors in textile such as textile home, threads, carpets,
confection, finishers, commercial, and companies specialized in technical textiles and machinery.
ATEVAL began its work in European project calls in 2004 with the LIFE program. From this experience
onwards ATEVAL has participated actively, putting forward its proposals to various programs of European

PUBLIC
Page 37 of 255
E-COMPASS
calls. These projects have been and are a tool for placing ATEVAL and textile companies representing the
starting position in international markets, equipped with all the necessary options for the development of
R+D+I in order to be competitive in the globalized world in which we live.

PUBLIC
Page 38 of 255
E-COMPASS
3
Methodology for collecting user requirement
The objective of the user requirement analysis is to harvest knowledge from e-commerce entrepreneurs
and SME professionals regarding their needs and specific challenges in web-transactions security and sales
operations. The objective is also to gain a complete understanding of the current data sets being collected
and produced in their day-to-day web operations as well as how (and if) they are currently analysing these
data sets for the benefit of their SMEs.
To this end, a framework as well as a methodology for collecting the user requirements have been defined.
This framework consists of four main elements: questionnaires, info days, focus groups and interviews.
 Questionnaires
In order to collect basic information about companies’ profiles, companies’ habits and companies’
interest in the services developed by the SME ECOMPASS project, a basic questionnaire (or e-survey)
has been developed. The questionnaire has been created using Google drive and can be completed
online. Therefore, companies find it easier to complete the questionnaire and answers are
automatically stored and ready to process. Table 2 shows a summary of the basic questionnaire
queries. Appendix A includes the complete questionnaires in its three different versions: Greek,
Spanish, and English (extended). All queries are closed, i.e. several alternative answers are suggested.
Thus, companies only have to select one of the answers given, reducing the time needed to complete
the questionnaire.
 Infodays
The main objective on the infodays is to bring together several companies in order to present them the
objectives of the project and to know the level of interest of each company to participate and/or
collaborate with the project. Organizing an infoday is an alternative or a complement to send the
questionnaire link to the companies. During the infoday the members are asked to fill-in a form of
interest (not a full questionnaire) so that their contact details can be collected and which ones are
interested in participating in the forthcoming focus group can be seen. From the collected contacts,
the full questionnaires will be send. For SMEs not attending the infoday, the SME-AGs will make some
phone calls for informing them on the project and ask them to fill-in the e-questionnaire.
 Focus groups
Once the information on companies’ profile and companies’ interest in the services developed by the
SME E-COMPASS project through questionnaires and/or info days, has been compiled, SME-AGs with
the assistance of their RDT partners will select a group of companies. These companies will be those
considered most suitable for describing and transferring their data to the project and for piloting
applications developed within the project.
This group of selected companies are convened at a focus group where, by means of a discussion,
several technical questions are answered. Two technical sessions will be organized, one for the antifraud service and one for the data mining for e-sales service. Questions in each technical session are
divided into different categories. Table 3 and Table 4 show these categories. Furthermore, companies
will be asked about the possibility of participating in the project as pilot users.

PUBLIC
Page 39 of 255
E-COMPASS
 Interviews
Companies which show an interest in piloting the project application, can arrange an interview where
more technical details can be given and a better understanding about the company profile and interest
can be developed.
Table 2. SME e-compass survey questionnaire: Common questions to all versions
SME e-compass survey questionnaire: Common questions to all versions
Questions
1. Which types of products and services do you offer?
2. Please tick the most relevant of the e-commerce sectors that your company belongs to
3. Which kind of product do you offer?
4. How many full time employees dedicated to e-commerce does your company employ?
5. Which is the total 2013 annual revenue from online sales?
6. Which is the total 2013 annual revenue from total (online and offline) sales?
7. Which was the annual volume of orders received in 2013? (including the orders that were not
executed for any reason)
8. How long has your company been doing business online?
9. Which languages does your e-shop currently support?
10. Which are the main Websites where you compare prices?
11. How often do you need to compare prices?
12. How do you adjust your prices?
13. Are you interested in a service which compares your products prices, sends alerts to you when
prices exceed certain price limits, and supports your price adjustments either manually or
automatically?
14. Are you currently using any service or tool for customer behaviour analysis, churn prediction
and/or prevention?
15. Are you interested in a service which analyses the customer behaviour, provides feedback on
how to improve your e-shop and supports/optimizes your cross-selling activities?
16. Are you interested in a service which analyses the behaviour of groups of customers, discovering
habits, and detecting new e-shopping tendencies?
17. Is fraud a concern which prevents you from selling your products/services online?
18. Is fraud a concern which prevents you from expanding your online market access to the entire
EU?
19. Is fraud a concern which prevents you from using online payment transactions or e-card systems?
20. What is the typical proportion of fraudulent cases in your total volume of transactions?
21. How do you deal with online payment fraud?

PUBLIC
Page 40 of 255
E-COMPASS
22. Do you use a specific software to deal with online payment fraud?
23. Do you use your own software or assessment method to deal with online payment fraud?
24. What type of transaction-specific information does your antifraud personnel or antifraud tool
take into account?
Table 3. SME e-compass Focus Groups question categories: Anti-fraud


SME e-compass Focus Groups question categories
Anti-fraud
1. Current fraud management practices
2. Cost-efficiency concerns
3. Efficiency of the overall fraud management process (performance indices)
4. Future actions for combating fraud
Table 4. SME e-compass Focus Groups question categories: Data Mining for e-sales
SME e-compass Focus Groups question categories
Data Mining for e-sales
1. Current challenges in e-sales
2. Current situation in web analytics/ visitors‘ behaviour analysis
3. Current situation in competitors‘ analysis
4. Requirements for web analytics/ visitors‘ behaviour analysis
5. Requirements for competitors‘ analysis
6. Requirements for automated actions
7. Pilot Users

PUBLIC
Page 41 of 255
E-COMPASS
4
Results of the user data collection process
4.1
Results and Analyses from Greek Chamber (EPK)
Corinth Chamber of Commerce (EPK) lists around 10,000 SMEs and entrepreneurs as members of the
chamber. As happens in the whole country, statistics and mapping of e-commerce activities are still weak
and the same situation exists in Corinth prefecture. The chamber is aware of the larger and more active eshops with membership, but has not listed smaller retailing SMEs that the last two years have expanded
their operations in e-commerce. Given this fact in Corinth, the Chamber decided to initially co-organise an
infoday with the Greek RTD performers (EXUS and E-TRAVEL) at the very start of the project in order to
raise the awareness of Ε-COMPASS in the region and attract those SMEs that are newly active in ecommerce. By this activity the Chamber could achieve a better mapping of SMEs with strong involvement
in e-commerce and communicate to the interested members the scope and benefits of the project for the
local SMEs. Thus, this infoday served as the starting point of the user and data requirements phase of the
project and the chamber had the chance to announce and promote during the Info day the organisation of
a focus group, possible face to face interviews with e-commerce SMEs as well as the launch of the
electronic survey on e-commerce statistics, needs and trends for the chamber’s SMEs. In the next
paragraphs follows a description of the activities that EPK carried out with the support of the RTD
performers for WP2.
4.1.1 Organisation of SME E-COMPASS INFODAY in Corinth region by EPK, EXUS & ETR
The infoday in Corinth was decided to take place on Monday, 17th of February, at the conference hall of the
Chamber in the center of Corinth city. The event was organized in the afternoon with duration of two hours
and a half following a structured agenda. For the event to be more appealing, the Chamber and the Greek
RTD performers decided to invite two external speakers outside the project consortium; one from the
private and one from the public sector. The external speakers were asked to make an introductory
presentation a) about the current status of e-commerce in Greece and b) the opportunities provided by the
new Framework Programme (Horizon 2020) for research and development SME initiatives. The coordinator
and EPK invited Mrs Helen Spyropoulou, executive of Athens Chamber of Tradesman and National Contact
on SMEs related calls of FP7. Mrs Spyropoulou’s presentation focused on the participation of Greek SMEs in
FP7 calls and projects as wells as the funding opportunities and instruments that HORIZON 2020 provided
to local SMEs. ETR invited Mr Panagiotis Gezerlis, General Director of the Greek Association of e-Commerce
(GRECA), for giving a speech entitled “ E-Commerce: the hope of Greek SMEs to development”. The infoday
was opened by an introductory speech to the audience by the President of Corinth Chamber of Commerce,
Mr. Vasilis Nanopoulos, highlighting the opportunities given to the local SMEs to participate in the SME ECOMPASS Project and the motives and plans that the Chamber has for exploiting the e-commerce
applications that the project will develop. The main presentations of the infoday were given by the RTD
performers’ representatives. The project’s coordinator Mrs. Elena Avatangelou (EXUS) presented in detail
the SME E-COMPASS project structure, consortium, objectives and benefits for the local SMEs. Then, two
more technological presentations were held, one from Mr. Orestis Papadopoulos (ETR) regarding the data
mining application and technologies that the project develops and a second one from Dr Nikolaos
Thomaidis on the risks and threats that cyber-fraud and identity theft entails for e-commerce SMEs and

PUBLIC
Page 42 of 255
E-COMPASS
how the anti-fraud software that the project will provide can monitor and secure the transactions of the
local SMEs. The RTD performers’ presentations of the infoday are available at the annex of this document.
After the end of the first session, EPK representative Mr. Sotirios Korovilos elaborated on how the
Chambers members can participate in the project and benefit from the applications and requested from
the interested SMEs to fill-in the available form of expressing interest in the project. The expression of
interest form requested the following information: profiling and contact details of the SME and of the
participant, a declaration with respect to the two e-commerce applications offered by the project. A copy
of the expression of interest form is attached in the annex of this report. Mr. Korovilos also announced the
implementation of an electronic survey for the local SMEs active in e-commerce and motivated the
chambers members to fill-in the questionnaire that would shortly receive via e-mail. A Q&A session
followed with the local SMEs requesting more information on the project, the partners’ expertise, the
technologies implemented as well as how they can participate during the pilot phase of the project and
access the applications after the project’s completion. After the end of the last session, a small reception
with coffee and sweets was organised by the chamber for all participants and an informal networking
session took place among the local e-commerce community, executives of the chambers, the key-note
speakers and project’s partners.
The following three photographs are taken from the Infoday event and a video5 of the event is also
available from the project’s web-site and You-tube (https://www.youtube.com/watch?v=V92wOxzDr2w).
Figure 19. Infoday Panel: Mrs. Elena Spyropoulou (FP7 SME National Representative) on the podium, panel members from the left:
Mrs Vasilis Nanopoulos (President of EPK), Mr. Orestis Papadopoulos (ETR), Dr Nikolaos Thomaidis (EXUS)


This video-report (in Greek) is produced and distributed by www.eCorinth.gr including interviews with Orestis Papadopoulos (ETR),
Panagiotis Gezerlis (invited speaker GRECA) and Elena Avatangelou (Coordinator).

PUBLIC
Page 43 of 255
E-COMPASS
Figure 20. Corinth Chamber of Commerce SME members attending SME E-COMPASS Infoday
Figure 21. Corinth Chamber of Commerce SME members attending SME E-COMPASS Infoday
Figure 22. Interviews from local TV channels were taken to (from left) i. Mrs Elena Avatangelou, Project Coordinator (EXUS), ii. Mr
Orestis Papadopoulos, Fraud Specialist (ETR), iii. Mr Panagiotis Gezerlis General Director of GRECA (invited speaker)
Promotion of the Infoday Event in Corinth
Almost before the date where the actual event took place, EPK started the preparation work with the
support of the Greek RTD performers (drafting of the agenda, invitation, invitation lists, press releases,
logistics of the event and media coverage). The following figure depicts the final format of the invitation

PUBLIC
Page 44 of 255
E-COMPASS
and of the agenda. The invitation was sent by e-mail to all SME members of the Chamber and a hard copy
was posted to selected companies and key persons of the business community in the prefecture. The
invitation included a short introduction to the project, the agenda with the key-note speakers, the titles of
their presentations, the logos of the project and the consortium members.
Furthermore, a week before the infoday, press releases were published in local electronic media,
newspapers and the web-page of the Chamber6. The press release of the event is available at the annex of
the report. Furthermore all three local TV channels were invited for the media coverage of the event. Most
speakers gave interviews to the local journalists and pictures of the event and project’s work were
broadcasted at the local TV news.
Figure 23. The cover pages of the folding invitation


http://www.Corinthiacc.gr/Corinthimages/DELTIOTYPOUIMERIDAS_F20326.pdf

PUBLIC
Page 45 of 255
E-COMPASS
Figure 24. The inner pages of the folding invitation
Evaluation of the Infoday event
Both the organisation and the outcomes of the infoday were considered to be successful and full meet our
expectations. More than sixty (60) local SMEs and entrepreneurs participating in the event managed to get
up-to-date and accurate information on various areas, including:
 the E-COMPASS’s objectives and work plan
 possible ways of collaboration via the chamber at the demonstration and evaluation phases of the
project in 2015
 current trends and national statistics in e-commerce operations
 funding instruments available for SMEs in the framework of Horizon 2020.
Additionally with the extensive media coverage and promotion of the project through the event, the local
business community became aware of the E-COMPASS activities and its applications.
For the objectives of WP2, this infoday provided the opportunity to EPK and to the RTD performers to:
 List the active and interest local SMEs in the project: Around thirty five (35) expression-of-interesttype forms were filled-in, containing contact details for each SME and the responsible contact
person. Some of them, around 10, arrived from start-up e-commerce SMEs that are not yet fully
functional to provide their user requirements but can serve as potential pilot-users during the
evaluation phase of the project.
 Shortlist local SMEs for organising focus group and interviews for collecting user requirements as well
as for providing them the e-survey to fill-in.
 Map those SMEs active in e-commerce.
 Meet in person many of the e-shop owners and directors in Corinth region and get to know their
plans and current challenges.

PUBLIC
Page 46 of 255
E-COMPASS
4.1.2
Organisation of e-Survey in Corinth prefecture by EPK
The e-survey organised by the Chamber of Corinth lasted for approximately two months from the beginning
of March till early May. Initially, an official e-mail was sent to the SME members that are active in ecommerce introducing them the objectives of the project and providing the links to relevant material
(including the web-site). A second round of contacts as organised with the group of 25 out of 35 SMEs that
filled in the expression of interest form during the February’s infoday.
Nine (9) e-questionnaires were filled-in by this e-mail campaign (approx. 10% response rate). By the mid of
April, a reminder e-mail was sent to those that hadn’t replied yet and a series of follow-up phone calls were
made to selected e-merchants by the Chamber staff. Another set of five questionnaires were taken from
the participants of the focus group organised in the Chamber’s premises. In total 23 questionnaires were
filled-in and, after a systematic screening process for the consistency and validity of responses, we
managed to get 20 questionnaires that formed the sample of the e-survey in Corinth region.
The rest of this section is devoted to a presentation of the main findings of this survey.
Figure 25. Position of the Responders in the SME in Corinth
Figure 25 illustrates the position of the responders in the SME. Almost six out of ten e-responders were the
founders/owners of the e-shop. The rest were either directors or sales managers.
Company Profile
Q1: Which types of products and services do you offer?

PUBLIC
Page 47 of 255
E-COMPASS
Figure 26. Types of Products and Services offered online in Corinth

The majority (55%) of e-shops based on Corinth offer commodity products, often fast selling. Most of these
are clothing, shoes, accessories, sport and outdoor equipment, cosmetics, household, appliances,
electronics, products for pets. Exclusive slow-selling products of high quality are provided by the 25% of the
e-shops. These kinds of products are jewellery, handmade crafts and gifts, watches, special and traditional
food/ beverage. The proportion of personalised products and services is 30%, mainly referring to travel and
accommodation services such as hotels, hostels and online travel agencies (OTA).
Q2: Please tick the most relevant of the e-commerce sectors that your company belongs to.
Figure 27. e-Commerce sectors in Corinth
70% of the e-merchants are offering products to consumers, while the 15% of them are also supplying
online B2B products. Services (mostly touristic) are offered by 3 out of 10 of the SMEs.

PUBLIC
Page 48 of 255
E-COMPASS
Q3: How many full time employees dedicated to e-commerce does your company employ?
Figure 28. e-Commerce personnel in Corinth
Three out of four SMEs in Corinth employ less than 5 full-time employees, usually no more than two. One
out of five SMEs has more than five but less than ten. These SMEs are cross-border vendors or providing
international tourism services and accommodation or are hotels. Exists one SME that is active in Corinth
and has a representation-logistics office, with premises in Athens and abroad with more than 50 employees
internationally.
Q4: Which is the total annual revenue from online sales?
Figure 29. Annual online revenue in Corinth
The annual revenue from online sales for the majority (55%) of e-shops in Corinth is up to € 50K. More than
the half of this group of respondents does not exceed €10K annually. Two out of ten SMEs had revenue
between € 50K and 100K, while 5 SMEs of the sample exceeded €100K.

PUBLIC
Page 49 of 255
E-COMPASS
Q5: Which was the annual volume of orders received in 2013? (Including the orders that were not executed
for any reason)
Figure 30. Annual volume of orders in Corinth
60% of the online SMEs does not exceed more than one thousand orders per annum. 3 out of the 20 SMEs
questioned receive 5-10K orders annually and one out of ten has got an order volume of 50K.
Q6: How long has your company been doing business online?
Figure 31. Years of online business
The majority of e-merchants in Corinth sell online for no more than two years, while two out of ten started
their operation in 2013. Five SMEs of the sample (equal to 25%) have an experience of 3 to 4 years, while
four SMEs (20%) are more experienced with more than 5 years of active e-commerce.

PUBLIC
Page 50 of 255
E-COMPASS
Web Analytics Application: Price Optimization
Q7: Which are the main Websites where you compare prices?
Figure 32. Sources of price comparison
The majority of the online SMEs are comparing their prices against the competition by both relevant search
engines (70%) and by checking the competitor’s e-shops (65%). Four out of ten monitor the price index
through dedicated eMarketplaces, while only a 5% of the sample receives prices directly from the industry.
Q8: How often do you need to compare prices?
Figure 33. Frequency of price comparison among Corinth e-shops
Only one in twenty SMEs has the capacity to monitor in real-time the prices of its competitors. 15% of the
respondents need to check the price index daily, 25% every two days and 20% weekly. Five SMEs of the
sample compare prices only once per month.

PUBLIC
Page 51 of 255
E-COMPASS
Q9: How do you adjust your prices?
Figure 34. Pricing adjustment in Corinth e-shops
One in two e-shops in Corinth still adjust their e-shop prices manually. The rest follow automatic
procedures, with 60% of them being carried online in real-time.
Q10: Are you interested in a service which compares your products prices, sends alerts to you when prices
exceed certain price limits, and supports your price adjustments either manually or automatically?
Figure 35. Interest of online SMEs in Corinth for pricing optimisation application
Seventeen in twenty SMEs (85%) were interested in such an application that supports the pricing
comparison and adjustment for their e-shops.

PUBLIC
Page 52 of 255
E-COMPASS
Web Analytics Application: Visitor Behaviour
Q11: Are you currently using any service or tool for customer behaviour analysis, churn prediction and/or
prevention?
Figure 36. Use of customer behaviour analysis, churn prediction and/or prevention tool(s) in Corinth
50% of the online SMEs do not use such a tool. Three out of ten are performing these operations manually,
while only two out of ten are using a real-time and automatic application.
Q12: Are you interested in a service which analyses the customer behaviour, provides feedback on how to
improve your e-shop and supports/optimizes your cross-selling activities?
Figure 37. Interest of online SMEs in Corinth for customer behaviour analysis application
Nine out of ten SMEs find interesting such an application. One SME set as a requirement the respect of the
private data.

PUBLIC
Page 53 of 255
E-COMPASS
Q13: Are you interested in a service which analyses the behaviour of groups of customers, discovering
Figure 38. Interest of online SMEs in Corinth for an application that analyses the behaviour of groups of customers,
discovers habits, and detects new e-shopping tendencies
95% of the SMEs find interesting such an application. One SME sets as a requirement the respect of the
private data.
Anti-Fraud Application
Q14: Is fraud a concern which prevents you from the following actions? (Multi-responses question)
Figure 39. Which online actions fraud threat can possibly force Corinth SMEs to abandon?
In this critical set of questions, 7 out of 10 online SMEs in Corinth consider fraudulent activities as a
significant threat which can force them out of the e-commerce business. Furthermore approximately 3 out
of 10 acknowledge fraud as a factor that can make them stop entirely their online operations including
cross-border trade of products and services.

PUBLIC
Page 54 of 255
E-COMPASS
Q15: What is the typical proportion of fraudulent cases in your total volume of transactions?
Figure 40. Which online activities can fraud threat force Corinth SMEs to abandon
Seven out of ten Corinth online SMEs face minimal fraudulent cases (less than 0.1% of their total volume of
transactions). This may happen either because they tend to use safer ways of payment (cash on delivery,
PayPal) or their volumes of transactions or size of the e-shop is very small or became online the last 1-2
years. As observed by the data analysed, the fraudulent cases are directly related to the years of operations
and volume of transactions. Three SMEs out of twenty of the sample face serious fraud threats (3.1%-5%).
Q16: How do you deal with online payment fraud? (Multi-responses question)
Figure 41. Actions taken from online SMEs for their fraud protection
Almost one in two online SMEs reviews incoming transactions manually (one-by-one) without the support
of specialised software. One in four respondents transfer this risk to a third party (i.e. PayPal). One out of
five does not deal with online payment fraud (most of them deliver their goods only in a “cash on delivery”
mode). Only one out of four uses either a real-time (15%) or offline (10%) fraud assessment tool. This
profiling highlights the absence of actions taken against fraud as well as the ignorance of the fraud risks and

PUBLIC
Page 55 of 255
E-COMPASS
worldwide expansion (mostly among micro and newly established online SMEs that do not offer online
payment options).
Q17: What type of transaction-specific information does your antifraud personnel or antifraud tool take
into account? (Multi-responses question).
Figure 42. Parameters of a transaction that SMEs (manually or with an assessment tool) take into consideration or
collect for the validation of an e-payment
Most of the SMEs (7 out of 10monitor the customer’s country of origin, e-mail and country of the issuing
bank of the card while assessing the riskiness of a transaction. Suspicious / dubious patterns of the
customer’s online behaviour are assessed at second order. Two out of five SMEs consider the 3-D Secure
protocol as a proof of validity for each transaction. Less importance is assigned to the rest of the
parameters such as bank details, IP geo-location etc.
Q18: Which languages does your e-shop currently support?
Figure 43. E-shops supported languages in Corinth

PUBLIC
Page 56 of 255
E-COMPASS
Close to the highest majority, Corinth e-shops provide their online products and services in the Greek
language (19/20). Also most of them (79%) have its e-shop translated in English. Additional languages are
provided by a short minority, mainly providing tourism services.
4.1.3
Focus group and interviews organised in EPK
On Wednesday 12th of March 2014, a focus group took place at the premises of Corinth Chamber of
Commerce with the purpose of discussing and collecting user requirements from local e-shops. The focus
group started at 18:00 and lasted till 21:00, while the last hour was dedicated to face-to-face interviews
with selected participants. Five local e-shops and SMEs active in e-commerce were present, namely:
www.eleftheriouonline.gr, www.koxyli.gr, www.stylebrands.gr, www.v-cubes.com, www.tourix.gr. Their
analytical profile can be found at section 6.2. The first four were retailers from different markets
(jewelleries, food, clothing, and gaming) and the fifth one is an e-tourism consultant providing e-commerce
mentoring and IT services to OTAs and hotels in the region. The focus group was coordinated by
representatives of EPK, EXUS and ETR.
The focus group started with a presentation of the project and its objectives as well as with a brief
description of the applications’ technology and decision support features. Each SME representative
presented briefly its profile and operations and the main part of the focus group was based on discussions
and Q&As focused on the Greek questionnaire of the e-survey as well as the guide drafted from the RTD
performers for the focus group. Each SME participant had the opportunity to express current challenges
with online business concerning marketing, advertising, national legislation, transactions and payment
methods, analytics, typical cases of fraud and ways to deal with them. In the last session, two working
groups were formed in which more in-depth discussions took place on the project’s applications and how
SMEs can benefit from them. Through the interviews, the facilitators helped SMEs to define more precisely
their user requirements and identify the online data that they collect and potentially can be processed by
the project’s applications. The following paragraph illustrates in tables the responses and insights collected
from the focus group discussions and personal interviews. The following tables presents the responses
regarding the web-analytics application with the use of data mining techniques. The following paragraph is
concerned about the fraud management practices and concepts of the EPK SMEs.
Web Analytics Application.
 Current State in Web Analytics: Visitor Behaviour
Table 5. Responses from EPK concerning current state in web analytics: visitor behaviour
Questions
What metrics are you analyzing?
(focusing on visitors‘ behaviour)
What tools do you use?
(e.g. e-shop, Web analytics)
How often do you check the metrics?
(frequency)

Responses









website origin of visitors (1 e-shop owner out of 4)
geographical origin of visitors (1 e-shop owner out of 4)
popular products / product trends (2 e-shop owners out of 4)
duration of visit (1 e-shop owner)
own ranking on Google (2 e-shop owners)
Google Analytics (3 e-shop owners out of 4)
GRIPS: Facebook-analytics (2 e-shop owners out of 4)
weekly (2 e-shops out of 4)
daily (1 e-shop out of 4)
PUBLIC
Page 57 of 255
E-COMPASS
What actions/activities do you derive?
How much effort do you spent for
web analysing?
What visitor’s/customer’s data do you
store in the e-shop?











content optimizing (1 e-shop out of 4)
target market analysis (1 e-shop out of 4))
marketing activities (3 e-shops out of 4)
4 hours per week (1 e-shop owner)
1 man day per months (3 e-shop owners)
names (4 e-shop owners out of 4)
addresses (4 e-shop owners out of 4)
email addresses (4 e-shop owners out of 4)
phone no. (4 e-shop owners out of 4)
frequency of spends (1 e-shop owners out of 4)
frequency of visits (2 e-shop owner out of 4)
 Current State in Web Analytics: Competitor’s Analysis
Table 6. Responses from EPK concerning current state in web analytics: competitor’s analysis
Questions
What competitors‘information are
you analysing?





competitors’ product prices (2 e-shop owners out of 4)
competitors’ product quality (2 e-shop owners out of 4)
competitors’ ranking on Google (1 e-shop owner out of 4)
competitors’ campaigns on Google (1 e-shop owner out of 4)
nothing (2 e-shop owners out of 4)
(e.g. price comparison portal)

no tool, do it manually (4 e-shop owners out of 4)
How often do you check the
competitors‘information? (frequency)



weekly (1 e-shop owner out 4)
monthly (2 e-shops out of 4)
less frequently (1 e-shop out of 4)

Changes to own prices (2 e-shop owners out of 4)
How much effort do you spent for the
competitors‘analysis?

1/2 man-day per month (2 e-shop owners out of 4)

Responses
PUBLIC
Page 58 of 255
E-COMPASS
 New Requirements for Web Analytics: Visitor Behaviour
Table 7. Reponses from EPK concerning new requirements for web analytics: visitor behaviour
Questions
Responses








What additional metrics do you wish
to analyze?
How often would you like to check the
metrics? (frequency)
customers' profiles incl. topics of interest (4 e-shop owners out of 4)
precise amount of traffic (3 e-shop owners out of 4)
visibility of own e-shop in the www (4 e-shop owners out of 4)
daily (1 e-shop owner out of 4)
weekly (2 e-shop owners out of 4)
monthly (1 e-shop owner out of 4)
looking for new channels (3 e-shop owners out of 4)
campaign analysis (2 e-shop owners out of 4)
 New Requirements for Web Analytics: Competitor’s Analysis
Table 8. Responses from EPK concerning new requirements for web analytics: competitor’s analysis
Questions
Responses





What competitors‘information do you
wish to analyse?





How often would you like to check
the competitors‘information?
What actions/activities do you
derive?
competitors' prices (3 e-shop owners out of 4)
amount of sold products (2 e-shop owner out of 4)
traffic on competitors' websites (2 e-shop owners out of 4)
keywords for competitors' websites (1 e-shop owner out of 4)
visibility of competitors (Google, Social Media) (2 e-shop owners out
of 4)
weekly (3 e-shop owners out of 4)
monthly (1 e-shop owner out of 4)
Coupons (2 e-shop owners out of 4)
recommendations (1 e-shop owner out of 4)
more targeted campaigns (3 shop owners out of 4)
 Pilot Users: Competitor’s Analysis
Table 9. Pilot users willingness in EPK
Questions
Could you provide us an access to your
metrics? (e.g. access to your Web
analytics)
May we install an additional web
analytics tool?
Would you be able to provide us the
database schema of your e-shop?
(maybe a couple of data sets as
examples)

Responses


yes (3 e-shop owners out of 4)
no ( 1 e-shop out of 4 N/A metrics)


yes (2 e-shop owners out of 4)
not applicable (2 e-shop owners out of 4)

yes (4 e-shop owners)
PUBLIC
Page 59 of 255
E-COMPASS
Analysis of the focus group and interviews
The majority of the e-shops interviewed do a weekly check on their current metrics (visitor’s data and
competitor’s data). The tool that is most often used during this process is Google Analytics. Stored data in
the e-shop systems, which could be used for the data mining use case, are names, addresses, phone
numbers of the customers as well as the frequency of expenditure and visits.
None of the interviewed e-shop owners has an automatic tool for competitors’ analysis; they typically
collect information on competitors’ pricing policies and marketing tactics manually by checking the prices
and quality of the same or similar products on other suppliers’ website.
They would like to automatically provide coupons to customers as well as to be able to make more targeted
spending on digital marketing campaigns (i.e. Google words and Google ads).
In general, all 4 e-shops interviewed as well as Tourix that participated in the focus group discussion, were
very positive in testing and acquiring a data-mining application with the features of SME E-COMPASS datamining application. Tourix customers, such as hotels and local OTA’s will also be supported for their daily
operations with such a tool. Furthermore, only one e-shop (V-Cubes) is an advanced user of e-commerce
relevant applications and has well established procedures for monitoring markets, visitors and competitors
but only via Google Analytics and without the use of any data-mining technique.
All e-shops participating in the focus group were to some extent aware of the risks entailed by e-commerce
fraud. E-shops selling cross-border had many more fraud cases to share than the ones doing business
online on a local or national level. All of the participants have experienced at least one chargeback case.
Only one e-shop (the one that is active in international selling) was fully aware of the fraud dimensions and
technical difficulties for its online monitoring. None of the e-shops were using an in-house or commercial
anti-fraud application, but from the discussion it was evident that all e-shops were applying common rulesof-thumb for manually dealing with cybercriminals. Tourix which consults hotels and local OTAs revealed
that in the tourism service industry its customers face fraud weekly in their operations. For tourism services
NCP fraud cases that are not reimbursed via charge backs are very costly for the service providers, since the
commission fee that they receive from flight or ferry tickets is very limited. The discussion of the project’s
anti-fraud application was focused on the features, rules development and customisation, data
management and compatibility of the system with the e-shops’ platform and technologies. Furthermore, all
four e-shops revealed their willingness to provide samples of transactions with information about most of
the parameters that are automatically stored (excluding those that are generally considered as confidential
and non-disclosable). Tourix agreed to provide transactions data from hotels and OTAs after receiving
authorisation from its clients. In section 5.1 we present analytically the user requirements and in section
6.1 the parameters of the data that EPK SMEs offered for the ant-fraud application design and
development.
4.2
Results and Analyses from British Chamber (HALTON)
The local methodological approach for the Halton region (UK) comprises interviews with SME partners ETravel in Greece and MeetNow in Germany in order to quickly get an in-depth understanding of the current
situation and challenges of e-shops in terms of sales with focus on payment/fraud, price and customer

PUBLIC
Page 60 of 255
E-COMPASS
behaviour analysis. The knowledge which have been gained from the SME partners in Germany (second
largest e-commerce market in Europe after UK) and in Greece (second fastest growing market in Europe
after Turkey), built the basis for preparing the activities in UK. Three qualitative interviews have been
conducted within February and March in order to shape the services which can be offered to e-shop
owners (concept for the future) adapted to their real-world needs (initial situation). The interviews were
held based on a structured interview guideline which has been developed and discussed among the RTD
partners within the E-COMPASS project.
The results and the know-how which could be derived from the interviews have built the basis for
developing the quantitative survey by applying an online-questionnaire. The online-questionnaire contains
questions which clarify the e-shop as well as its background and focus on relevant aspects of anti-fraud and
data mining. The online-questionnaire has been published beginning of April prior to the focus group
workshop.
4.2.1
Organisation of Infoday and Focus Groups by HALTON and FRA
In mid of April, a focus group workshop with 8 e-shop owners took place. The e-shop owners were grouped
into two groups with 4 e-shop owners each. The two topics, anti-fraud and data mining, were introduced
and discussed for one hour within two groups in parallel sessions. Afterwards the groups were swapped
and the topics discussed in this new constellation.
For the data mining part, a workshop guideline was developed which allows the structured gathering of
information of (1) the initial situation in terms of current challenges, used business processes, applied ITsystems with a strong focus on e-shops and web analytic tools, (2) the concept of the future situation which
includes the new online data mining services and their functions, improved processes and additional
service-based solutions, and (3) finally the motivation of making pilot users and their capability and
willingness of sharing relevant information, such as web analytics metrics with the E-COMPASS online data
mining services.
Figure 44. Time line schedule for collection information from Halton companies
The main interviews with e-Shop owners in UK have been conducted on the 15th of April in Halton. A
picture of the workshop and its participants is shown below.

PUBLIC
Page 61 of 255
E-COMPASS
Figure 45. Focus groups info days in Halton
Interviews as basis for online-questionnaire and focus group workshops
In order to prepare the online-questionnaire and the focus group workshop, interviews with MeetNow
(Germany), e-Travel and A&E Fornaro (both from Greece) have been conducted. The main objective was
the identification of the crucial aspects and issues for e-commerce in general and data mining in specific
which drive owners of small e-shops. MeetNow consult e-shops in Germany in terms of their marketing
activities. With its existing customers they have a broad overview over many internet-based businesses in
Germany.
E-Shops in Germany offer in average a range of 1,000 articles. The product descriptions are taken from the
manufacturers. For approx. 50 articles an individual maintenance of the offers and the content is
conducted in order to differentiate the articles form the offers of competitors.
E-shop solutions offer different functionalities. A premium e-shop is for example Intershop, medium range
e-shops are Oxid, Magento, Gambio, Kosmoshop, Omekoshop and Shopware, and small e-shop solutions
are Strato and eBay. An average e-shop has a turn-over of 30 Mio. Euro and employs 4 people. The profit
margins of electronic appliances are very tight, whereas for clothes they vary a lot and may be greater.
Price analysis and comparisons with competitors
A differentiation between standard products which can be easily compared by checking product ID, such as
GTIN and EAN, products which need to be compared on the basis of their attributes, and customer-specific
products. Especially, for the first group of products price comparisons are often made on an automated
basis. Here, especially the medium to large e-shops use tools which provide them with appropriate
information. For small e-shops, those tools are too complicated and not very wide spread used among
companies with this size. For the second type of products, the price comparison tools get rare. Therefore,
most of the e-shops limit their price comparisons on a few competitors and products and manually conduct
those activities.

PUBLIC
Page 62 of 255
E-COMPASS
An easy to use web-based tool for price comparisons among e-shops was considered as very valuable. The
tool needs to especially focus on supporting small e-shops. The owner should be able to define minimum
and maximum thresholds for the price analysis. If a competitor price exceeds the thresholds, the owner is
notified. An automated price adjustment is technical possible to implement. Different pricing strategies
need to be implemented and must be configurable by the e-shop owners. However, the e-shops might not
be willing to provide write access to the e-shop to an additional tool on an automated basis.
Customer Behaviour Analysis
In order to optimize an e-shop concerning the communication with its users, the offered products, and the
personalization of information, an understanding of the e-shop users and their objectives is very helpful.
Therefore, the development of user profiles and clusters depending on the user behaviour on the website
and the e-shop becomes interesting. Since many users just stay for a short time (one or two clicks), it is
important to analyse the search terms which are applied prior to enter the e-shop. A clustering of users
depending on their purposes might be useful for optimizing the e-shop and the communication with the
users. When discussing those issues with e-Travel and A&E Fornaro, the feedback towards the issues which
have been addressed was very similar.
Three of the participating e-shops in Halton agreed to get involved into SME E-COMPASS as pilot users.
They will provide user behaviour data which are collected via Google Analytics. The data are real-world
data from the e-shops and an interface between their Google Analytics tools and the SME E-COMPASS
cockpit which allows a real-time connectivity for the data mining services needs to be discussed among the
involved project partners and the pilot users. The collection of additional data such as competitors’ product
prices are implemented with a web scraping tool.

PUBLIC
Page 63 of 255
E-COMPASS
4.2.2
Organisation of e-Survey in Halton region by HALTON and FRA
The questionnaire of our survey was filled out by 15 e-shop owners. 10 (67%) of these e-shops are working
in the B2C sector and 5 e-shops (33%) in the B2B sector.
1) Which types of products and services do you offer?
7%
commodity products
13%
exclusive products
46%
7%
configurable products
personalised products
online network
membership
27%
Figure 46. Question 1) in online questionnaire. English version for HALTON
Nearly half of the e-shop owners questioned are selling commodity products, almost 27% sell exclusive
products, one or two e-shops are trading with personalised or configurable products and one e-shop is a
member of an online network.
2) Please tick the most relevant of the e-commerce sectors that your company belongs to
33%
B2C
B2B
67%

PUBLIC
Page 64 of 255
E-COMPASS
67% of the e-shop owners questioned quoted to work in B2C sector, 33% in B2B sector.
3) Which kind of product do you offer?
The products offered by the e-shops questioned are very diverse.
-
Women Clothes
-
Solid Fuel
-
Packaging Supplies
-
Marine Safety Equipment
-
Health Products and Pharmaceuticals, Pet products
-
Parcel Delivery Options
-
Books
-
Group Memberships
-
Babies & Children Wear & Bespoke Nursery Items
-
Audio Visual Equipment
-
Domestic Appliance Consumables
-
Bathroom Products
-
Print and Design
-
Sport Equipment, Specialist Gloves
-
Fashion Accessories
4) How many full time employees dedicated to e-commerce does your company employ?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
1
2-3
4-5
6-10
> 50

PUBLIC
Page 65 of 255
E-COMPASS
More than 50% of the e-shops are one-person enterprises, 20% employ 4 to 5 employees. Only one e-shop
has more than 50 employees. Here, the main target group of SME E-COMPASS has been perfectly reached
and motivated to participate in the project activities.
5) Which is the total 2013 annual revenue from online sales?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
< 10K€
10K-50K€
50K-1M€
The total annual revenue from online sales in 2013 of more than the half of the e-shops was between
10,000 € and 50,000 €, 33% had an annual revenue of less than 10,000 € and the annual revenue of 13% (3
e-shops) has been between 50,000 € and 1 Mio. €.
6) Which is the total 2013 annual revenue from total (online and offline) sales?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
< 10K€
10K-50K€
50K-1M€

PUBLIC
Page 66 of 255
E-COMPASS
The total annual revenue from online and offline sales in 2013 of 40% of the e-shops was between 10,000 €
and 50,000 €, 27% had an annual revenue of less than 10,000 € and the annual revenue of 33% (3 e-shops)
has been between 50,000 € and 1 Mio. €.
7) Which was the annual volume of online orders received in 2013?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
< 100
101 - 1.000
1.001 - 5.000
10.001 - 50.000
80% of the e-shops have an annual volume of max. 5,000 online orders, 20% (3 e-shops) have an annual
volume between 10,000 and 50,000 online orders.
8) How long has your company been doing business online?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
< 1 year
1-2 years
3-4 years
5-10 years
> 11 years

PUBLIC
Page 67 of 255
E-COMPASS
Most (10 e-shops or 67%) of the e-shop owners questioned run online business between 1 and 4 years, 3 of
the e-shop owners (20%) are in online business even for less than one year. Therefore, e-shops have been
addressed in Halton which are mostly of small size and quite new in the business. Especially, those e-shops
need the support by chambers and other public organizations in order to successfully compete against the
larger player in the e-commerce market.
9) Which languages does your e-shop currently support?
All e-shops questioned mentioned to only support English language.
10) Which are the main Websites where you compare prices?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
price/product
search engines
e-marketplaces
websites of
competitors
market
intelligence
solutions
none
The main websites, which are used for the price comparisons are the websites of direct competitors. 80%
of the e-shop owners questioned are looking on their competitors’ websites for comparing their prices.
Other websites for price comparisons are e-market-places (20%), price and product search engines (7%) or
market intelligence solutions (7%). One e-shop owner (7%) does not do price comparisons, yet.
The main search engines used for price comparison by the e-shop owners questioned are shown in the
following figure.

PUBLIC
Page 68 of 255
E-COMPASS
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Google
Yahoo
none
Search engines used by the e-shop owners questioned are Google and Yahoo. 27% of the e-shop owner use
Google, 7% use Yahoo, the others do not use a search engine for making price comparisons. As one can see
in Figure 55 the main eMarketplaces, which are used by the e-shop owners questioned are E-bay and
Amazon.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
ebay
Amazon
none
27% of the e-shop owners use E-bay, 13% use Amazon, the others do not use an e-marketplace for their
price comparisons.

PUBLIC
Page 69 of 255
E-COMPASS
11) How often do you need to compare prices?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
real-time
daily
every 2
days
weekly bi-weekly monthly
never
Most of the e-shop owners do the price comparison weekly (33%) or monthly (33%), one e-shop owner
does the price comparison daily and another does it in real-time. Two e-shop owners do not compare their
prices. In this case, many of the e-shop owners do not realize price changes of competitors and thus, will
not have the opportunity to appropriately react on those competitor activities and develop an appropriate
price strategy.
12) How often would you like to compare prices?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
real-time
daily
every 2
days
weekly bi-weekly monthly
never
If the e-shop owners would have more time or an electronic tool for comparing their prices, 40% (6 e-shop
owners) would do the price comparison daily, 20% (3 e-shop owners) would do it weekly and another 20%
would do it monthly. One e-shop owner would do it bi-weekly, another e-shop owner would do it in realGrant Agreement 315637

PUBLIC
Page 70 of 255
E-COMPASS
time and another one would still not do it. However, the e-shop owners recognize the opportunity of
regularly monitor competitors’ product prices and would like to more regularly use an appropriate
instrument for their business activities.
13) How do you adjust your prices?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
manually
automatically, offline
automatically, real-time
The adjustment of prices is done manually by the majority of the e-shop owners questioned (93%). Only 2
are doing that automatically by using a software tool, one of them is doing that online and also in real-time.
a. How many products do you compare regularly at competitors e-shops?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Figure 59. Question 13a) in online questionnaire. English version for HALTON

PUBLIC
Page 71 of 255
E-COMPASS
Currently 40% (6 e-shop owners) are comparing 5 to 10 of their products with products of their
competitors, 13% (2 e-shop owners) are comparing 11 to 20 products, 20% are comparing more than 100,
and one e-shop owner (7%) is comparing only one product. The other e-shops (3 e-shops) have not
provided information to that question.
b. How many products would you like to compare regularly at competitors e-shops?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Figure 60. Question 13b) in online questionnaire. English version for HALTON
If the e-shop owners would have more time or an electronic tool for comparing their prices, the e-shop
owners tend to compare more prices. Thus, 20% of the e-shop owners would compare 5 to 10, 13% would
compare 20 to 50, another 13% would compare as many products as possible and additional 13% would
compare all of their products. 1 e-shop (7%) have not provided any information to that question.
c. How many competitors’ e-shops do you observe?
Figure 61. Question 13c) in online questionnaire. English version for HALTON

PUBLIC
Page 72 of 255
E-COMPASS
47% of the e-shop owners questioned observes the products of 5 to 10 competitors for price comparison.
27% of the e-shop owners compare their products with products of 3 to 4 competitors. One e-shop owner
compares his products with the products of 10 to 20 competitors and one e-shop owner does not compare
his products.
d. How many competitors’ e-shops would you like to observe?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Figure 62. Question 13d) in online questionnaire. English version for HALTON
If the e-shop owners would have more time or an electronic tool for comparing their prices, all e-shop
owners would compare their products with products of their competitors. Thus, 33% would compare their
products with the products of 5 to 10 competitors, another 33% would compare their products with
products of 10 to 40 competitors, one e-shop owner would compare his products with the products of as
many competitors as possible and 2 e-shop owners (13%) would compare their products with products of
all their competitors.
e. Do you use a service or an electronic tool to observe and compare the prices of
competitors?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
manually
electronically
Figure 63. Question 13e) in online questionnaire. English version for HALTON

PUBLIC
Page 73 of 255
E-COMPASS
87% (13 e-shop owners) compare the prices of their products with the prices of their competitors manually
by watching the websites of their competitors. Only 13% (2 e-shop owners) use a software tool for doing
that comparison electronically.
14) Are you interested in a service which compares your products prices, sends alerts to you when
automatically?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
interested
conditionally interested
not interested
60% of the e-shop owners questioned are interested in a service for price comparison and alerting, 13% of
the e-shop owners are conditionally interested and 27% are not interested in such a service.
15) Is fraud a concern which prevents you from?
27%
33%
expanding online market
access to the entire EU
selling products/services
online
using online payment
transactions or e-card
systems
40%

PUBLIC
Page 74 of 255
E-COMPASS
27% of the shop owners want to expand their online product offerings to the entire EU market and 33% are
already using some form of online payment transaction
16) What is the typical proportion of fraudulent cases in your total volume of transactions?
7%
< 0.1%
0.2% - 1%
93%
7% of the e-shop owners have observed a fraudulent case instance that ranged in 0.2% to 1% of their total
volume of transactions.
17) How do you deal with online payment fraud?
13%
Manual review
6%
7%
7%
Automatic
Automatic - real time
Transferring
67%
Do not deal with online
payment fraud
67% of the participants do not deal with online payment fraud as they have the feeling and confidence that
this is a concern by their third party payment provider. 13% of the users review manually the transactions
and a total 7% plus 6% are using real-time automatic or semi-automatic mechanisms.

PUBLIC
Page 75 of 255
E-COMPASS
18) Do you use specific software to deal with online payment fraud?
13%
Yes
No
87%
87% of the e-shop owners are not using a specific software or service to deal with online payment fraud.
19) Do you use your own software or assessment method to deal with online payment fraud?
All e-shop owners questioned quoted to do not have an own software or assessment method.
20) What type of transaction-specific information does your antifraud personnel or antifraud tool take
into account?

3D secure (2 e-shops)

Repeat transaction (1 e-shop)

Handled by PSP (2 e-shops)

Suspicious or dubious behaviour (2 e-shops)

E-mail addresses (2 e-shops)

Device identity (2 e-shops)

None (2 e-shops)

Not applicable (2 e-shops)

PUBLIC
Page 76 of 255
E-COMPASS
21) What do you monitor, plan to monitor or do not monitor on your Website/e-shop? Information of
the visitor's origin
visitor's geographical origin
8
4
3
visitor's origin: search phrases for search
engines
7
visitor's origin: entry pages
7
6
2
visitor's origin: keywords for search engines
7
6
2
visitor's origin: website
7
6
1
4
5
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
montoring
plan to monitor
not monitoring
Most of the visitor’s origin information questioned is either already monitored by the e-shop owners or
they plan to monitor them. Only the visitor’s origin website is not monitored by 5 e-shop owners.
22) What do you monitor, plan to monitor or do not monitor on your Website/e-shop
visitors' attribute
number of visits per visitor
6
technical equipment of visitors (e.g. browser
version)
6
5
9
number of new visitors
9
number of visitors per week
4
4
number of frequent visitors
Information of
5
6
0
5
11
1
4
0
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%100%
montoring
plan to monitor
not monitoring
Interesting visitors’ attributes, which are not monitored by nearly 30% of the e-shop owners are number of
visits per visitor and the visitors’ technical equipment.

PUBLIC
Page 77 of 255
E-COMPASS
visitors' behaviour
Information of
most common view sequences (click paths)
6
6
3
most common exit pages
6
6
3
keywords within own e-shop search
6
duration of visit
6
5
4
6
number of page views per visit
3
9
most often viewd pages
5
8
1
7
0
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
montoring
plan to monitor
not monitoring
Concerning the visitors’ behaviour only the number of page views per visit and the most often viewed
pages are already tracked by the e-shop owners or they plan to monitor it.
purchasing behaviour
average time of stay until purchasing products
6
5
Information of
4
number of visitors who break up the
purchasing process
4
7
4
number of visitors who put a product into the
basket
4
7
4
average value of shopping cart
8
number of visitors who did a purchase
5
9
2
5
1
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%100%
montoring
plan to monitor
not monitoring
Concerning data of purchasing behaviour nearly all e-shop owners monitor or plan to monitor only the
average value of the shopping cart as well as the number of visitors who did a purchase.

PUBLIC
Page 78 of 255
E-COMPASS
for monitoring
mouse-tracking 0
comparative tests (e.g. A/B-tests)
5
10
1
6
form field analysis 0
qualitative user survey
9
7
2
categorization of user groups
montoring
6
6
5
0%
8
7
4
access of mobile devices
8
6
page oriented user feedback 0
New function
5
8
2
10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
plan to monitor
not monitoring
Concerning new functions for monitoring only the access of mobile devices is monitored by the e-shop
owners. For all other functions there is currently no monitoring by nearly all e-shop owners. For mouse
tracking, form field analysis and page oriented feedback the e-shop owners do currently not have a solution
for monitoring.
26) How much do you invest in your web activities?
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
< 10K€
10K€ - 50K€
The amount of investment for web activities of the e-shops questioned is less than 10,000 €, only one eshop spent between 10,000 € and 50,000 € for online activities.

PUBLIC
Page 79 of 255
E-COMPASS
27) What additional information are you interested in?
mouse-tracking
comparative tests (e.g. A/B-tests)
form field analysis
page oriented user feedback
qualitative user survey
categorization of user groups
access of mobile devices
average time of stay until purchasing products
number of visitors who break up the purchasing process
number of visitors who put a product into the basket
average value of shopping cart
number of visitors who did a purchase
most common view sequences (click paths)
most common exit pages
keywords within own e-shop search
duration of visit
number of page views per visit
most often viewd pages
number of visits per visitor
technical equipment of visitors (e.g. browser version)
number of frequent visitors
number of new visitors
number of visitors per week
visitor's geographical origin
visitor's origin: search phrases for search engines
visitor's origin: entry pages
visitor's origin: keywords for search engines
visitor's origin: website
percentage of e-shop owners
who wish to have that
information
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
All questioned data for visitors’ analysis, but only the visitors’ origin website, is very relevant for the e-shop
owners questioned, because more than 70% wish to have that data.
28) Are you interested in participating as a pilot user in E-COMPASS
Nine (or 60%) of the e-shop owners questioned are interested in participating as a pilot user in E-COMPASS.
4.2.3
Results of Focus Groups and Interviews by HALTON and FRA
Anti-fraud
During the focus group the participants were introduced with the Anti-Fraud fundamentals of:
 Automated screening,
 Manual Review,
 Order dispositioning (Accept/Reject),
 Fraud Claim Management and
 Tuning and Management
The participants were also informed with very interesting statistics about the current situation in the UK.
Moreover the participants were informed with antifraud terms such as country blocking; perceived risky
code etc. and they were shown in detail the top anti-fraud high risk indicators in the UK.

PUBLIC
Page 80 of 255
E-COMPASS
With fraud management budgets static for the majority in 2013 in the UK, it was made clear by the
participants that is very hard for SME’s to invest money or person resources for manual fraud detection.
New markets and sales channels are explored by the participants and despite the fact that none of the
participants has faced a serious fraud threat, it was apparent that growth in sales statistically raises fraud
events. A threat that in a local and controlled e-commerce environment is less imminent.
Thus identifying good customers sooner, building on positive data and lists, removing friction from the
checkout process. In doing so, businesses can better control short and longer term profits and improve the
overall experience. In addition, improving automated front line screening can dramatically reduce the need
for manual review, allowing budgets to be reallocated elsewhere.
The participants throughout the participation asked a lot of questions and discussion and exchange of
experiences added a lot of value for the focus group. Following are the main points constituting the
participants’ profile.
1. No more than 2 employees dedicated part time to e-commerce.
2. The total revenue from the online sales had a range from 2,500 Euros to 250,000 Euros.
3. All participants did not have more than 5 year’s online presence.
4. Because of the local sales channel and small revenues fraud was not a limitation for the participants
to sell online.
Current Fraud Practices and issues.
5. One issue of email scan was mentioned from a participant and all agreed that a lot of attention
should be paid to the problems of phishing and affiliate channel schemes.
6. All incoming transactions are manually reviewed before the companies proceed to send the item that
has been bought.
7. All participants pass the transaction risk to a third party provider and all are using PayPal in
combination with their bank (Northwest Bank in many local cases).
8. The ecommerce platforms used are wecommerce (wordpress), Magento, Zencard and Bigcommerce.
All have modules connecting to third party providers such as Paypal.
9. All participants were interesting in high risk fraud indicators such as postal addresses, postcodes,
non-UK IP addresses and multiples of same items in the basket (two of the participants deal with
heavy items 80 pounds median per delivery).
Data Mining
The focus group interviews were held in mid of April at the Halton Chamber of Commerce & Enterprise with
a focus group of 8 e-shop owners of the following product sectors:
 printworks
 clothing
 venture packaging
 human hair extensions, fashion accessories, cosmetics, scarves, jewellery, footwear

PUBLIC
Page 81 of 255
E-COMPASS
 gloves, mittens, medical devices
 cue sports, indoor leisure
 marine safety equipment
Table 10. Responses from HALTON concerning current state in web analytics: visitor behaviour
Questions
What metrics are you analysing?
(frequency)
web analysing?
Responses













website origin of visitors (2 e-shop owners)
geographical origin of visitors (2 e-shop owners)
popular products/product trends (1 e-shop owner)
most visited e-shop sites (1 e-shop owner)
duration of visit (1 e-shop owner)
followers (1 e-shop owner)
own ranking on Google (1 e-shop owner)
Google Analytics (4 e-shop owners)
GRIPS: Facebook-analytics (2 e-shop owners)
Zen Cart (1 e-shop owner)
Terapeak (1 e-shop owner)
Bigcommerce (1 e-shop owner)
ClockworkCommerce (1 e-shop owner)

weekly (3 e-shops)




content optimizing (2 e-shops)
target market analysis (2 e-shops)
marketing activities (1 e-shop)
Spending many times in learning to analyse different things (1 eshop owner)
names (5 e-shop owners)
addresses (5 e-shop owners)
email addresses (5 e-shop owners)
phone no. (5 e-shop owners)
frequency of spends (1 e-shop owner)
frequency of visits (1 e-shop owner)






Table 11. Responses from HALTON concerning current state in web analytics: competitor’s analysis
Questions
What competitors‘ information are
you analysing?
competitors‘ information?

Responses





competitors’ product prices (3 e-shop owners)
competitors’ product quality (3 e-shop owners)
competitors’ ranking on Google (1 e-shop owner)
competitors’ campaigns on Google (1 e-shop owner)
nothing (1 e-shop owner)

no tool, do it manually (4 e-shop owners)

when having time (1 e-shop owner)
PUBLIC
Page 82 of 255
E-COMPASS
(frequency)

not applicable
competitors‘ analysis?

couple of days (1 e-shop owner)
What are the main competitors‘ eshops which you consider?



10 products similar to/same as own products for 10 competitors
same products as own products of 3 competitors
products of one direct competitor
Table 12. Reponses from HALTON concerning new requirements for web analytics: visitor behaviour
Questions
Responses







to analyse?
customers' profiles incl. topics of interest (3 e-shop owners)
precise amount of traffic (1 e-shop owner)
visibility of own e-shop in the www (1 e-shop owner)
daily (1 e-shop owner)
weekly (2 e-shop owner)
looking for new channels (2 e-shop owners)
campaign analysis (1 e-shop owner)
Table 13. Responses from HALTON concerning new requirements for web analytics: competitor’s analysis
Questions
Responses
What competitors‘ information do
you wish to analyse?





competitors' prices (2 e-shop owners)
amount of sold products (1 e-shop owners)
traffic on competitors' websites (1 e-shop owners)
keywords for competitors' websites (1 e-shop owners)
visibility of competitors (Google, Social Media) (1 e-shop owners)
the competitors‘ information?

weekly (3 e-shop owners)
derive?



vouchers (2 e-shop owners)
reward cards (1 e-shop owner)
recommendations (1 e-shop owner)
Table 14. Pilot users in HALTON
Questions
analytics)
analytics tool?

Responses


not applicable (3 e-shop owners)



PUBLIC
Page 83 of 255
E-COMPASS
examples)

Analysis of the online questionnaire
The analysis of the survey data shows that the majority of the e-shop owners questioned are small
enterprises with 1 to 5 employees in B2B or B2C sector, selling commodity or exclusive products, being less
than 4 years in online business, having a maximum online order number of 5,000 orders a year and a
annual revenue of max. 50,000 €. They invest less than 10,000 € in their web activities.
The most of the e-shop owners questioned, do competitors observation and price adjustment manually.
The main websites for price comparison are the websites of the competitors and eMarketplaces. Used
search engines for price comparison are Google and Yahoo, used marketplaces for price comparison are
EBay and Amazon.
The current average number of products is between 1 and 20 products, having the possibility to do the
comparison automatically the number would be higher. The current number of competitors for price
comparison is between 3 and 10 competitors, some e-shop owners even do not compare their products
with competitor’s products. Having the possibility to do the comparison automatically every e-shop owner
would do the comparison and would compare the own prices with more competitors than they actually do.
The current frequency of price comparison is mostly weekly or monthly, having the possibility of an
automated comparison the frequency would be daily or weekly. In this regard, 60% of the e-shop owners
are interested in a service for price comparison and alerting. Additional 13% are conditionally interested in
such a service. Named conditions are the guarantee that it works properly or that it works for the e-shop of
a special competitor. Furthermore the results of the survey show that nearly all of the e-shop owners
currently do the price comparison manually and that they would the comparison more often as well as for
more products and competitors as they actually do. These results show, that an online data mining tool for
comparing the prices of the own e-shop with the prices of competitors is very relevant for the e-shop
owners.
Less than 30% of the e-shop’s owners do currently monitor and categorize user groups from their visitors’
data, about 40% plan to do that and more than 30% do not monitor it, yet.
The results of the survey show that all data for visitors’ analysis is very important for the e-shop owner.
Excepting the visitors’ origin website, all visitors’ metrics questioned are relevant to the e-shop owners and
more than 70% of the e-shop owners wish to have that data.
Analysis of the focus group interviews
The majority of the e-shops interviewed do a weekly check of their current metrics (visitor’s data and
competitor’s data). The most often used tools for the check are Google Analytics and GRIPS: Facebookanalytics.
Currently derived actions are content optimising, target group analysis and marketing activities. Stored data
in the e-shop systems, which could be used for the data mining use case, are names, addresses, phone
numbers of the customers as well as the frequency of spends and visits.

PUBLIC
Page 84 of 255
E-COMPASS
The majority of the interviewed e-shop owners do not have a tool for competitors’ analysis, but do it
manually by checking the prices and quality of the same or similar products on the competitors’ websites.
They weekly track 1 to 10 competitors and would not check more competitors or more often if they could
do it automatically. They would like to automatically provide voucher or reward cards to customers, who
have ordered a lot or they would like to automatically recommend products to visitors based on visitors’
analysis. Half of the e-shop owners interviewed would like to have the service for visitors’ and competitors’
analysis integrated into their e-shop system.
Conclusions of the analysis results
The analysis results of the survey and the focus group interviews lead to the following requirements for the
data mining and analysis tool (E-COMPASS cockpit) which will be created in the data mining use case:
 The tool should be able to access the price and product data of the competitors’ websites given by
the e-shop owners.
 The tool should be able to access the price and product data of the eMarketplaces mainly used for
price comparison by the e-shop owners questioned, which are Amazon and EBay.
 The tool should be able access competitors’ price and campaign data as well as the competitors’
ranking on Google and Yahoo, which are the search engines mainly used by the e-shop owners
questioned.
 The tool should do the price and product comparison automatically.
 The tool should be able to compare the prices of as much products as possible.
 The tool should be able to schedule the price and product comparison like a Windows Task or a
Cron Job.
 The tool should be able to derive predefined actions (e.g. recommendations, reward cards,
vouchers) from visitors’, customers’ and competitors’ data by using predefined rules. The e-shop
owners should be able to choose from predefined rules and actions.
 The tool should be able to identify products on the websites of the competitors, which are similar
or the same like that of the e-shop owners.
 The tool should be simple to integrate the existent software solutions of the e-shops.
The results of the focus group workshop show that the e-shop owners are especially interested in having or
generating profiles of customers including their topics of interest. 50% of the e-shop owners questioned
mentioned the relevance of that metric to their business.
For deriving customers’ and visitors’ profiles they need to monitor visitors’ and customers’ data.
For 33% of the e-shop owners questioned competitors’ prices are very important.
On these metrics the e-shop owners questioned would like to derive automated actions like vouchers,
reward cards or recommendations for customers. For running these actions real-time data would be
needed.
The required data and actions of the focus group shows the importance of a data mining tool which offers
the functions delivering these data as well as the required actions.

PUBLIC
Page 85 of 255
E-COMPASS
4.3
Results and Analyses from Spanish SME Association GAIA
4.3.1
Organisation of e-Survey and Focus Groups by GAIA and CIC
The local methodology proposed for the analysis of requirements has been to conduct an electronic survey
with a sample of 86 SMEs, all of them being associated with GAIA. The responses have been stored in an
UMA database. Of these 86 SMEs, 4 companies were nominated for an in-depth interview after the
completion of the e-survey. These interviews were conducted via telephone or face-to-face at the SME’s
own offices.
The interviews with the organizations that participated in the extended questionnaire were conducted
between April and May 2014 and lasted between 50 and 90 minutes approximately.
Taking into account the results of the e-survey, the traceability, is examined between information classified
as: existing data; semi-existing data; new data and the variables related to the same. These variables are
broken down as follows:
 Untreated: gathered directly from the data source
 Treated: gathered directly from the data source but will be treated by our system
 Generated by Data Mining Service: created by the Data Mining system itself
Table 15. Traceability matrix (classified information and related variables)
4.3.2
Results of e-Survey by GAIA and CIC
Company Profile
Figure 76 captioned, “Commodity products, very often fast selling” shows that this is the main type of
product offered by e-shops.

PUBLIC
Page 86 of 255
E-COMPASS
Figure 76. Question 1) in online questionnaire. Spanish version for GAIA
Continuing with the company profile, question 2 asks about the most relevant commercial sectors in which
companies work. As illustrated in Figure 77, a large number of GAIA e-shops belong to the Business to
Consumer (B2C) sector (53%), but with an important presence in Retail commerce, which is explained by
the fact that they are SMEs.
2) Please tick the most relevant of the e-commerce sectors that your company belongs to:
In this sense, as shown in Figure 78 (related to question 3), a high percentage of companies (41%) have less
than 5 employees, 37% of them have between 5 and 10. As is usual in SMEs, there are no companies with
more than 50 employees.

PUBLIC
Page 87 of 255
E-COMPASS
A high percentage of companies (41%) have less than 5 employees, 37% of them have between 5 and 10.
There are no companies with more than 50 employees. Most of the companies in the association (GAIA)
are SMEs which are the main beneficiaries of this project.
4) Which is the total annual revenue from online sales?
Twenty-six per cent of e-shops have annual revenue of lower than 5K euro, which indicates that they are
still small companies in initial phases of operation. Nevertheless, 24% of companies showed annual
revenues of between 11K and 50K euro, most of them corresponding to consolidated e-shops.

5) Which was the annual volume of orders received in 2013? (including the orders that were not
Thirty-seven per cent of companies showed an annual volume of orders, received in 2013, of between 101
and 1,000 euro and the rest of the rank values are less than 21%. The majority of the e-shops in GAIA have
been active between 1 to 5 years. Although there are a few (35%) that started their activities in 2013 (see
graphic of Figure 81).

PUBLIC
Page 88 of 255
E-COMPASS
6) How long has your company been doing business online?

A high percentage of GAIA companies compare their prices weekly (34%) or even bi-weekly (22%). In
almost all cases they calculate the comparison using a web competitor.

PUBLIC
Page 89 of 255
E-COMPASS
A low percentage of them compare prices in real-time (9%) or every few minutes (1%), although they
declared to be, in general, quite interested in having the possibility to compare and update prices
automatically and in real-time (Figure 84 & Figure 85).
To the question of how they adjust the prices in their e-shops, as can be observed in Figure 84, a high
percentage (36%) of companies replied that they do it manually. However, some of them (47%) added that
they currently use automatic tools for adjusting prices in online environments.
automatically?
Some requirements are: Provide information about direct competitors, not generals; that fits my needs, my
customers’ profile.

As commented (and Figure 85 illustrates), a high percentage (81%) of companies declared themselves to be
interested in using an automatic service for price notification and comparison. Sixteen per cent of e-shop
owners replied that they do not need a new tool for price comparison because, as shown in the previous
question, they are already using an automatic tool for online price comparison. A few companies (3%)

PUBLIC
Page 90 of 255
E-COMPASS
reported some requirements that automatic tools have to accomplish, such as: providing information about
direct competitors, and having the feature of being easily customized for a particular e-shop.
In terms of the visitors’ behavior analysis, questions 11), 12) and 13) are dedicated to discovering how GAIA
e-merchants are currently using services to detect or predict their visitors’ behavior, and to what degree
they are interested in using a new automatic tool for this purpose.
Figure 86 shows a bar graph of replies with regards to question 11). A first interesting observation in this
figure is that 33% of companies do not use any kind of tool for visitor tracking or behaviour analysis.
However, there are also a number of companies (40%) that use automatic online tools. Specifically, as also
revealed from the personal interviews, among these automatic tools, the favourite one is Google Analytics
and its use is merely informative.
11) Are you currently using any service or tool for customer behaviour analysis, churn prediction and/or
prevention?

12) Are you interested in a service which analyses the customer behaviour, provides feedback on how
to improve your e-shop and supports/optimizes your cross-selling activities?

PUBLIC
Page 91 of 255
E-COMPASS
13) Are you interested in a service which analyses the behaviour of groups of customers, discovering
Requirements: tight to the geographic area; it is free; unpersuasive service; provided they do not involve
direct customer involvement.

Questions 12) and 13) are closely related in such a way that they ask for the interest of the e-shop’s owners
in new services for the analysis of the customer behavior (or groups of customers) to improve cross-selling
activities and to discover tendencies in e-sales. In this regard, Figure 87 and Figure 88 show the graphs
generated for these two questions from GAIA e-merchants’ replies, where it is easily observable that in
both cases, 80% of companies declared themselves to be interested in these analytical services. A low
percentage of replies (5%) gave a series of requirements that services might cover: be freely available and
they do not entail direct customer involvement.
As shown in Figure 89, the possibility of fraud prevents them (28%) from using online payment transactions
of e-card systems, so they mainly prefer to offer additional payment options like direct anticipated bank
transactions, and to analyse them manually.
14) Is fraud a concern which prevents you from:
Figure 89.Question 14) in online questionnaire. Spanish version for GAIA

PUBLIC
Page 92 of 255
E-COMPASS
In this regard, the proportion of fraudulent cases in their total volume of transactions was lower than 0.1%,
for the majority of companies (53%) as shown in Figure 90. 14% of them registered fraudulent movements
in more than 1.1% of their transactions.
As mentioned, a high percentage (26%) of e-shop’s owners check their incoming transactions manually,
followed by an automatic online fraud evaluation tool for online transactions (25%) as shown in Figure 91.

PUBLIC
Page 93 of 255
E-COMPASS

into account?
Most of the SMEs in GAIA give priority to device identity (17%), bank country (15%) and country device
(13%) according to Figure 92.


PUBLIC
Page 94 of 255
E-COMPASS
As expected, almost all GAIA e-shops are providing their online products in services in the Spanish language
(57%). There are a number of them (24%) that offer the option to change the language of their e-shop to
English.
4.3.3
Results of Focus Groups and Interviews by GAIA and CIC
Now, information gathered from focus groups and interviews is also provided in the scope of GAIA chamber
companies.
Web Analytics Application
Table 16. Responses from GAIA concerning current state in web analytics: visitor behaviour
Questions
(frequency)
Responses












Most common search phrases used
Geolocalization (country, city…)
Average visit time
Number of visits
Client device
Incoming site
Conversion rate
Cost by click
Web site that referenced
Google Analytics (3 companies)
Mastertool (1 company)
Every 3 week about 250 days (4 companies). In season, it is more
sporadic

None
web analysing?

3-4 hours


Billing/Register confidence information from clients
No information from visitors (no clients), besides this information
gathered/processed by Google Analytics
Table 17. Responses from GAIA concerning current state in web analytics: competitor’s analysis
Questions
Responses
you analysing?



Top selling products
Prices
Products of outlets

Search in Google

In season: Summer, Autumn, … and about outlets

Mainly the price is determined by the supplier
competitors‘ information? (frequency)

PUBLIC
Page 95 of 255
E-COMPASS

Middle day

Detection of the variation in the price of outlets
Table 18. Reponses from GAIA concerning new requirements for web analytics: visitor behaviour
Questions
Responses
to analyse?







The types of consumer behaviours
Average value of a shopping cart of the e-shop
Number of visitor who put a product into the basket
Number of visitor who break up the checkout (purchasing) process
Average time of stay in e-shop until purchasing products
Number of visitor who did a purchase
Most common page view sequence.

Daily

Notices after detecting an anomalous behaviour
Table 19. Responses from GAIA concerning new requirements for web analytics: competitor’s analysis
Questions
What competitors‘ information do
you wish to analyse?
the competitors‘ information?
derive?
Responses

Competitors outlets’ price

In seasons: Summer, sales…

Notification in the change of the price of products by competitors
Table 20. Pilot users in GAIA
Questions
analytics)
analytics tool?
examples)

Responses

Yes (4 company)


Yes (1 company)
No (3 companies)


Yes (2 company)
No (2 companies)
PUBLIC
Page 96 of 255
E-COMPASS
 Current State in Fraud Detection
Table 21. Responses from GAIA concerning the current state in fraud detection
Questions
Responses
What kind(s) of payment fraud is your
company mainly exposed to?
What is the typical proportion of
fraudulent cases in your total volume
of transactions?
How do you deal with online payment
fraud?
What kind of intelligence/analytical
tools does your fraud monitoring
system imbed?
How do you maintain/update your
anti-fraud system?

Mainly, problems at the bank transfer

Approx. 1%

They deal with the third-party: Bank,..

None

None
What transaction parameters do your
fraud specialists particularly look at in
order to assess the legitimacy of an
order?










Suspicious or dubious behaviour
Device country
Device identity
Bank country
3D secured transaction
Bank name provided by the user
E-mail address
Address city
Address country
Telephone country
What type of transaction-specific
information does your antifraud tool
take into account?


Name the top-five key transaction
attributes that can quickly help detect
fraud





E-mail address
Bank country
Address city
 Cost Efficiency Concerns
Table 22. Reponses from GAIA concerning the current state in fraud detection
Questions
What kind(s) of payment fraud is your
company mainly exposed to? (that can
differ depending on
The 2013 CNP fraud revenue loss was
less or more compared to the previous
two years?
How much money did you spend in
2013 for fraud management?

Responses

Mainly, problems at the bank transfer

Equal, 1% (4 companies)

0 (4 companies)
PUBLIC
Page 97 of 255
E-COMPASS
How was the budget distributed
among antifraud systems and / or
personnel dealing with fraudulent and
suspicious cases?
How many employees are involved in
fraud management?


0 (4 companies)
0 (4 companies). Mainly, it is the same person that it is responsible of
the e-shop
 Efficiency Of The Overall Fraud Management Process
Table 23. Reponses from GAIA concerning the efficiency of the overall fraud management process
Questions
Responses
What is the overall proportion of
transactions screened automatically?
How many minutes does it take on
average an employee to investigate a
suspicious order?
How much time does it take to
manage a fraud complaint?
What is the proportion of transactions
sent to fraud staff for review?
What is the rate of inbound orders
that are automatically processed
without manual screening?
What is the rate of inbound orders
that are automatically rejected
without manual screening?
What is the typical rate at which
reviewers reject orders that have
initially been marked as suspicious?
Do you see any significant difference
in rejection rates between domestic
and international orders?

0 (4 companies)

3 hours – 5 hours

Approx. one day

All (4 companies)

0 (4 companies)

0 (4 companies)

0 (4 companies)

Yes, the payment of the transport, customs duty…
 New Requirements For Anti-Fraud (About Future)
Table 24. Responses from GAIA concerning new requirements for Anti-fraud
Questions
Responses
Would you hire more fraud analysts?

No (4 companies)

No (4 companies)

No (4 companies)

No (4 companies). The fraud is not a priority.
Would you spend more money on
training your existing personnel on
new types of fraud?
Would you consider gradually
reducing the fraud staff budget and
switching to automatic security
assessments tools?
Would you prefer to develop an inhouse fraud-detection tool rather
than using a web-based software-as-a-

PUBLIC
Page 98 of 255
E-COMPASS
service? (here it is important to know
the exact reasons behind each choice )
Would you be interested in investing
in a customised security assessments
tool from the market?
Is fraud a concern which prevents you
from selling your products/services
online?
Is fraud a concern which prevents you
from expanding your online market
access to the entire EU?
What is the most important direction
in terms of R&D initiatives? Rank the
choices:
Ranking (most important in top)
B
C
A
F
E
D

No (4 companies)

No (4 companies)

Yes (2 companies)
A. Modernizing the fraud management process
(e.g. by introducing advanced data mining tools,
new types of fraud analytics)
B. New business practices for optimising the
performance of review teams (review protocols,
tactics to reduce false acceptance or rejection
rates, behavioural concerns, quality assurance
indicators)
C. Improving user interfaces for background
checks (graphical tools, consolidation of
information from external sources, etc)
D.
Develop
new
online
transaction
monitoring/analytics tools (device /velocity
monitoring, IP analysis)
E. Improving the cost-benefit relationship of
cooperative schemes (involving humans and
machines)
F. Utilising information from heterogeneous
sources (social networks, geo-analytical
services, etc.)
B
D
C
E
A
F
(4 companies)
Conclusions of the online questionnaire within GAIA’s SMEs
The analysis of the survey data shows that the majority of the e-shop owners questioned are small
enterprises with less than 5 employees in B2C sector, selling commodities or quick sale or very comparable
across number of products, being 1-2 years in online business, having a maximum number of online orders
between 101-1,000 a year and an annual revenue of less than 50,000 €. Basically, the main languages
supported by the e-shops are Spanish and English.
The most of the e-shop owners questioned carry out competitors’ observation and price adjustment
automatically & online and manually. The main websites for price comparison are firstly, only the websites
of the competitors and secondly the price/product search engines, e marketplaces & web sites of the
competitors.
The current frequency of price comparison is mostly weekly. Moreover, 81% of the e-shop owners
questioned are interested in a service for price comparison, additional 2% are conditionally interested.
Most of questioned e-shop’s owners are interested in a service which analyses the customer behaviour,
those representing an 81%. Being a 3% of the owners that are conditionally interested. In this case, the
requirement is the simplicity of the system. Although, most of them are currently using a service or tool for
customer behaviour analysis in an automatically and online way more or less, in a 40%.

PUBLIC
Page 99 of 255
E-COMPASS
Besides, most of questioned e-shop proprietors are interested in a service which analyses the behaviour of
groups of customers, discovering habits and detecting new e-shopping tendencies, in an 80%. Being a 5% of
the owners that are conditionally interested.
Respecting to the antifraud application, the majority of the e-shops owners answered that fraud is a
concern which prevents them from selling their products/services online and expanding their online market
access to the entire EU and using online payment transactions e-card systems, in a 28%. However, the
proportion of fraudulent cases in their total volume of transactions is less than 0,1% in the 53% of the eshop owners questioned and 32% deal with online payment fraud manually, review of incoming
transactions, or transferring risk to a third party.
The variables which have been taken into account for this proposal are mainly: device identity, country
bank and suspicious country device.
Analysis of the interviews with GAIA’s SMEs
Analysing the companies that have done the interview, it appears that all of them offer minimally
processed, high-demand products. Furthermore, 50% of the companies surveyed do not know exactly the
main reasons why the customers make purchases at its e-Shops. It also highlights the fact that only these
SMEs employ a person to ensure online sales and do little investments in their online presence (<10K €). In
short, these are companies with a small volume of business and its revenues from online sales do not make
a big impact on their business.
The e-shop remarks that it has been less than 5 years selling their products via the Internet and, therefore,
it is a process that must evolve towards maturity. The e-shops' owners consider the on-line service as an
added service to its business performance but not a priority.
In addition, it appears that the e-shop respondents do not use electronic tools to observe and compare the
prices of its competitors because it is the provider which set them. And the market to which it aims is the
Spanish-speaking because their web sites only support a single language: Spanish. The incursion of new
languages could facilitate the opening up new markets provided control logistics costs.
From the opposite point of view, it is noted that most of the surveyed companies do not address the fraud
in online payments, nor they has any specific software that allows dealing with fraud in the online payment.
This is justified because the fraudulent cases in the total volume of transactions do not exceed 1% and they
leave it in the hands of third parties: banks and gateways for payment.
Finally, it can be seen that the e-shops have a certain level of infrastructure because they have mostly
outsourced the hosting of their website and almost all use Google Analytics as a web analytics tool. Of the
e-shops that use Google Analytics, actually very few take advantage of the tool. Remarkably, all interviewed
companies would be willing to share data anonymously with project partners. However, not all of them are
willing to share the schema of your database.

PUBLIC
Page 100 of 255
E-COMPASS
4.4
4.4.1
Results and Analyses from Spanish Chamber (ATEVAL)
Organisation of e-Survey and Focus Groups by ATEVAL and UMA
In the case of ATEVAL (Valencia), a series of activities have been conducted that comprises the local
methodology for the requirements specification as follows: first, the online questionnaire (in Spanish) was
distributed to all the SMEs of this chamber from which, a number of 36 of them filled all the questions with
numeric responses and additional comments. Second, several focus group/info day sessions were
organized in the office place of ATEVAL, where 14 companies assisted and were informed about the
project. In these informative sessions, several interviews were also made with two selected companies
(Nessys and Gestiweb) in order to obtain more in depth information. Figure 94 shows pictures of focus
groups sessions in ATEVAL. Resulting statistical information from data collection by UMA is reported next.
Figure 94. Focus group informative sessions and interviews in ATEVAL
Interviews with ATEVAL companies were organized after focus groups sessions on 20th/21st of May in
Ontinyent (Valencia). A later and extensive interview with a software services provider (e-shops designer)
was also performed, in which we planned the way to proceed for requesting real data to the e-shop’s
owners.
It is worth noting that one of the interviewed companies in ATEVAL is more focused on software
development, and mostly on e-shop design and management. This company is willing to provide (if
possible) the SME-E-COMPASS project with real data from their own customers. In this regard, UMA
prepared an additional info flyer of the project to inform all these new e-shops owners by email. An image
of this flyer can be observed in Figure 95.

PUBLIC
Page 101 of 255
E-COMPASS
Figure 95. Info Flyer for real data collection from e-shop’s owners (Google Analytics and Piwik reports, etc.)
The statistical information is organized in replies collected from questionnaires and responses gathered
from focus groups and interviews. The information is then processed by following the same structure of
questionnaires, that is, company profile questions, web analytics (price optimization and visitor behaviour)
and fraud detection applications.
4.4.2
Results of e-Survey by ATEVAL and UMA
Company Profile
Figure 96 shows the percentage of options that ATEVAL companies selected with regards to question 1 in
the online questionnaire. In concrete, it is referred to the types of products and services that these
companies offer. In this regard, 46% of them offer commodities and very often fast selling products,
followed by personalised and exclusive products with 18% a 15% of responses, respectively. ATEVAL is
mostly oriented to textile factories and e-shops, which is in the line of these responses.

PUBLIC
Page 102 of 255
E-COMPASS
Figure 96. Question 1) in online questionnaire. Spanish version for ATEVAL
Following with the company profile, question 2 is concerning the most relevant commercial sectors on
which SMEs compete. As illustrated in Figure 97, a large part of ATEVAL’s e-shops belong to Business to
Consumer (B2C) sector and retail (64%).
2) Please tick the most relevant of the e-commerce sectors that your company belongs to:
In this sense, as shown in Figure 98 (related to question 3), a high percentage of companies (75%) have less
than 5 employees in staff, and the 15% of them have between 5 and 10. As usual in small enterprises, there
are no companies with more than 50 employees.

PUBLIC
Page 103 of 255
E-COMPASS
Questions 4 and 5 are concerning to the annual revenue from e-sales and the volume of orders received in
2013, respectively. Figure 100 shows the graphics of these two related questions, from which, there is a
direct relation between those companies that obtained a high revenue in 2013 (6% obtained more than
100K euro) and those ones that received the largest number of orders (6% of e-shops with more than
50,000 orders in 2013). 35% of e-shops have annual revenue lower than 5K euro, which indicates that they
are still small companies in initial phases of operation. Nevertheless, 24% of companies showed annual
revenues between 11K and 50K euro, most of them corresponding to consolidated e-shops.
4) Which is the total annual revenue from online sales?
5) Which was the annual volume of orders received in 2013? (including the orders that were not
Figure 99. Question 4) in online questionnaire. Spanish
version for ATEVAL
Figure 100. Question 5) in online questionnaire.
Spanish version for ATEVAL
The great amount of e-shops in ATEVAL are working from 1 to 5 years, although there exist several of them
(6%) that started their activities in 2013 (see graphic of Figure 101).

PUBLIC
Page 104 of 255
E-COMPASS
6) How long has your company been doing
business online?
Figure 101. Question 6) in online questionnaire.
Spanish version for ATEVAL
Concerning Web Analytics applications, question 7 is focused on discovering the sources of information and
sites that e-shops’ owners usually visit in order to compare prices. Figure 102 shows the percentage of
companies’ replies with regards to the kind of website with which they compare their own prices. The
highest percentage of replies (42%) corresponds to e-shops owners that frequently visit the web sites of
their direct competitors. In fact, as reflected in next section devoted to focus groups and personal
interviews, they demand automatic systems able to gather usual competitor keeping a list of prices per
product and direct competitor.
A high percentage of ATEVAL companies compare their prices weekly (26%) or even monthly (26%). A low
percentage of them compare prices in real-time (9%) or every few minutes (3%), although they declared to
be in general interested to have the possibility of comparing and updating prices automatically and in realtime. Figure 103 reports the summary of responses concerning question 8) as follows:

PUBLIC
Page 105 of 255
E-COMPASS
To the question of how they adjust the prices in their e-shops, as can be observed in Figure 104, a high
percentage (59%) of companies answered that they do it manually. However, some of them (32%) added
that they currently use automatic tools for adjusting prices in online environments.
automatically?


PUBLIC
Page 106 of 255
E-COMPASS
As commented before (and Figure 105 illustrates), a high percentage (68%) of companies declared to be
interested in using an automatic service for price notification and comparison. 29% of e-shop’s owners
responded that they do not need any new tool for price comparison since, as shown in the previous
question, they correspond to those ones that are already using an automatic tool for online price
comparison. A few of companies (3%) reported some requirements that automatic tools have to
accomplish, such as: providing information about direct competitors, and having the ability of being easily
customized for a particular e-shop.
In terms of visitor behaviour analysis, questions 11), 12) and 13) are devoted to discover how ATEVAL emerchants are currently using any kind of service to detect or predict the visitor behaviour, and in which
degree they are interested to use a new automatic tool with this purpose.
Figure 106 shows a bar graph of replies with regards to question 11). A first interesting observation in this
figure is that more than 50% of companies do not use any kind of tool visitor tracking or behaviour analysis.
However, there are also a number of companies (36%) that use automatic online tools. In concrete, as also
extracted from personal interviews, these automatic tools are: Google Analytics, Prestashop ATs, and Piwik,
which use is merely informative.
11) Are you currently using any service or tool for customer behaviour analysis, churn prediction and/or
prevention?


PUBLIC
Page 107 of 255
E-COMPASS
12) Are you interested in a service which analyses the customer behaviour, provides feedback on how
to improve your e-shop and supports/optimizes your cross-selling activities?
13) Are you interested in a service which analyses the behaviour of groups of customers, discovering
Figure 107. Question 12) in online
questionnaire. Spanish version for ATEVAL
Figure 108. Question 13) in online questionnaire. Spanish version
for ATEVAL
Questions 12) and 13) are strongly related in such a way that they ask for the interest of e-shop’s owners
on new services for the analysis of the customer behaviour (or groups of customer) to improve cross-selling
activities and to discover tendencies in e-sales. In this regard, Figure 107 and Figure 108 show the graphs
generated for these two questions from ATEVAL e-merchants’ replies, where it is easily observable that in
both cases, 85% of companies declared to be interested on these analytical services. A low percentage of
replies (6%) exposed a series of requirements that services might cover: tight to the geographic area, be
freely available, unpersuasive service, and they do not involve direct customer involvement.
Anti-fraud services are also a focus of interest for e-merchants in ATEVAL chamber. As shown in Figure 109,
fraud prevents them (36%) to use online payment transactions of e-card systems, so they mainly prefer to
offer additional payment options like direct anticipated bank transactions and cash on delivery, to lately
analyse them manually. This is possible in current companies since, as shown in previous analyses, the
volume of order and revenue per year is still moderated for almost all companies in ATEVAL. However, this
practice would become infeasible as long as the number of transactions grow, which will lead them to use
automatic security methods for their operations.
14) Is fraud a concern which prevents you from:

PUBLIC
Page 108 of 255
E-COMPASS
In this regard, the proportion of fraudulent cases (to the best of their knowledge) in their total volume of
transactions is lower than 0.1%, for the majority of companies (74%), as shown in Figure 110. Only 3% of
them, registered fraudulent movements in more than 5% of transactions, which is in the line of the
proportion of companies with the highest volume of revenue per year (see Figure 99 in company profile).
As aforementioned, a high percentage (32%) of e-shop owners check their incoming transactions manually,
followed by those companies that transfer this risk to a third party, like PayPal or other kind of automatic
real-time tool for online transactions. Figure 111 reflects this behaviour in the scope of ATEVAL e-shops.
Most of the SMEs in ATEVAL give priority to the customer’s suspicious of dubious behaviour, device country
and identity from which started the operation (see Figure 112). Then following the bank country and
address of the transaction. Other indicators such as the IP address or Geo-location seems to be of
importance for companies as they specified some times in option “Other”.

PUBLIC
Page 109 of 255
E-COMPASS
into account?
As expected, almost all ATEVAL e-shops are providing its online products in services in the Spanish language
(94%). Also a number of them (50%) offer the option to shift the language of e-shops to English. It is worth
mentioning that some e-shops use the Valencian language, since they are located in this specific region and
they offer services to their local customers (3%). Figure 113 shows the percentage of languages used in
ATEVAL e-shops.

PUBLIC
Page 110 of 255
E-COMPASS
4.4.3
Results of Focus Groups and Interviews by ATEVAL and UMA
The following tables (Table 25 to Table 30) summarises the responses and insights collected from the focus
group discussion and personal interviews. These tables presents the responses regarding the web-analytics
application with the use of data mining techniques and follows a paragraph regarding the practices and
concepts of the ATEVAL SMEs regarding fraud management.
Table 25. Responses from ATEVAL concerning the current state in web analytics: visitor behaviour
Questions
(frequency)
Responses



















web analysing?





Client device
Number of visits
Average visit time
Geo localization
Incoming site
Google analytics (7 companies)
Piwik (1 company)
Prestashop analytics tools (3 companies)
Zircus BI (1 company)
Pentahop (1 company)
Mastertool (1 company)
Propietary tools (2 companies)
Real-time (6 companies)
Weekly (2 companies)
Monthly (2 companies)
Sporadic
Direct indication of interesting products for clients (manually)
Minipages of a given zone of the e-shop with keywords regarding
products in this region to later control the number of visits looking for
those keywords
Linked blogs with related content/hidden text (SEOs should bring out
to the rank if meets these kind of activities)
Improving product’s descriptions and prices
Half an hour per day by average (7 companies)
3 hours by average in weekends
Billing/register confidence information from clients
No information from visitors (no clients), besides this information
gathered/processed by Google Analytics
Table 26. Responses from ATEVAL concerning the current state in web analytics: competitor’s analysis
Questions
Responses
you analysing?



Top selling products
Vip client’s accounts on competitors e-shops to discover offers, special
products, and new trends

Prices
PUBLIC
Page 111 of 255
E-COMPASS
competitors‘ information? (frequency)












Last selling products and destination
Most popular products
Amount of selling products
Reputation: votes, comments, starts
Number of visits
Manually, Excel with summary and ranking of competitors
Benchmarking web sites with price comparatives
Daily
Every weekend
Monthly
Price updating
Improve product’s descriptions

In all cases, personal dedication of e-shops owners




Companies of the same commercial sector
Manufacturers having their own e-shops
Direct competitors with similar products
Multinationals with the better rankings in SEO systems
Table 27. Responses from ATEVAL concerning new requirements for web analytics: visitor behaviour
Questions
Responses
What additional metrics do you wish to
analyses?



Mouse location tracking in the e-shop
Tracking of clicks in the e-shop
Changes in the module distribution of the e-shop in order to discover
where the visitors pay attention, mostly when they enter the e-shop for
the first time

Locating hot regions in the e-shop site to locate special offers

Daily




Stock updating in real-time
Modifying the e-shop design to place popular links more visible
Locating other interesting products close to most popular ones
Calculating the conversion rate
Table 28. Responses from ATEVAL concerning new requirements for web analytics: competitor’s analysis
Questions
Responses
What competitors‘ information do you
wish to analyse?








competitors‘ information?

Number of actual visits and sales
Actual conversion-rate
What competitor they visit just when leaving my e-shop
Daily
Weekly. On weekends
Modify the price
Improve product’s descriptions and images
Increasing the catalogue of products including popular products in
competitors e-shops
PUBLIC
Page 112 of 255
E-COMPASS

Show the number of sales for top ten products
 Requirements for Automatic Actions
Table 29. Responses from ATEVAL concerning requirements from automatic actions
Questions
Responses
What actions or rules would initiate
which actions?


How invasive may be the actions for
the e-shop?



Detecting products without offers in competitors in order to promote
exclusive offers in our own e-shops
Exploring trends, shopping seasons for triggering warnings messages
and reports
Automatic search of new shopping tendencies
Automatic negotiator for improving prices and offers
Automatic, although asking the e-shop owner for confirmation of the
action proceed
Table 30. ATEVAL pilot users
Questions
Responses
analytics)


analytics tool?
examples)
Yes (4 companies)
Software e-shop designer is willing to ask their own clients in order to
get authorized to provide the project with real data from metrics

Most of accessible information comes from Google Analytics

Yes, if software providers make it in a transparent way (4 companies)

N/A (the remaining of companies)


Yes, if software providers make it in a transparent way (4 companies)
N/A (the remaining of companies)
This analysis is also structured in two main parts: statistics gathered from questionnaire responses, and
interpretation of comments in focus groups and interviews.
Analysis of responses from online questionnaire
Concerning questionnaire responses, a first analysis is regarding the company’s profiles, in order to obtain a
general view of the commercial sectors in which most of them are involved, as well as their dimensionality.
According to graphs in questions 1) and 2), most of companies are devoted to commodity products and
belong to Business to Consumer (B2C) sector. In a lower number, there also exist SMEs dealing with
personalised and exclusive products, which are often involved in sectors of services and retail.
In terms of dimensionality, a high percentage of companies (76%) have less than 5 employees in staff, and
the 15% of them have between 5 and 10. There are no companies with more than 50 employees. 35% of eshops have an annual revenue lower than 5K euro (see graph in question 4), which indicates that they are
still small companies in initial phases of operation. Nevertheless, the 24% of companies showed annual

PUBLIC
Page 113 of 255
E-COMPASS
revenues between 11K and 50K euro, then corresponding to consolidated e-shops. The annual volume of
orders received in 2013 is around 100, since 32% of them are below 100 and between 101 and 1,000.
Nevertheless, 24% of companies received more than 1,000 orders in the last year. These previous
percentages correlate with the time the companies are doing online business. That is, 32% of companies
operate from 1 or 2 years, and 32% of them are working from 5 and 10 years. In general, it is clear that
most of companies in this chamber are SMEs, which are active from the last few years in e-commerce
environment and they are still in an initial phase of operation.
A last question in this regard is concerning the language/s currently used in ATEVAL e-shops. Obviously, the
higher percent of e-shops sites are available in Spanish (94%), although a number of them provide options
in English (50%) too. It is worth mentioning that some e-shops use the Valencian language, since they are
located in this specific region and they offer services to their local customers (1%).
The second block in the questionnaire consists on Web Analytics applications. In this regard, questions are
first refereed to price comparison systems, for which the main websites examined by e-shop’s owners are
those of direct competitors, followed by price products search engines and eMarketplaces. E-shop’s owners
adjust their product’s prices weekly (26%) and monthly (26%), and only a 9% revise their prices in real-time.
In fact, the way of adjusting prices is mostly manually for more than 50% of companies, although there are
also a number of them (32%) that responded automatically online. In this sense, as graph of question 10)
reflects, a high percentage of them (68%) are interested in using a new application for automatic price
adjustment.
The remaining questions concerning Webs Analytics are focused on visitor’s behaviour applications. In this
regard, question 11) is about the use of tools for churn prediction and/or prevention, for which 50% of
companies do not use any tool and around 35% of them declared that they use automatic online tools. Of
course, most of them (85%) are quite interested on using a service for customer behaviour analysis, but
they often added that this service should be free available and easy to use. A similar response (85%) is
received when these companies are asked about a service to discover tendencies and common habits in
groups of clients.
The third block corresponds to anti-fraud questions. Question 14) is concerning actions that fraud prevents
e-shop’s owners to proceed in a given direction. In most of cases, responses are focused on online
transactions with card systems, for purchasing (29%) as well as for selling (32%) actions. Nevertheless,
when they are asked for the typical proportion of fraudulent cases in their transactions, a high percent
(74%) of companies confirmed to be lower than 0.1%. In fact, most of transactions are revised manually
(32%) by e-shops owners, and the 15% of them declared that they do not deal with online fraud payment at
all. In the case of companies that are currently using an anti-fraud tool, they argued that they can obtain
information from suspicious or dubious behaviour, and also secondary information such as the device
country, IP, etc.
Analysis of focus groups and interview comments
In the case of focus groups and interviews, comments obtained from e-shops are more regarded to web
analytics tools, since all the participant e-shops stated that they transfer the task of anti-fraud analysis to a
third party (mostly PayPal). In fact, only one of the interviewed e-merchants could be interested in using a

PUBLIC
Page 114 of 255
E-COMPASS
new application (or payment gateway) for anti-fraud, although it must guarantee the security and be easy
to use.
In general, we can stand out Google Analytics as the most used tool for almost all interviewed companies,
although they also use additional tools suggested by their software providers such as: Piwik, Prestashop
analytics tool, and Pentahop. Therefore, the set of common metrics e-shop owners in ATEVAL usually
analyse are those computed by Google Analytics, e. g., number of visits, average visit time, geo-localization,
country, and client device. These metrics are usually checked in real-time, weekly, and monthly. From these
simple analysis, e-shop’s owners carry out a series of interesting actions that could led SME-E-COMPASS
project to develop specific services/applications in this regard: direct indication of interesting products for
clients, linked blogs with related content/hidden text, improving product’s descriptions and prices, mouse
location tracking in the e-shop, changes in the module distribution of the e-shop in order to discover where
the visitors pay attention (mostly when they enter the e-shop for the first time), and Locating hot regions in
the e-shop site to locate special offers.
Concerning competitor’s analysis, e-shop’s owners usually check the list of top selling products of their
competitors, last selling products and destinations, and current prices. In concrete, companies pay special
attention on reputation comments and ranking votes (commonly known as stars ranks). Additional actions
that e-shop’s owners usually perform are related to Improving product’s descriptions and images,
increasing the catalogue of products including popular products in competitor’s e-shops, and showing the
number of sales for top ten products. In general, these attributes from competitors are collected manually
in Excel files, and this process is often done on weekends.
In all cases, it involves the personal dedication of e-shop’s owners. In this regard, they demand automatic
solutions such as: stock updating in real-time, modifying the e-shop design to place popular links more
visible, locating other interesting products close to most popular ones, and especially automatic conversion
rate calculation. Moreover, additional requirements that interviewed companies declared to wish are:
detecting products without offers in competitors in order to promote exclusive offers in our own e-shops,
exploring trends or tendencies in shopping seasons for triggering warnings messages and reports, and
automatic negotiator for improving prices and offers. In addition, a common requirement for all companies
is that new generated applications should be integrated with their systems in a transparent way.
Finally, from these focus groups, only two companies acceded to be pilot users and to provide us with their
analytics data (mostly Google Docs reports). Moreover, one of these companies is a Software e-shop
designer that was willing to ask their own clients in order to get authorized to provide the project with real
data from metrics.
Conclusions of the analysis results
A series of conclusions can be extracted from analysis of questionnaires and interviews as follows:

The tool should be integrated with existing e-shop’s systems in a simple and transparent way.
The tool should provide the companies with useful information to improve the e-shop sites, like
optimum module location, hot visible regions, and products rankings.
 The tool should be able to access and gather the price and product data of the competitors’
websites given by the e-shop owners.
 The tool should offer a robust, secure, and simple enough anti-fraud service in order for e-shop’s
owners to change their current payment gateways.


PUBLIC
Page 115 of 255
E-COMPASS

4.5
The tool should be able to generate information about new tendencies in e-markets.
Common Results and Analyses
In this section, common statistics and analyses from all questionnaires and interviews are described in
order to obtain general requirements and conclusions. These analyses are then performed with previous
information gathered from all SMEs e-commerce chambers: EPK, HALTON, GAIA, and ATEVAL.
Company Profile
In general, the majority of studied companies are micro enterprises with 1 to 5 employees and working in
B2B or B2C sector. They are mostly oriented to commodity or exclusive products, carrying out their
activities from less than 4 years in online business.
The products offered by the e-shops questioned are numerous and diverse: women clothes, solid fuel,
packaging supplies, marine safety equipment, health products and pharmaceuticals, pet products, parcel
delivery options, books, group memberships, babies-children wear and bespoke nursery items, audio visual
equipment, domestic appliance consumables, bathroom products, print and design, sport equipment,
specialist gloves, fashion, shoes, sport and outdoor equipment, cosmetics, household, appliances,
electronics, exclusive products with high quality often slow selling are also provided: jewellery, handmade
crafts and gifts, watches, special and traditional food and beverage, personalised products and services are
referring mainly to travel and accommodation services: hotels, hostels and online travel agencies (OTA).
In terms of commercial volume, most of SMEs have a maximum number of 5,000 orders per year (2013)
and an annual revenue of maximum 10,000 € from online sales (more than 30% of companies, as shown in
Figure 114). They invest less than 10,000 € in their web activities.
Annual revenue from online sales in 2013
60%
50%
40%
30%
20%
10%
0%
< 10K €
11Κ -50K € 51K -100K € 101Κ - 200Κ
€
> 200k €
Figure 114. Annual revenue from online sales in 2013 for all the studied SMEs
Spain (Basque Country): GAIA
Spain (Valencia): ATEVAL
A last question in this regard is concerning the language/s currently used in e-shops. Obviously, the higher
United
Kingdom:
HALTOM
Greece:
EPK(Spanish, Greek and
percent of e-shops sites are
available
in native languages
for each chamber
region
English), although a number of them also provide options to shift to English (50%). Other languages like

PUBLIC
Page 116 of 255
E-COMPASS
German, French, and Portuguese are also used in e-shops. In addition, some e-shops use the Valencian and
Basque languages, since they are located in their specific regions offering services to local customers.
Figure 115. Question 8), general results from all questionnaires


Data Mining, Web Analytics Application
Concerning web analytics applications, questions are first refereed to price comparison systems. The
majority of the online SMEs are directly comparing their prices with the competition by relevant search
engines: Google and Yahoo (70%), and by checking the competitor’s e-shops. They are also monitoring the
price index through dedicated eMarketplaces (EBay and Amazon), while only 5% receive directly the prices
from the industry. The current average number of products is between 1 and 20 products, having the
possibility to do the comparison automatically the number would be higher.
As plotted in Figure 115, e-shop’s owners adjust their product’s prices mainly monthly (31%) and weekly
(26%), and only a 9% revise their prices in real-time. In fact, the way of adjusting prices is mostly manually
for more than 60% of SMEs (TOTAL label in Figure 116), although there are also a number of them (28%)
that responded automatically online. More in depth, most of SMEs from EPK, ATEVAL and HALTON perform
a manual adjustment of prices (more than 50% of them), although in the case of GAIA, companies use as
main option automatic offline methods for price comparison. However, as reflects the graphs of question
10), a high percentage of them are interested in using a new application for automatic price adjustment.
This last result is repeated for SMEs in all studied regions.

PUBLIC
Page 117 of 255
E-COMPASS
Figure 116. Price adjustment method of all SMEs
Concerning applications of web analytics focused on visitor’s behaviour, question 11) is about the current
use of tools for churn prediction and/or prevention. As Figure 117 summarizes, for the four studied SMEAGs, 47% of companies do not use any tool for churn analysis. On the contrary, a percentage of 29%
declared that they use automatic online tools, and 22% make this task manually. For EPK, HALTON, and
ATEVAL, the highest percentage of replies correspond to negative option in questionnaire, followed by
automatic online tools. For GAIA, it seems that a great amount of SMEs are currently using online services
for churn analysis. Of course, most of them (>80%) are quite interested on using a service for customer
behaviour analysis, but they often added that this service should be without charge and easy to use. A
similar response (>80%) is received when these companies are asked about a service to discover tendencies
and common habits in groups of clients.
Derived actions are content optimising, target group analysis and marketing activities. Stored data in the eshop systems, which could be used for the data mining use case, are names, addresses, phone numbers of
the customers as well as the frequency of spend and visits.

PUBLIC
Page 118 of 255
E-COMPASS
Figure 117. Customer behaviour analysis for all SMEs
In general, it can be stood out that Google Analytics is the most used tool for almost all interviewed
companies, although they also use additional tools suggested by their software providers such as: Piwik,
Prestashop analytics tool, and Pentahop. Therefore, the set of common metrics e-merchants usually
analyse are those computed by Google Analytics, e. g., number of visits, average visit time, geo-localization,
country, and client device. These metrics are usually checked in real-time, weekly, and monthly. From these
simple analysis, e-shops owners carry out a series of “manual” actions that could led SME-E-COMPASS
project to develop specific services/applications in this regard: direct indication of interesting products for
clients, linked blogs with related content/hidden text, improving product’s descriptions and prices, mouse
location tracking in the e-shop, changes in the module distribution of the e-shop in order to discover where
the visitors pay attention (mostly when they enter the e-shop for the first time), and locating hot regions in
the e-shop site to locate special offers. Remarkably, a number of interviewed companies would be willing
to share data anonymously with project partners. However, not all of them, would be willing to share the
schema of your database.
Fraud Detection Application
The third block corresponds to anti-fraud questions. Question 14) is concerning actions that fraud prevents
e-shop’s owners to proceed in a given direction. In most of cases, responses are focused on online
transactions with card systems, for purchasing as well as for selling actions. Nevertheless, when they are
asked for the typical proportion of fraudulent cases in their transactions, a high percentage of SMEs
(between 53% in the case of GAIA and 93% for HALTON) confirmed to be lower than 0.1%. This may happen
either because they tend to use safer ways of payment (cash on delivery, PayPal), most of them are micro

PUBLIC
Page 119 of 255
E-COMPASS
enterprises and their volumes of transactions or size of the e-shop is very small or became online the last 12 years. As observed by the data analysed, the fraudulent cases are directly related with the years of
operations and volume of transactions.
In general, as shown in Figure 118 for all chambers (label TOTAL), most of e-shop’s owners responded that
they do not deal with online payment fraud (47%) or they transfer this task to a third party (47%), e. g.
PayPal. This fact can be interpretated as an indicator of ignorance of the fraud risks among mostly micro
enterprises. As global statistics reveal and similar conclusion for the project’s survey is deduced, when the
sales start to increase proportionally and the fraudulent transactions emerge, more evidently when the eshop starts cross-border e-commerce. A moderate percentage of e-merchants revise their transactions
manually (29%). A special result can be observed in the case of HALTON, which 66% of SMEs declared that
they do not deal with online payment fraud. Similarly, a highest percentage of e-shop’s owners in EPK
(45%) and ATEVAL (33%) associations answered that they revise manualy all their transactions. As
aforementioned, these SMEs are currently able to perform manual review of transactions, since they are sill
young companies with a low volume of orders per year. In the case of companies that are currently using
an anti-fraud tool, they argued that they can obtain information from suspicious or dubious behaviour, and
also secondary information such as the device country and IP.
Figure 118. How do you deal with online payment fraud?

PUBLIC
Page 120 of 255
E-COMPASS
This overall profiling highlights the absence of actions taken against fraud as well as the ignorance of the
fraud risks and worldwide expansion (mostly among micro and newly established online SMEs that do not
offer online payment options).
General Conclusions
In the light of these results, a series of general conclusions can be extracted that lead to requirements for
the web analytics, as well as for the anti-fraud tools:

The tool should be integrated with existing e-shop’s systems in a simple and transparent way.

The tool should offer a robust, secure, and simple enough anti-fraud service in order for e-shop’s
owners to change their current payment gateways.

The tool should be able to access the price and product data of the eMarketplaces mainly used for
price comparison.

The tool should be able to access competitors’ pricing and campaign data. Both, the price and
product comparison should be made automatically.

The tool should be able to compare the prices of as much products as possible.

The tool should be able to derive predefined actions (e.g. recommendations, reward cards,
vouchers) from visitors’, customers’ and competitors’ data by using predefined rules. The e-shop
owners should be able to choose from predefined rules and actions.

The tool should be able to identify products on the websites of the competitors, which are similar
or the same like that of the e-shop owners.

The tool should provide the companies with useful information to improve the e-shop sites, like
optimum module location, hot visible regions, and products rankings.

The tool should be able to generate information about new tendencies in e-markets.

The tool should perform e-card transaction’s checking automatically. Certain actions and/or
automatic warning concerning clients in cash transactions are also desirable.

New anti-fraud modules should be easily integrated with current payment gateways like PayPal.

PUBLIC
Page 121 of 255
E-COMPASS
5
5.1
User requirements implications to the project applications
Online fraud detection
One of the main activities of the SME-AG’s participating in WP2 was the organisation of focus groups and
specialised physical or remote interviews. The aim of this task was to contact a number of members in each
SME-AG and elicit useful information not only about the state-of-the-art in fraud detection systems but also
about future needs and challenges. As a guide for the coordination of focused discussions, interviewers
used as an extensive questionnaire which had been specially prepared for this purpose by EXO and ETR
under the supervision of SME-AGs. This included over 40 questions, organised in 6 sections, and performed
a more in-depth analysis into several aspects of currently in-place fraud detection systems that are not
covered by the shorter, general-purpose, questionnaire available on-line. Among those are the cost
breakdown of fraud detection operations, commonly used performance indicators, efficiency of the overall
fraud management process, etc. The main motivation behind the establishment of common interview
guidelines was to maintain uniformity and consistency in this process, which turned out to be of vital
importance for this stage of E-COMPASS taking into account the large diversity in the nationality and
technological proficiency of market survey participants. Below, we provide a summary list of user (technical
and operational) requirements posed by e-merchants and fraud analysts that participated in various focus
group discussions and interviews organised by SME-AGs:
1. Integration and assimilation of information sources. Many interviewees agreed that an important
requirement for future anti-fraud technologies is their ability to integrate, support and facilitate the
processing of heterogeneous data types and information sources. These typically range from purely
technical parameters collected during the normal day-to-day operation (e.g. browser or proxy
server settings) to geo-analytical attributes available from global databases.
2. Knowledge sharing. The majority of interviewees supported the view that increasing professional
awareness and knowledge sharing could help deal with fraud more effectively in the future. This
highlights the importance of developing reputation databases exposing known cases of malicious
IPs or credit cards to a wider community of fraud analysts.
3. Timeliness of response. A key performance measure indicated by market specialists is the
timeliness of system response, i.e. its ability to perform order screening and risk assessment online
within reasonable time limits. These are typically a function of the overall customer’s tolerance to
transaction execution waiting time and often dictated by good sales practices.
4. Time and cost efficiency. A common ascertainment among chamber members was that an
automatic fraud detection tool can be considered “worth trying” to the extent that it manages to
restrict the need for human intervention. This places an upper limit on the actual percentage of
orders that are typically directed to reviewers for further inspection. In the context of the
envisaged anti-fraud architecture presented in D1.1, the human intervention rate is effectively
controlled by the spread of the two thresholds that determine the boundaries between the
acceptance, review and rejection decision region.
5. Reliability. Many respondents pointed to system availability as a crucial condition for unlimited and
without interruption 24/7 service.

PUBLIC
Page 122 of 255
E-COMPASS
6. Customisation: Our experience from e-shop focus groups revealed the vast diversity of markets/
products which are served/sold through the online sales channel. This significantly hinders the
development of a “global” and “generic” anti-fraud system applicable to all markets, customer
groups and products. On the contrary, it makes more sense to provide a basic structure and then
prompt end-users to customise the system to their own needs and the particularities of their
business environment. At a first level, this could be achieved by supplementing the fraud
assessment platform with modules allowing e.g. the creation of new risk scoring rules, the
modification of existing ones or the changing of the rule execution order and hierarchy.
7. Security. Much of the order screening process involves analysing information, such as customer
data, IP address and geo-location, which is generally regarded as personal and non-disclosable. This
makes of paramount importance the “armouring” of the system with the necessary security
controls, which prevent not only access to “sensitive” customer information or transaction
parameters but also the unauthorised configuration of system settings (e.g. activation/deactivation
of filtering rules).
8. Communication and reporting. Most of the interviewed security professionals consider fraud
assessment an interactive process, in which human experts collaborate with computer programmes
to increase not only the formers’ insight into each case but also the overall performance in terms of
early fraud detection. This stresses the importance of a well-designed user interface that will
facilitate the man-machine communication and allow the user to gain a better understanding of
cybercrime practices. Several functionalities brought out by the interviewed experts to this end are:
a) Easy report generation
b) Dashboards showing key statistics and performance indicators (e.g. acceptance vs
rejection rates)
c) Information on case-by-case user navigational patterns, set of activated rules, risk
score
breakdown,
justification
for
the
classification
result
(acceptance/rejection/review), etc.
9. Technical level adjustment. Many were the experts who pointed out that in everyday e-commerce
operations, order-reviewing and fraud-assessment employees typically refrain from going into the
technical details of classification algorithms and data mining tools underlying the risk scoring
process. Even if they had the essential knowledge background to do so, most of them would resort
to readily applicable assessment rules in their effort to provide a timely and accurate verdict for
each suspicious case. This has important implications for the knowledge representation forms that
should be selected for the WP3 anti-fraud architecture.

PUBLIC
Page 123 of 255
E-COMPASS
5.2
Data mining for e-sales operations
In order to implement the project data mining for e-sales solution, five modules will be developed:

DM1: Data collection and consolidation

DM2: Competitor price data collection

DM3: Business Scorecard – optimization potential analysis

DM4: Automated procedures by applying rule-based actions

DM5: Visualization – SME E-COMPASS cockpit
The following tables show the functional as well as the non-functional requirements derived by analysing
the requirements given by the questionnaires and the focus group workshops as described in section 4. The
module which should implement each requirement is specified.
Table 31. Functional requirements for data mining (web analytics) applications
Functional Requirements
ID
Functional Requirements
Module
F01
Retrieving from digital footprint to analysis of visitors’ behaviours
DM1
F02
Scraping prices from competitors’ websites
DM2
F03
Cleansing and linking of all retrieved data (provided by DM1 itself and DM2)
DM1
F04
Storing variables or metrics (provided us by DM1) and data processing to carry out the
data quality assurance
DM3
F05
Applying Data Mining techniques over data: behaviour visitors, products and
competitors
DM3
F06
Run actions if a pre-defined rule matches for the analysed data
DM4
F07
Visualization of the analysed data (Frontend)
DM5
F08
Storing user input from e-shop owners/user administration (Backend)
DM5
F01 Retrieving from digital footprint to analysis of visitors’ behaviours
Collecting and physical gathering of saved and calculated data (according to predefined metrics) referred to
visitors digital footprints.
F02 Scraping prices from competitors’ websites
Scraping of own and competitors’ product and price data from own and competitors’ websites for price
comparison (see section 6.c.ii Data Mining).

PUBLIC
Page 124 of 255
E-COMPASS
F03 Cleansing and linking of all retrieved data
New generated data from web analytic tools as well as from scraping methods have to be cleaned,
organized, and converted to standard formats (RDF) for smart information processing. These data have to
be also linked within a semantic model with other existing repositories containing useful information.
F04 Storing variables or metrics and data processing to carry out the data quality assurance
The raw data, from DM1, are stored in a database where it comes to cleaning. For example: visits
within zero time, robot behaviours and outliers.
F05 Applying Data Mining techniques over data: behaviour visitors, products and competitors
Taking into account, the results of the extended interviews, it will describe the main user requirement and
his enhanced version.
1. User requirement about the information of visitors’ behaviour
The information about the visitors’ behaviour nowadays is not supervised by the majority of the enterprises
and the added value of the data mining for e-sales operations is the recognition of the different typologies
of visitors depending mainly on the page view per visit and time of visit. The data modelling technique to
obtain the typologies will be the clustering.
2. User requirement about the information of visitors’ behaviour enhanced
According to the information showed in the traceability matrix (see table 1), it will describe the improved
version of the user requirement. For this, the next table describes the dimensions whereby the analysis and
outcomes of the data mining for e-sales operations service will be shown and explained.
In this sense, the almost features that corresponding to the each dimension are considered in the
traceability matrix (see Table 1), but in the case of How dimension’s features, these are related to the
variable which is generated by the data mining service.
Table 32. Dimensions for data mining e-sales operations
Dimensions for data mining e-sales operations
By these dimensions, the main requirement will have a significant improvement and the outcomes of the
data mining for e-sales operations service will contribute in a better way in the project.
3. User requirement about the information of products

PUBLIC
Page 125 of 255
E-COMPASS
The added value of the data mining for e-sales operations is the recognition of the different typologies of
products depending mainly on price product and origin. The data modelling technique to obtain the
typologies will be the clustering.
4. User requirement about the information of competitors
In the same way, it can find types of competitors depending mainly on average price product and origin.
The data modelling technique to obtain the typologies will be the clustering.
5. User requirement about the information of visitors’ behaviour, products and competitors
For this user requirement is necessary the attributes of visitor behaviour, products and competitors:
number the visits of the visitor, average time, number of purchased product and monetary value of
purchase and average price product based on the competitors. Also, the data modelling technique to
obtain the typologies will be the clustering.
F06 Run actions if a pre-defined rule matches for the analysed data
E-shop owners can use pre-defined rules to run an action if a rule matches for the analysed visitors’ or
competitors’ data, e.g. send an e-mail to e-shop owner if the prices for a specified product of all
competitors are lower than the own price.
F07 Visualization of the analysed data (Frontend)
Visualizing the business figures of F07 as well as some of the actions of F08 in form of graphs.
F08 Storing user input from e-shop owners/user administration (Backend)
Administration website for e-shop owners. Here they can add the data of their competitors for price
scraping (URLs of competitors) or select pre-defined rules and actions.
Table 33. Non-Functional requirements for data mining (web analytics) applications
Non-Functional Requirements
ID
Non-Functional Requirements
Module
N01
Interface from/to Google Analytics in order to exchange data of user behaviour
DM1/DM4
N02
Interface to Linked Open Data
DM1
N03
VPN-Service for changing IP-address during scraping of competitors’ data
DM2
N01 Interface from/to Google Analytics in order to exchange data of user behaviour
Section 6.a.ii shows the table of the data, which can be collected by the interface to Google Analytics.
N02 Interface to Linked Open Data
Development of a set of Mapping methods and linking with other external tables with the aim of
establishing a common/transparent interface to access data from heterogeneous sources. Final accesses
will follow standards like SPARQL queries.

PUBLIC
Page 126 of 255
E-COMPASS
N03 VPN-Service for changing IP-address during scraping of competitors’ data
The VPN-Service is used to change the own IP address during the scraping of competitors’ data, because
some web servers will automatically close the connection if an IP address opens the connection many
times.

PUBLIC
Page 127 of 255
E-COMPASS
6
Existing data used by Projects pilot SMEs and their collection for future analysis
6.1
Existing Data Used by SME
6.1.1 Anti-fraud
Work package (WP) 2, apart from collecting and defining the user requirements of SMEs in their combat
against fraudulent transactions, has also as additional scope: the mapping of the data that SMEs are
collecting, storing and handling through their transactions. The exercise concerning WP2 data description
and collection was through the focus groups and personal interviews with selected e-shops and ecommerce executives of the SME-AGs, to inquire the SMEs regarding the type of data that they typically
collect and identify during the transactions. Table 34 aggregates all relevant data parameters and presents
their qualitative characteristics. This data comes from SMEs (e-shops) selling retail products and offering
services online (mostly travel and accommodation). In WP3 for the development of the anti-fraud
application, the data already collected by M5 of the project as well as new data that will be provided by the
pilot SMEs will be camouflaged (for privacy and security reasons), and pre-processed for the application of
data mining and computational intelligent models.
Table 34. Data parameters and qualitative characteristics used by selected SMEs
Data description
E-commerce Category
Data format
Data type
Address Post Code
Address Street Name
Address Street Number
Address Country
Address Telephone Country
Address Telephone Number
Bank Name (as set by the user)
Card Holder's Name
Card First 6 Digits (i.e. BIN: Bank
Identification Number)
Card Hash
BIN Bank Name
BIN Country
Passenger
Passenger Date of Birth
Passenger Passport #
User E-Mail
User First Name
User Last Name
User Middle Name
User Nickname
Acquisition
Address information
Address information
Address information
Address information
Address information
Address information
Address information
Card Information
String
String
String
String
String
Country
Country
String
String
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Card Information
Card Information
Card Information
Card Information
Card Information
Customers' Details
Customers' Details
Customers' Details
Customers' Details
Customers' Details
Customers' Details
String
String
String
String
Country
String
String
String
Email
String
String
Raw Data
Raw Data
Raw Data
Calculated
Calculated
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data

PUBLIC
Page 128 of 255
E-COMPASS
Data description
E-commerce Category
Data format
Data type
Username
Device Hostname
Device ID
Device IP Address
Device City
Device Country
Flight destination city
Flight destination country
Flight booking window
Flight departure city
Flight departure country
Hotel - Number of rooms
Hotel booking window
Hotel check-in date
Hotel check-out date
Hotel city
Hotel country
Hotel number of days to
cancellation fees
Hotel number of nights
Purchase Product Name
Purchase Product Quantity
Purchase Product Type
User behind proxy
Shipping Address City
Shipping Address Post Code
Shipping Address Street Name
Shipping Address Street Number
Shipping Country
Shipping Telephone Country
Shipping Telephone Number
Payment type
Card used is 3D Secure
Purchase Amount
Purchase Date
Number of different Cards used
during the same Session
Flight one-way reservation
Number of Checkouts from same
device in the last two days
Number of checkouts from same
Customers' Details
Customers' Details
Customers' Details
Device
Device
Device
Device
Device
Flight Travel Details
Hotel Travel Details
String
String
String
String
String
String
String
Country
String
Country
Number
String
Country
Number
Number
Date
Date
String
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Calculated
Calculated
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Product
Product
Product
Proxy
Shipping information
Transaction Details
Transaction Details
Transaction Details
Country
Number
Number
String
String
String
Flag
String
String
String
String
Country
Country
String
String
Flag
Number
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Calculated
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Raw Data
Transaction Details
Transaction Analysis
Date
Number
Raw Data
Calculated
Flag
Raw Data

PUBLIC
Page 129 of 255
E-COMPASS
Data description
device the last 2 hrs.
Number of Checkouts from same
IP in the last two days
Global number of Checkouts
within the last 2 days for Credit
Card
Global number of Failed Payment
Attempts within the last 2 days
for Device
Global number of Failed Payment
for IP
Global number of Purchases
within the last 2 days for Device
Global number of Purchases
within the last 2 days for IP
Number of Checkout within the
last 2 days for Credit Card
Number of Failed Payment
for Device
Number of Failed Payment
for IP
Number of Purchases within the
last 2 days for Device
Number of Purchases within the
last 2 days for IP
Card used successfully 4 months
ago or before
Global: Device attempt to checkout more than 5 times in a day

E-commerce Category
Data format
Data type
Number
Velocities
Number
Velocities
Number
Velocities
Number
Velocities
Number
Velocities
Number
Velocities
Number
Velocities
Number
Velocities
Number
Velocities
Number
Velocities
Number
Velocities
Number
Velocities
Number
Velocities
PUBLIC
Page 130 of 255
E-COMPASS
6.1.2 Data mining
As it can see in the table of existing and new data for gathering in the previous point, these are the present
variables that it will use for the data mining service: Language, city, region, country, continent, page view
per visit, time of visit.
Table 35. Existing data used for web analytics
Existing Data Used for Web Analytics
web analytic
tool
Google Analytics
PIWIK
survey
Visitor's
origin
- referrer/origin (URL)
- medium (search engine,
AdWords, direct, referral)
- keywords (search engine)
- landing/entry pages
- geographical origin (continent,
country, region, …)
- campaign (that led the user
here)
- social network
- number of visitors/visits that
started/finished on this page
- referrer/origin (URL)
- medium (search engine, AdWords,
direct, referral)
- campaign
- keywords (external engine)
- social network
- geographical origin (continent,
country, region, …)
-
Visitor's
behaviour
-
-
- Pages which are most often viewed
- Number of page views per visit
(depth of visits)
- Time of stay per visit (duration of
visits)
- Applied key words within the own eshop search
- Most common exit pages
- Most common page view sequence
(click paths). These variables are
necessaries: Previous, Next page or
action.
- Most common landing page
- Average time on site
- Bounce
- Average time on page
- Time on page
- Page Views
- Page View per visit
- Visit bounce rate
- Percent visit with search
-
pageviews
pageviews per session
session duration
internal search (search terms,
start/destination page)
exit pages
previous and next pages (to
determine click paths)
time on page
count of sessions
days since last session
bounces (sessions with only one
page hit)
time (When did the user
something: hour, weekday …)
social network activities

pageviews/visits
number of actions per visit
number of visits per visit duration
internal search (search terms, search
terms without result, destination
page)
- exit pages
- transitions (previous/following pages)
PUBLIC
refer
Key words
entry pages
search phrases
Geographical origin (Continent,
Region, Country, City, location)
- Hostname
- Network Domain
- Network Location
Page 131 of 255
E-COMPASS
Visitor's
attributes
- number of visitors (per day,
week, month, …)
- user type (new or returning)
- device
- operating system
- browser
- flash version
- Java enabled?
- screen (resolution, colours)
- language
- number of visitors (per day, week,
month, …)
- number of returning visitors
- number of visits by visit count
- device
- operating system
- browser
- plugins
- screen (resolution, colours)
- language
-
visitors per week
new visitors
Number of Dark frequent visitors
Technical equipment of the visitors
(e.g. browser-version, screen
resolution, browser, operating
system, operating system version,
immobile)
- Number of visits per visitor
- Most common language
- New visits. Maybe like new visitor
purchasing
behaviour
(needs
ecommerce
tracking
activated)
- conversion rate
- average order value
- time to transaction
(days/sessions between users'
purchases and the related
campaigns)
- shopping cart abandonment
(needs Funnels)
-
conversion rate
average order value
purchased products
best products
best product categories
abandoned carts metrics (visits with
abandoned carts, revenue left in
carts)
- conversion rate by visit/session
length
- Number of visitors who did a
purchase
- Average value of a shopping cart of
the e-Shop
- Number of visitors who put a
product into the basket
- Number of visitors who break up the
checkout (purchasing) process
- Average time of stay in the e-Shop
until purchasing products
New function
for
monitoring
- Users Device Type (mobile,
tablet, …)
- Device Type (desktop, smartphone,
tablet)
- Analysis of the access of mobile
devices
- Categorization of user groups
(visitors segmentation, type of
resource, name of resource)
- Qualitative user survey
- Page-oriented user feedback
- Form field analysis
- Comparative tests (e.g. A/B-tests)
- Mouse-tracking
The table above is just an excerpt of all data available in Google Analytics and PIWIK. Google Analytics
offers over 130 metrics in over 200 dimensions. Although only certain dimensions and metrics can be used
together to create valid combinations, these are still too much to mention them all. This is true especially
for Google Analytics as it allows to query up to 7 dimensions. Furthermore not all valid combinations are
meaningful in our use case.




classic Google Analytics (ga.js)
classic Google Analytics (ga.js) with ecommerce tracking (_trackTrans() )
universal Google Analytics (analytics.js)
universal Google Analytics (analytics.js) with ecommerce tracking (ga('ecommerce:send') )

PUBLIC
Page 132 of 255
E-COMPASS
Table 36. Pilot e-shop's technical features
Pilot e-Shop’s Technical Features
pilot e-shop
e-shop-Software
used web analytic tools
*1
how to activate
Google Analytics ecommerce
tracking
simply activate in Magento and
provide Google Analytics Account
Data
Gripaid
Magento
classic Google Analytics
Quantcast
Merseyfuels
Zen Cart
is possible with the free zen cart
plugin “Easy Google Analytics”
none, unknown
Data
Venture Packaging
Supplies Limited
Magento
Rachel Wears
WordPress 3.9.1 with
WooCommerce plugin classic Google Analytics
2.1.9
simply activate in WooCommerce
and provide Google Analytics
Account Data OR edit functions.php
like here
Terry's Boutique
unknown
universal Google Analytics
unknown
Printworks Chester
unknown
unknown
Atlas Sport S.L.
Magento
Data
Custom
Classic Google Analytics
with eCommerce
tracking and universal
Google Analytics with
eCommercce tracking
Custom
eTravel
Companies providing data and Pilot Users
As a result of the focus groups and interviews, several companies are willing to share data anonymously
with the project’s members, as well as take part in piloting the initial developed prototypes. The objective
is that these companies will evaluate the performance and functionality of the initial prototypes using their
own data. The following tables describe all SMEs which have explicitly indicated that they are interested in
participating in the project as a piloting company. For each company a description of its activity, the sector
which it belongs to, the kind of product it sells and the SME-AG it is member of are provided.

PUBLIC
Page 133 of 255
E-COMPASS
Pilot Users
Table 37. Pilot e-shop's descriptions
Pilot e-Shop’s descriptions I
Name
SME-AG
HALTON
http://www.terrysboutique.co.uk/

Trading of human hair
extensions, fashion
accessories, cosmetics,
scarves, jewellery and
footwear.
Hair, nails & beauty,
jewellery, bags & purses,
HALTON
Business Cards, Flyers,
Leaflets, Compliment Slips,
Letterheads, Posters, Display
& Exhibition
HALTON
Rachelwears.com provides luxury clothing for
grown up women in easy to wear classic
styles that will not date. All the garments are
made in the UK using only the finest fabrics
and yarns and each style is meticulously
designed to flatter a real woman’s body.
These classic pieces are all beautifully gift
wrapped before dispatch so that every
customer receives their own little ‘present’.
Customers can also use a telephone service
for advice on styles and fit and will be kept
up to date with news and offers and even the
occasional recipe!
dresses & skirts, knitwear,
tops, trousers & leggings,
accessories
HALTON
Grip Aids products are manufactured to aid
the end user enhance the grip on a multitude
of objects in sports and in recreation. Each
glove is specifically designed to help the end
user improve the control they have on an
object by taking the strain away from the
wrist and hand with the use of the unique
patented strapping system. Sufferers of short
or long term loss of grip through injury or
medical condition find this product appealing
as it allows the end user to continue in sports
or recreation without the worry that their
grip will be lost. Stroke patients and arthritis
sufferers are amongst those users who
benefit as the patented strapping system
helps to ease pain and strain when carrying
out repetitive or strenuous tasks
gloves, mittens and medical
devices
http://www.rachelwears.com
Terry’s Boutique is one of the UK’s leading
retailers of human hair extensions, fashion
accessories, cosmetics, scarves, jewellery and
footwear all across the UK, Ireland & Europe.
It serves brands such as: Remi Goddess, Pink
Lust, Directions, Crazy Colour, Moroccan Oil,
Tangle Teezers, sleep in rollers, LYDC hand
bags, Impulse earrings + Jewellery, Valasio
Scarves, Hunter Willies to Silky Tights &
Socks.
Sector & Products
Servicing the print and design industry for
over 20 years. Offering the same level of
service from short-run printing of 1 or 2
posters to 500,000 flyers, nationwide, fast
turnaround, and quality print at affordable
prices. These qualities have been the core of
the business for the last 20 years.
http://www.printworkschester.com/
https://gripeeze.com/
Description
PUBLIC
Page 134 of 255
E-COMPASS
Table 38. Pilot e-shop's descriptions II
Pilot e-Shop’s descriptions II
Name
SME-AG
HALTON
www.merseyfuels.co.uk
HALTON
http://www.venturepackagingsu
pplies.co.uk/
www.smartpet.gr
EPK
Description
Venture Packaging provides products like All
Night Burners, Bottom Grates, Calor Gas Bottles,
Cast Iron Soot Box ,Chimney Balloon, Chimney
Cowls, Coal Bunker , Cold Hods (Scuttles),
Companion Set, Defra Approved Stoves, ECO
PANEL, FIRE GUARDS(SPARK GUARDS) ,
Fireclay/Firebacks,
Fire
Bricks,
Fireside
Accessories And Tools, FLUE PIPES, Flying
Dutchman Stores, Fuels, Coal, Logs,Peat, Garden
Section, Gift ideas, Gully Grids, Log Baskets, Log
Splitter, Mobile Heater, Paraffin Odourless, Soild
Fuels Frets, SOLID FUEL KITS, Solid Fuel Section,
Stove Glass Replacement, STOVE SPARES/PARTS,
Woodburning/Multifuel Stoves
Venture Packaging Supplies Limited have been
trading as an online packaging supplier since
February 2009, we proudly make the claim that
we are now one of the largest mailing bag
suppliers on EBay with a positive feedback score
of 99.9% extending in excess of some 55,000
sales. The Directors combined have over 45 years
of experience in the packaging industry and fully
understand the requirements of the end user,
you the customer.
The smartpet.gr is the online store for exhibition
and sale pet products of the physical pet shop
Stamatis. Headquartered in Saint Theodore in
Corinth owned store 300sq.m on the Old
National Road Athens-Corinth. The store was
established in 2005 and aims to provide to
animal lovers anywhere with quality products for
their little ones at affordable prices for everyone.
Sector & Products
fuels, heating and gardening
supply
Venture Packaging
A4 Sticky Printer Address
Labels, Acid Free Tissue
Paper, Blue Mailing Bags,
Board Backed Envelopes,
Brown Paper Kraft Bags,
Bubble Wrap, Document
Wallets, Grey Mailing Bags,
On-line pet shop providing
food and accessories for
most kind of pets such as
dogs, cats, fishes, mice,
reptiles.
Jewellery and watches
EPK
www.eleftheriouonline.gr

Eleftheriou online shop started its operations in
2010. This need has inspired by John Eleftheriou
who took the decision to make the big step of
entering the world of internet. You just have to
let us guide you to the depths of Jewellery,
watches and Gift. Here you will discover a wide
range of jewelry for Women, Men and Child with
excellent quality.
PUBLIC
Rings, bracelets, earrings,
necklaces,
pendants,
crosses, leg chains, neck
chains, cufflinks, charms,
amulets,
zodiacs,
monograms,
baptismal
crosses of gold, silver, alloy
metal and steel, diamond
rings, glittering diamonds
and jewellery with pearls,
sapphires,
rubies
and
emeralds.
Page 135 of 255
E-COMPASS
Table 39. Pilot e-shop's descriptions III
Pilot e-Shop’s descriptions III
Name
SME-AG
EPK
www.koxyli.gr
EPK
www.stylebrands.gr
EPK
www.v-cubes.com
EPK
www.tourix.gr

Description
The company Koxyli is a Greek family business,
focused on producing high quality chocolate
products, having as main objectives the
innovation, high product quality and continuous
development. The company's headquarters is
located in Loutraki Corinth private facilities. The
company operates in the food sector without
sugar in the last seven years, bringing numerous
innovations. Countries exportingare. England,
Germany, Bulgaria, Cyprus, Kazakhstan, Denmark,
Romania.
Sector & Products
Food industry (Chocolate)
Products are based on
chocolate but also produces
juices, jams, pastels, cakes,
following
these categories
based on ingredients: bio,
sugar free, lactose free, gluten
free, super foods
This is the online shop of the jeweller and craft
Kontzamichalis based in Corinth, selling online Jewelleries and watches B2C
watches, jewelleries and accessories.
V-CUBE™ is a worldwide registered trademark of
VERDES Innovations S.A. The company was
founded in 2008 and is located in the Corinthos
prefecture of Southern Greece. Verdes
Innovations S.A’s vision is to provide people a
unique opportunity to extend their capabilities
and enhance their cube solving experience
through the use of the V-CUBE™ Technology and
to provide them with new means to express their
creativity. All contemporary, safety and high-tech
standards are used in order to accommodate
customers, including authorized V-CUBE™ dealers
and cubers worldwide, so that they acquire our
latest technology cube games efficiently and with
all the necessary support.
Tourix specializes in providing strategic emarketing
services
to
businesses
and
organizations in the tourism sector, supporting
them to design and implement their digital
strategy and marketing campaigns. Some of the
services offered from Tourix are strategic
eMarketing plan, website construction, search
engine
marketing,
online
reputation
management, social media marketing, corporate
identity.
PUBLIC
Gaming Industry
V-CUBE™ products are a
uniquely
designed
and
constructed series of skill
games.
They
are
3D
mechanical cubic puzzles that
rotate smoothly on the 3
based axes of the coordinate
system.
V-CUBE™ technology made
possible the construction of a
cube with an unlimited
number of layers, providing
safe and smooth rotation!
Consulting in e-tourism
Services such as e-marketing
plans,
SEM,
web-site
development. Will provide to
project access to its customers
(mainly Hotels of Corinthos)
for becoming pilot users of the
applications.
Page 136 of 255
E-COMPASS
Table 40. Pilot e-shop's descriptions IV
Pilot e-Shop’s descriptions IV
Name
www.coozina.gr
SME-AG
Description
Sector & Products
FBA
(SME
Partner)
COOZINA. GR is the online shop of the wholesaler
in Greece Fornaro Business Agency (FBA).
Operating as a representations agency on an
exclusive basis, FBA assists global manufacturers
to create and grow their business in the Greek
market by providing them with independent local
market insights, performing sales activities and
systematically building a strong brand for them in
the Greek market. The new era of e-commerce
has been an opportunity for the company to
directly enter the retail sector with the
development, launch and operation of an e-shop
for kitchenware –an important market sector for
the company.
Kitchenware
appliances
GAIA
Atlas began as clothing, footwear and accessories
shop that was born in late 1996 in San Sebastian,
a city associated with the image and the surf and
skate culture. Shortly after that first store in San
Sebastian, another store opened at the provincial
level in Gipuzkoa. Moreover, Atlas has a career of
more than 15 years, and we want to continue the
business also through Internet
GAIA
KLIK BAT is an online ecommerce platform
specially developed for the shops in Elgoibar, a
small village located in the Basque Country,
Spain. KLIK BAT offers the shops a free online
platform to publish the products they offer in
their shops.
Clothes for women and
men, shoes, toys, CDs and
DVDs, sport equipment, etc.
GAIA
ECHEBARRIA SUMINISTROS is the biggest
Industrial Supply company in Alava, North of
Spain. They offer all kind of products related with
constructions, screws, labour protection, etc. The
new era of e-commerce is a new opportunity for
this company which allows them to open their
market and offer as much products as they offer
in their offline shop with the same effectiveness.
Building materials, labour
protection products, screws,
industrial machinery, etc.
http://atlasstoked.com/
http://elgoibar.klikbat.com/es
http://echebarriasuministros.com

PUBLIC
and
clothing, footwear
accessories
kitchen
and
Products: Urban Clothing:
clothing skate, surf and
urban wear , surf and skate
shoes, streetwear shoes,
streetwear
accessories,
skateboard: skateboards and
skates
Page 137 of 255
E-COMPASS
Table 41. Pilot e-shop's descriptions V
Pilot e-Shop’s descriptions V
Name
SME-AG
Description
Sector & Products
GAIA
Lencería La Vascongada is an offline shop located
in Vitoria-Gasteiz, which saw an opportunity to
open their business in the online shop through
which they offer the same products as they have
in the offline shop.
Swim
wear,
women´s
clothes, summer clothes,
lingerie.
GAIA
A company that belong from the communication
sector, they have large experience offering all
kind of services to companies that need to
improve their communication and marketing
strategy.
Marketing
services,
consultancy, e-learning.
ATEVAL
Gestiweb is a company founded in 2002 with aim
of integrating different types of solutions (free
software) and oriented towards universal access
to the Internet, ie business needs in the area of
Information Technology and Communication
(ICT). It operates in Europe. Gestiweb has a group
of experienced professionals in research and
software development applied to business
management services, using cutting edge
technologies. It has a very active role in the
development of solutions on free software
systems, aimed at optimizing business processes
and improving administrative and commercial
enterprise results.
http://www.lavascongada.com/
http://www.quick.es/
http://www.gestiweb.com/
ATEVAL
www.tormodel.es
ATEVAL
www.macronutricion.com

Sector:
Information and
communication technologies
Products:
Technology
Consulting,
Free
ERP
software,
Software
as,
Monitoring spaces and
enclosures, web Design, web
programming, Development
advanced programming.
Positioning Internet, Hosting
and Domains.
Tormodel is dedicated to the sale of radio
controlled and model airplanes through internet,
all around Europe but mainly in France. In it, you
can find wide variety of accessories and products
for the electric model airplane.
Sector: Toy industry
Products:
Aircrafts,
Helicopters , MultiRotores,
Cars / Motorbikes , Ferries /
Boats ,engines Regulators,
servos, Radio / Rx Batteries,
Chargers, helices, crystals
Electronic , Accessories FPV
&
AV,
Accessories,
adhesives, tools, Depron /
EPP , fuels ,Parts Models,
Static models
Pro-nutrition is dedicated to the marketing and
distribution of accessories and nutrition sports.
Distribution network operates in Spain, England,
France, Italy and Germany, offering customers a
commercial and logistics personalized services.
Sector: Sports nutrition
Products:
Stimulant
/
Energy, Nitric Oxide, Amino,
proteins, Fat Burners,
Creatine
,
accessories,
carbohydrates,
Vitamins and minerals,
Hormonal,
Meal
replacement, Energy Bars /
Protein.
PUBLIC
Page 138 of 255
E-COMPASS
Table 42. Pilot e-shop's descriptions VI
Pilot e-Shop’s descriptions VI
Name
SME-AG
Description
Sector & Products
Sector: Information and
communication technologies
ATEVAL
http://www.nessys.es/
NESSYS IT is an ICT managed services company
established in Valencia that operates under very
traditional business principles: commitment,
service and reliability. Founded in April 2010,
Nessys is dedicated to contribute with other
companies or organizations in the management
of their Systems as consultants and service
projects all around Europe.
Products: Shared Hosting,
VPS servers, Corporate
email, SSL certificates ,
domains,
Mobile
Applications
services : Technical Support,
sysAdmin, ICT consultancy,
Marketplace
Operations,
Custom Development
Sector: Erotic products
ATEVAL
www.mipuntito.com
6.2
Mipuntito.com is the sex-shop leader in online
sales of erotic toys. Sells its products throughout
Spain. MiPuntito Sex Shop offers the most
extensive variety of sex toys with more than
7500 items in stock. Leader in selling dildos,
vibrators, Chinese balls, lingerie, accessories for
your parties and more.
Products:
aphrodisiacs, erotic games,
toys xxx, Linea BDSM,
lubricants and creams,
condoms, lingerie, sheets,
shirts, accessories and
costumes, etc.
Data that will be produced in the Project
6.2.1 Anti-fraud

The real-time anti-fraud application that will be designed and developed in WP3 following an hybrid
architecture, pilot-tested and fine-tuned in WP5, is to be considered as an advanced and intelligent expert
system that will support the decision making process and review automatically day-to-day transactions.
Thus, the system’s objective is not to generate meta-data for the user, but to provide a score for each
transaction that will help the fraud expert to decide its degree of suspiciousness.
Furthermore, the application will incorporate a set of expert rules for monitoring each transaction, fully
configurable from the user in order to meet the requirements of each e-shop and customisable to each ecommerce market. Therefore it is very crucial to guarantee a sufficient level of quality and quantity for the
input data in order for the application to operate properly and reliably:
In terms of input-data Table 34 (Section 6.1.1) describes the data parameters already defined and collected
by the consortium for the designing and development of the anti-fraud system. For (some) pilot SMEs that
do not have the technology to track and collect the data characterised as “Velocities” (Data Type) or

PUBLIC
Page 139 of 255
E-COMPASS
“Transaction Analysis” (E-commerce Category), the project’s anti-fraud application will enclose a special
module called “Transactions Analytics Toolkit” (Task 3.3) that will generate, collect and store this data for
each transaction, communicate with the rules and inference engines.
6.2.2 Data mining
For competitors’ analysis, the following data have to be automatically generated by web scraping as well as
manually by the e-shop owners:
Table 43. Competitors’ analysis: data have to be automatically generated
Competitors’ Analysis: data have to be automatically generated
Data
Creation
Description
e-shop owners
For reference to the following datasets a
dataset for each e-shop owner is needed
- eshopID
- eshopName
- eshopURL
- eshopDescription
- eshopOwner
- eshopType
competitors
The e-shop owners have to specify the their
competitors’ e-shops (competitorsEshopID)
for comparing their prices with the prices of
their competitors
- competitorID
- eshopID
- competitorsEshopID
- relationshipDescription
products
The tool has to scrape the products of the
e-shop owners and those of their
competitors
- productID
- productLabel
- productType
product prices
The tool has to scrape the product prices of
the e-shop owners and those of their
competitors
- eshopProductID
- eshopID
- productID
- productDescription
- price
- currency
- availability
- creationDate
Taking into account Table 43 concerning the new data has to be automatically generated, it will be
produced in the project the following outcomes:
 The evolution of the number of cases of each typology about the information of visitors’ behaviour in
a period of time.
 The percentage of cases that have been grouped into the diverse types of visitors.
 The number of cases that have been classified into the diverse types of fidelity.

PUBLIC
Page 140 of 255
E-COMPASS
6.3
Semantic Data Model Initial Proposal
WP2 includes the design of the semantic model that will be used for the data exchange; it also explains how
this model relates to existing data as well as the transformation processes required. The goal is to enable
the integration of all the data used in the project using semantic technologies and its publication as
(private) Linked Data.
This section describes the initial SME E-COMPASS project’s semantic model. From a technical point of view,
the semantic model is an OWL ontology. The ontology is being developed to handle data related to online
transactions, including all the elements required to annotate the data produced and share it with different
applications. The ontology will be linked to the schemas of existing data sets by means of mappings,
enabling their integration into a common data model.
The first version of the SME E-COMPASS Ontology has 45 classes (groups of individuals sharing the same
attributes), 34 object properties (binary relationships between individuals) and 34 data types properties
(individual attributes).
Figure 119 shows a general overview of the ontology generated. More in detail, a description of the
ontology is provided below.
The main concept of our domain is an e-shop. An e-shop has one or several pages and also an e-shop
owner. Each e-shop owner has an address. Addresses are composed by address street, address number and
location, i.e. city, region, country and continent. The e-shop owner can have competitors, who are e-shop
owners of other e-shops. Visitors visit pages (which belong to an e-shop). A user is a visitor who is
registered. A customer is a user who makes a purchase. If it is the first purchase of a customer, he/she is a
new customer. Customers have an address. A visitor has a device, which has a browser, an operating
system, a proxy and an IP address. IP addresses have a location, an ISP provider and belong to an
organization. Pages contain items, i.e. products or services. A visit has an entry page and an exit page and
follows a path which has a next page and a previous page. During a visit a transaction may be completed,
both successful and failed transactions are possible. Transactions have an associated payment method.
Therefore it is clearer to understand that the ontology is divided into eight parts, all of them connected by
means of one class. These parts correspond to e-shop, address, location, visitor, visit, page device and IP
address. These elements are described below.

PUBLIC
Page 141 of 255
E-COMPASS
Figure 119. General scheme of the complete ontology

PUBLIC
Page 142 of 255
E-COMPASS
Figure 120 presents the classes, object properties and attributes for modelling e-shops and e-shop owners.
As we have mentioned, an e-shop has one or several pages and also an e-shop owner. Each e-shop owner
has an address. The e-shop owner can have competitors, who are e-shop owners of other e-shops. The
attributes for an e-shop are date of last transaction, number of new customers, visit bounce rate, number
of customers and number of transactions. The only attribute for an e-shop owner is name.
Figure 120. Classes, object properties and attributes for modelling e-shops and e-shop owners


Figure 121 represents visitors. A user is a visitor who is registered. A customer is a user who makes a
purchase. If it is the first purchase of a customer, he/she is a new customer. Customers have an address. A
visitor has a device. The attributes for visitors are bounced rate and number of visited pages. The only
attribute for users is username. The attributes for customers are number of successful transactions and
number of failed transactions.

PUBLIC
Page 143 of 255
E-COMPASS
Figure 121. Classes, object properties and attributes for modelling visitors
Addresses are modeled as shown in Figure 122. The ontology contemplates three types of addresses, i.e.
shipping address, contact address and billing address. An address has two attributes, street number and
street name. Location is modeled as a class because it will be used in another relationship. A location
(Figure 123) has city, region, country and continent. The only attribute for location is post code.

PUBLIC
Page 144 of 255
E-COMPASS
Figure 122. Address subclasses contemplated by the ontology
Figure 123. Class modeling a location

A visitor has a device, which has a browser, an operating system, a proxy and an IP address. IP addresses
have a location, an ISP provider and belong to an organization. Attributes for device are device ID and host
name. The only attributes for operating system and browser are OS version and browser version,
respectively.

PUBLIC
Page 145 of 255
E-COMPASS
Figure 124. Class Device in the ontology
Figure 125. Visit modeling in the ontology

PUBLIC
Page 146 of 255
E-COMPASS
Visitors visit pages. Visits are essential to our project because it captures the behavior of a visitor when
visiting our e-shop. Figure 125 shows how visits are modeled in our ontology. A visit has an entry page and
an exit page and follows a path which has a next page and a previous page. Class navigation step is used to
model the path that the user follows from the entry page to the exit page. Each new navigation step has
only one attribute number. During a visit, transactions (Figure 126) may be completed; both successful and
trustworthy transactions are possible. A purchase is a subclass of successful transaction. Transactions have
an associated payment method which can be PayPal, bitcoins, bank account or card. Credit card is a
subclass of card.
Pages (Figure 127) contain items, i.e. product or services to be sold. Products and services must be
extended in the next version of the ontology in order to model their common features. Specific items of an
e-shop should be modeled by defining a domain ontology for a specific domain, i.e., travel, books, music,
etc. The attributes for a page are date of last visit and title. The attributes for an item are price and price
currency.
The current version of the ontology contains all the necessary elements to represent and capture data that
will be used by the application being developed in the project, i.e. the online anti-fraud system and the
data mining for the e-sales system. Future versions of the ontology will include more attributes as well as a
better modeling for transactions, products and services.
Figure 126. Transaction class in the ontology

PUBLIC
Page 147 of 255
E-COMPASS
Figure 127. Class Page in the ontology

Epilogue
The aforementioned ontology is the initial step for moving towards the activities of the two last tasks of
WP2, namely Task 2.3 “Mapping Design” and T2.4 “Linked Data repository development” that both will
reported in Deliverable 2.2 “Semantic Model & Database Mappings” (first version).
As far as the user and data requirements collection process is concerned, this has been concluded with the
accomplishment of this deliverable and all conclusions and recommendations are injected in WP3 and WP4
design and functional requirements tasks. Direct channels of communication have been established
between the consortium and a satisfying number of interested and potential piloting SMEs in all four
regions. Therefore any additional data provision and elaboration on the user and functional requirements if
needed will be addressed to the SMEs during the design processes of both applications.
By the conclusion of the deliverable the consortium of SME E-COMPASS Project has achieved to validate
the hypotheses of the user requirements for micro, small and medium enterprises active in e-commerce
through dedicated and well-structured activities that raised the awareness of the project to SMEs and
highlighted the benefits that they can gain via the evaluation and participation as end-users in the project.
The consortium is pretty confident that with the completion of user and data activities, has adequately
achieved to conceptualize and in-depth interpret the needs and the challenges of the SME-AGs members.

PUBLIC
Page 148 of 255
E-COMPASS
7
References
1. Abensur, E. (2013). Do you know who I am? The importance of personalisation. Retrieved May
21, 2014, from https://econsultancy.com/blog/63050
2. Bamfield, J. A. N., (2013). Retail Futures 2018: Shop Numbers, Online and The High Street - A
Guide to Retailing in 2018.
3. Ecommerce Europe. (2013). Western Europe B2C E-commerce Report 2013 (light). Retrieved
May 15, 2014, from http://www.ecommerceeurope.eu/cms/streambin.aspx?documentid=4334
4. El comercio electrónico y el uso de las TIC - Instituto (2012). Retrieved May 23, 2014, of
http://www.ine.es/ss/Satellite?blobcol=urldata&blobheader=application%2Fpdf&blobheader
name1=ContentDisposition&blobheadervalue1=attachment%3B+filename%3DCifrasINEComo
pdf.pdf&blobkey=urldata&blobtable=MungoBlobs&blobwhere=972%2F963%2FCifrasINECom
opdf%2C1.pdf
5. Estudio sobre Comercio Electrónico B2C 2012 (Edición 2013) Retrieved May 23, 2014, of
http://www.ontsi.red.es/ontsi/sites/default/files/informe_ecomm_2013.pdf
6. Eurostat. (2014). Individuals having ordered/bought goods or services for private use over the
Internet in the last three months (tin00067). Retrieved May 15, 2014, from
http://epp.eurostat.ec.europa.eu/tgm/table.do?tab=table&init=1&plugin=1&language=en&p
code=tin00067
7. GAIA – mission & vision. (s.f.). Retrieved May 23, 2014, of http://www.gaia.es/english.html
8. Hesse, J. (2013). Seven e-commerce trends to look out for in 2014. Retrieved May 21, 2014,
from http://realbusiness.co.uk/article/24844
9. Informe sobre Medios de Pago y Fraude en Comercio Electrónico 2012. Retrieved May
23, 2014, of http://www.slideshare.net/adigitalorg/informe-medios-pago2012
10. Kalapesi, C., Willersdorf, S., & Zwillenberg, P. (2010). The Connected Kingdom – How the
Internet is Transforming the U.K. Economy. Retrieved May 15, 2014, from
http://www.bcg.com/documents/file62983.pdf
11. Khan, A., & Hunt, J. (2013). 9th annual UK eCommerce Fraud Report. Retrieved May 15, 2014,
from http://cybersource.com/ukfraudreport
12. Moth, D. (2012). Recommendations help drive 27.9% holiday sales growth at John Lewis.
Retrieved May 21, 2014, from https://econsultancy.com/blog/8904
13. Nielsen Holdings. (2012). Press Report: Three-quarters of UK consumers use the internet for
grocery shopping. Retrieved May 15, 2014, from
http://www.nielsen.com/uk/en/insights/press-room/2012/three-quarters-of-uk-consumersuse-the-internet-for-grocery-shop.html
14. Office for National Statistics [ONS]. (2012). E-Commerce and ICT Activity. Retrieved May 15,
2014, from http://www.ons.gov.uk/ons/dcp171778_342569.pdf

PUBLIC
Page 149 of 255
E-COMPASS
15. Office for National Statistics [ONS]. (2013). Internet Access - Households and Individuals, 2013.
Retrieved May 20, 2014, from http://www.ons.gov.uk/ons/dcp171778_322713.pdf
16. Office for National Statistics [ONS]. (2014). Internet Access Quarterly Update, Q1 2014
Release. Retrieved May 15, 2014, from http://www.ons.gov.uk/ons/dcp171778_362910.pdf
17. Page, M. (2012). The Internet Economy in the United Kingdom. Retrieved May 15, 2014, from
http://www.atkearney.com/documents/10192/b4017381-eeb9-4963-b575-8959020de3f1
18. Panorama de la Sociedad de la Información Euskadi 2013. Retrieved May 23, 2014, of
http://www.eustat.es/elementos/ele0011200/ti_Panorama_de_la_Sociedad_de_la_Informaci
on_Euskadi_2013_pdf_962_KB/inf0011206_c.pdf
19. Payvision. (2013). Factsheet 2012 – UK: E-Commerce Payments Landscape. Available via
http://www.payvision.com/infographic-online-shopping-cross-border-ecommerce-uk.
Retrieved May 21, 2014
20. Royal Mail. (2013). Press Release: UK online-only e-retailing has doubled as entrepreneurs
build online success. Retrieved May 21, 2014, fromhttp://www.royalmailgroup.com/ukonline-only-e-retailing-has-doubled-entrepreneurs-build-online-success
21. Seybert, H., Eurostat. (2012). Internet use in households and by individuals in 2012. Statistics
in focus 50/2012. Retrieved May 15, 2014, from
http://epp.eurostat.ec.europa.eu/cache/ITY_OFFPUB/KS-SF-12-050/EN/KS-SF-12-050-EN.PDF
22. The Paypers. (2013). United Kingdom Cross Border E-commerce Country Report - Critical Facts
and Insights for International Expansion. Available via http://www.thepaypers.com/crossborder-ecommerce/cross-border-ecommerce-report-UnitedKingdom/6. Retrieved May 20,
2014
23. Xu, A. J. (2014). 7 E-commerce Trends Small Businesses Need to Know in 2014. Retrieved May
21, 2014, from http://www.huffingtonpost.com/annie-jie-xu/1_b_4557164.html
24. yStats.com. (2012). UK B2C E-Commerce Report 2012, Abstract. Retrieved May 15, 2014, from
http://www.ystats.com/uploads/report_abstracts/970.pdf
25. Zablan, M., Oates, D., Jenkings, R., Bennett, N., Goad, R. (2011). The changing face of UK retail
in today’s multi-channel world. Retrieved May 15, 2014, from
http://www.experian.co.uk/assets/business-strategies/white-papers/RWC-whitepaper2.pdf
26. A. Rovira, A. Aznar, S. Esteban, C. Hernández, B. Martín, G. Valor, and I. Angulo. (2013).
Informe Anual de la Distribución Comercial Minorista. Comunitat Valenciana. Online available:
http://www.portaldelcomerciante.com/miafic/userfiles/Biblioteca/029e843cdf562afe3f35Info
rmeAnualDistribucionComercial_2013.pdf
27. Econsultancy.com. Ltd (2013). Internet Statistics Compendium. Econsultancy London (UK).
Available in URL https://econsultancy.com/reports/internet-statistics-compendium

PUBLIC
Page 150 of 255
E-COMPASS
8
8.1
Appendix
Supporting Material
8.1.1 Online questionnaire for Greek e-commerce SMEs
Live Form URL available:
https://docs.google.com/forms/d/1X94no-ukFjdC4ETPgjmTWB4SdFNGHskQi1VkncFphXU/viewform
D2.1 SME e-COMPASS. Questionnaire for Greek e-commerce SMEs V1.3
* Required
Χώρα *
Όνομα Επιχείρησης - Ιστοσελίδα *
Θέση Εκπροσώπου Επιχείρησης
1) Ποιούς τύπους προϊόντων και υπηρεσιών το ηλεκτρονικό σας κατάστημα προσφέρει; *
α) προϊόντα λιανικού εμπορίου ευρείας κατανάλωσης, τα οποία εύκολα εντοπίζονται και
συγκρίνονται μέσω των κωδικών τους (πχ bar code, GTIN, κτλ)
β) αποκλειστικά προϊόντα, υψηλής ποιότητας, μη ευρείας κατανάλωσης, με δυσκολία
εντοπισμού τους μέσω κωδικών (πχ bar code, GTIN, κτλ)
γ) προϊόντα που διαμορφώνονται ή συναρμολογούνται από τον πελάτη
δ) εξατομικευμένα προϊόντα και υπηρεσίες
ε) Άλλο:
2) Επιλέξτε σε ποιους από τους παρακάτω κλάδους ανήκει το ηλεκτρονικό σας κατάστημα: *
α) Λιανικό Εμπόριο (Business to Consumer -B2C )
β) Χονδρικό Εμπόριο (Business to Business - B2B)
γ) Παροχή Υπηρεσιών
δ) Πώληση Λογισμικού
ε) Άλλος:
Αν επιλέξατε Λιανικό ή Χονδρικό Εμπόριο, καταγράψτε τα είδη προϊόντων που εμπορεύεστε:
Αν επιλέξατε Παροχή Υπηρεσιών, καταγράψτε τα είδη των υπηρεσιών που παρέχετε:

PUBLIC
Page 151 of 255
E-COMPASS
Αν επιλέξατε Πώληση Λογισμικού, καταγράψτε τα είδη λογισμικού που εμπορεύεστε:
3) Πόσους εργαζόμενους πλήρους απασχόλησης διαθέτετε για το ηλεκτρονικό σας κατάστημα;
α) <5
β) 5-10
γ) 11-20
δ) 21-50
ε) >50
4) Ποιο είναι κατά μέσο όρο το ύψος των ετήσιων πωλήσεών σας (τζίρος) μέσω ηλεκτρονικού εμπορίου;
(Κ=1.000 πωλήσεις)
α) < 10K €
β) 11Κ-50€
γ) 51K-100K €
δ) 101Κ-500Κ €
ε) >501Κ €
5) Ποιό ήταν το ύψος των ηλεκτρονικών παραγγελιών που λάβατε το 2013; *
(συμπεριλάβετε και τις παραγγελίες που για οποιαδήποτε λόγο δεν ολοκληρώθηκαν)
α) < 100
β) 101-1000
γ) 1.001-5.000
δ) 5.001-10.000
ε) 10.001-50.000
ζ) >50.001
6) Πόσα έτη δραστηριοποιείται η επιχείρησή σας στο ηλεκτρονικό εμπόριο;
α) <1 έτος
β) 1 – 2 έτη

PUBLIC
Page 152 of 255
E-COMPASS
γ) 3 – 4 έτη
δ) 5 – 10 έτη
ε) >11 έτη
7) Με ποιους τρόπους συγκρίνετε τις τιμές των προϊόντων σας με τους ανταγωνιστές; *
α) μέσω μηχανών αναζήτησης για προϊόντα / υπηρεσίες
β) μέσω ηλεκτρονικών υπεραγορών (eMarketplaces)
γ) στα ηλεκτρονικά καταστήματα των ανταγωνιστών
δ) με άλλο τρόπο:
8) Πόσο συχνά χρειάζεται να πραγματοποιείτε σύγκριση τιμών με ανταγωνιστικά προϊόντα;*
α) σε πραγματικό χρόνο (real-time)
β) κάθε 5΄ – 15΄
γ) κάθε 16΄ – 30΄
δ) κάθε 31΄ – 60΄ λεπτά
ε) κάθε 1 – 6 ώρες
στ) κάθε 6 – 12 ώρες
ζ) ημερησίως
η) κάθε δεύτερη μέρα
θ) εβδομαδιαίως
ι) κάθε δύο εβδομάδες
ια) μηνιαίως
9) Πώς εισάγεται τις τιμές ηλεκτρονικά; *
α) χειρονακτικά, με μη αυτόματο τρόπο
β) αυτοματοποιημένα, σε μη πραγματικό χρόνο
γ) αυτοματοποιημένα σε πραγματικό χρόνο
δ) με άλλο τρόπο:
10) Θα σας ενδιέφερε μία υπηρεσία η οποία θα σύγκρινε τις τιμές σας με αυτές των ανταγωνιστών σας
και θα σας υποστήριζε στη διαμόρφωση των τιμών με χειρονακτικό ή αυτοματοποιημένο τρόπο; *
α) Όχι
β) Ναι
γ) Ναι, εφόσον η υπηρεσία πληροί τις παρακάτω προϋποθέσεις:

PUBLIC
Page 153 of 255
E-COMPASS
11) Χρησιμοποιείτε κάποια υπηρεσία/λογισμικό ανάλυσης/πρόβλεψης της συμπεριφοράς του πελάτη ή
ειδοποίησης εξόδου από το ηλεκτρονικό σας κατάστημα; *
α) Όχι
β) Ναι, όμως μη-αυτοματοποιημένα
γ) Αυτοματοποιημένα, σε μη πραγματικό χρόνο
δ) Αυτοματοποιημένα, σε πραγματικό χρόνο
12) Θα σας ενδιέφερε μία υπηρεσία που αναλύει τη συμπεριφορά του πελάτη σας, σάς παρέχει
συμβουλές για τη βελτίωση των πωλήσεων του καταστήματός σας και υποστηρίζει τις σταυροειδείς
πωλήσεις (cross selling);
α) Nαι
β) Όχι
γ) Ναι, εφόσον η υπηρεσία καλύπτει τον εξής όρο - ανάγκη:
13) Θα σας ενδιέφερε μία υπηρεσία που αναλύει τη συμπεριφορά ομάδων πελατών σας, ανακαλύπτει τα
ενδιαφέροντά τους και εντοπίζει νέες τάσεις στις ηλεκτρονικές αγορές των καταναλωτών; *
α) Όχι
β) Ναι
γ) Ναι, υπό προϋποθέσεις:
Εάν επιλέξατε το (γ), αναλύστε τις απαιτήσεις σας:
14) Οι ηλεκτρονικές απάτες μέσω πιστωτικών καρτών έχουν επηρεάσει τις αποφάσεις σας αναφορικά
με: *
α) την πώληση προϊόντων ή παροχή υπηρεσιών μέσω διαδικτύου

PUBLIC
Page 154 of 255
E-COMPASS
β) την επέκταση του ηλεκτρονικού καταστήματός σας σε άλλες αγορές ευρωπαϊκών χωρών
εκτός από την Ελλάδα
γ) τη χρήση διαδικτυακών συστημάτων πληρωμής ή ηλεκτρονικών συναλλαγών μέσω
πιστωτικών καρτών
15) Ποίο είναι το σύνηθες ετήσιο ποσοστό περιπτώσεων ηλεκτρονικής απάτης ως προς το συνολικό όγκο
πωλήσεών σας; *
α) < 0.1 %
β) 0.2 % – 1%
γ) 1.1% – 3%
δ) 3.1% – 5%
ε) >5.1%
16) Πώς αντιμετωπίζετε τον κίνδυνο ηλεκτρονικής απάτης; *
α) Δε λαμβάνω μέτρα για αυτόν
β) Ελέγχω μία-μία τις παραγγελίες που λαμβάνω
γ) Διαθέτω ένα αυτοματοποιημένο σύστημα εντοπισμού απάτης που λειτουργεί σε μηπραγματικό χρόνο
δ) Διαθέτω ένα αυτοματοποιημένο σύστημα εντοπισμού απάτης που λειτουργεί σε
πραγματικό χρόνο
ε) Αναθέτοντας το καθήκον αυτό σε κάποιον τρίτο (εταιρεία, φορέα)
στ) Συνδυασμός (β) & (γ)
ζ) Συνδυασμός (β), (γ) & (δ)
η) Άλλο:
17) Χρησιμοποιείτε κάποιο συγκεκριμένο λογισμικό – πρόγραμμα για να εντοπίσετε ύποπτες
συναλλαγές;
α) Όχι
β) Ναι
Εάν ΝΑΙ, παρακαλώ γράψτε μας ποιο είναι:
18) Χρησιμοποιείτε σύστημα αξιολόγησης επικινδυνότητας συναλλαγών που έχετε αναπτύξει εσωτερικά
στην εταιρεία σας; *
α) Όχι
β) Ναι
Εάν ΝΑΙ, παρακαλώ περιγράψτε μας με λίγα λόγια πως λειτουργεί:

PUBLIC
Page 155 of 255
E-COMPASS
19) Ποίες παραμέτρους της ηλεκτρονικής συναλλαγής λαμβάνετε υπόψη σας κατά την προσπάθεια
εντοπισμού περιπτώσεων απάτης;
α) Ταυτότητα μηχανήματος συναλλαγής (διεύθυνση IP Η/Υ)
β) Χώρα προέλευσης μηχανήματος (Η/Υ) συναλλαγής
γ) Πόλη προέλευσης μηχανήματος (Η/Υ) συναλλαγής
δ) Χώρα προέλευσης τράπεζας πιστωτικής κάρτας
ε ) Χώρα προέλευσης πελάτη από διεύθυνση
στ) Πόλη προέλευσης πελάτη από διεύθυνση
ζ) Αριθμός τηλεφώνου πελάτη
η) E-mail πελάτη
θ) Συναλλαγή “3D secure” (πληκτρολόγηση μυστικού κωδικού κατά την εισαγωγή των
στοιχείων της κάρτας από τον πελάτη)
ι) Όνομα εκδότριας τράπεζας πιστωτικής κάρτας που πληκτρολογεί ο πελάτης
ια) Ύποπτη ή αντιφατική συμπεριφορά πελάτη
ιβ) Άλλη:
20) Παρακαλώ αναφέρετε τις γλώσσες που υποστηρίζει το ηλεκτρονικό σας κατάστημα:

PUBLIC
Page 156 of 255
E-COMPASS

PUBLIC
Page 157 of 255
E-COMPASS
Figure 128. Data Base gathering of responses from online questionnaire, Greek version
8.1.2 Online questionnaire for Spanish e-commerce SMEs
https://docs.google.com/forms/d/1MV4_hnhfJSScRetkbsD5t3am9OJsMuMQkGxI_W32JXA/viewform
D2.1 E-COMPASS Cuestionario. SPANISH V 1.3
* Required
Nombre de la empresa *
Ubicación de la empresa *
Nombre de la persona de contacto
Teléfono de contacto
E-mail de contacto
¿Qué tipo de productos y servicios ofrece su empresa? *
artículo de consumo, muy comparables a través de un número de producto ( GTIN / EAN , etc )
, de venta rápida
productos exclusivos, de alta calidad , venta lenta, baja comparabilidad basada en un número
de producto ( GTIN / EAN , etc )
productos configurables , construidos mediante la combinación de varios componentes
productos individuales
Other:
Por favor, marque los sectores de comercio electrónico a los que su empresa pertenece: *
a) Business to Consumer (B2C )
b) Business to Business (B2B )
c) Distribución
d) Servicio
e) Software
Other:
Si ha marcado la opción c) indique el tipo de productos:
Si ha marcado la opción d) indique el tipo de servicios:
Si ha marcado la opción e) indique el tipo de software:
¿Cuántos empleados a tiempo completo dedica al comercio electrónico su compañía? *
<5

PUBLIC
Page 158 of 255
E-COMPASS
5-10
11-20
21-50
> 50
Ingresos anuales en compra online *
< 5K €
5K -10K €
11Κ -50 €
51K -100K €
101Κ - 200Κ €
> 200k €
Volumen de pedidos en 2013 (incluyendo las órdenes que no fueron ejecutados por cualquier razón ) *
< 100
101-1000
1001-5000
5001-10000
> 50001
Tiempo que lleva la empresa utilizando un canal de ventas online *
menos de un año
1 - 2 años
3-4 años
5-10 años
> 10 años
¿Cuáles son los principales sitios web que utiliza para comparar precios? *
los motores de búsqueda por Precio / producto
los mercados electrónicos
los sitios web de los competidores
Other:
¿Con qué frecuencia necesita comparar precios? *
en tiempo real
5 a 15 minutos
16 a 30 minutos
31 a 60 minutos
1-6 horas
6-12 horas

PUBLIC
Page 159 of 255
E-COMPASS
todos los días
cada 2 días
semanalmente
quincenalmente
mensulamente
¿Cómo ajusta sus precios? *
manualmente
de forma automática, offline
de forma automática, online
Other:
¿Está interesado en servicios de comparación, ajuste y notificación/alertas de precios? *
Si
No
Sí , si el servicio cumple con los siguientes requisitos:
Especificar requisitos
¿Utiliza actualmente servicios para el análisis del comportamiento del cliente, la predicción y/o prevención
de pérdida de clientes ? *
no
manualmente
de forma automática, offline
de forma automática, online
¿Está interesado en servicios de análisis del comportamiento de los clientes , que apoyen la venta cruzada
y optimicen el rendimiento de la tienda virtual? *
Si
No
Si, si el servicio cumple con los siguientes requisitos:
Especificar requisitos:
¿Está interesado en servicios que analizan el comportamiento de grupos de clientes para conocer hábitos
de consumo y detectar tendencias? *
Si
No
sí , si el servicio cumple con los siguientes requisitos:
Especificar requisitos:
¿Es el fraude una preocupación que le impide: *
la venta de sus productos / servicios en línea

PUBLIC
Page 160 of 255
E-COMPASS
la ampliación de su acceso a los mercados en línea para toda la UE
el uso de transacciones de pago en línea o sistemas de correo de cartas
¿Cuál es la proporción de casos fraudulentos en su volumen total de transacciones ? *
< 0,1 %
0,2 % - 1 %
1,1 % - 3 %
3,1 % - 5 %
>5%
¿Cómo controla el fraude en transacciones online? *
a) revisión manual de las transacciones de entrada
b) herramienta offline automática de evaluación fraude
c) herramienta online automática de evaluación fraude
transfiriendo el riesgo a un tercero
combinación de (a) y ( b )
combinación de (a ) , ( b ) y ( c )
no me ocupo de fraude en los pagos en línea
utilizo un software específico
utilizo su propio software o método de evaluación
Other:
En caso de usar un software específico indicar cuál:
En caso de usar su propio software o método de evaluación indicar cuál:
¿Qué tipo de información específica sobre transacciones tiene en cuenta el experto o software
antifraude? *
identidad de dispositivos
país del dispositivo
ciudad del dispositivo
país del banco
dirección del país
teléfono del país
e-mail
transacciones seguras 3D
nombre del banco proporcionado por el usuario
comportamiento sospechoso o dudoso

PUBLIC
Page 161 of 255
E-COMPASS
Other:
¿Qué idiomas soporta mentalmente su sitio de ventas online? *
Alemán
Búlgaro
Croata
Checo
Danés
Eslovaco
Esloveno
Español
Estonio
Finlandés
Francés
Griego
Holandés
Húngaro
Inglés
Irlandés
Italiano
Letón
Lituano
Maltés
Polaco
Portugués
Rumano
Sueco
Other:

PUBLIC
Page 162 of 255
E-COMPASS
Figure 129. Data Base gathering of responses from online questionnaire, Spanish version

PUBLIC
Page 163 of 255
E-COMPASS
8.1.3 Online questionnaire for British e-commerce SMEs
https://docs.google.com/forms/d/1lU-VT-nk2qrFArHe8ZqgTnCDFFcSNyKzE5VyaUa2d80/viewform
Sequence of versions:
 E-COMPASS. Extended. Survey Questionnaire V1.0
D2.1 E-COMPASS. Extended Survey Questionnaire – English V 3.3
* Required
Country *
Company Name *
1) Which types of products and services do you offer? *
a) commodity products and very often fast selling
b) exclusive products, high quality and very often slow selling
c) configurable products, build by combining several components
d) personalised products
Other:
2) Please tick the most relevant of the e-commerce sectors that your company belongs to: *
a) Business to Consumer (B2C)
b) Business to Business (B2B)
3) Which kind of product do you offer?
Women Clothes
Men Clothes
Shoes
Toys
Furniture
Books
Music CDs
Music and video download
Tickets for events

PUBLIC
Page 164 of 255
E-COMPASS
Hotels
Holiday trips
Last-minute travels
Movies on DVDs and videos
Train tickets
Flight tickets
Computer Hardware and other equipment
Computer software without games
Computer games
Health products and pharmaceuticals
Sport equipment
Other:
4) How many full time employees dedicated to e-commerce does your company employ? *
a) 1
b) 2-3
c) 4-5
d) 6-10
e) 11-20
f) 21-50
g) >50
5) Which is the total 2013 annual revenue from online sales? *
a) < 10K€
b) 10K-50K €
c) 50Κ-1M€
d) 1M-10M €
e) 10M-30M €
f) >30M
6) Which is the total 2013 annual revenue from total (online and offline) sales? *
a) < 10K€
b) 10K-50K €
c) 50Κ-1M€
d) 1M-10M €
e) 10M-30M €
f) >30M

PUBLIC
Page 165 of 255
E-COMPASS
7) Which was the annual volume of online orders received in 2013? (including the orders that were not
executed for any reason) *
a) < 100
b) 101-1000
c) 1.001-5.000
d) 5.001-10.000
e) 10.001-50.000
f) >50.001
8) How long has your company been doing business online? *
a) <1 year
b) 1 – 2 years
c) 3 – 4 years
d) 5 – 10 years
e) >11 years
9) Which languages does your e-shop currently support? *
Bulgarian
Croatian
Czech
Danish
Dutch
English
Estonian
Finnish
French
German
Greek
Hungarian
Irish
Italian
Latvian
Lithuanian
Maltese
Polish
Portuguese

PUBLIC
Page 166 of 255
E-COMPASS
Romanian
Slovak
Slovene
Spanish
Swedish
Other:
10) Which are the main Websites where you compare prices? *
a) Price/product search engines
b) eMarketplaces
c) Websites of competitors
Other:
if select a), Which search engines do you use?
if select b), Which eMarketplaces do you use?
11) How often do you need to compare prices? *
a) Real-time
b) 5 – 15 minutes
c) 16 – 30 minutes
d) 31 – 60 minutes
e) 1 – 6 hours
f) 6 – 12 hours
g) Daily
h) Every 2 days
i) Weekly
j) Bi-weekly

PUBLIC
Page 167 of 255
E-COMPASS
k) Monthly
l) Never
12) How often would you like to compare prices? *
a) Real-time
b) 5 – 15 minutes
c) 16 – 30 minutes
d) 31 – 60 minutes
e) 1 – 6 hours
f) 6 – 12 hours
g) Daily
h) Every 2 days
i) Weekly
j) Bi-weekly
k) Monthly
l) Never
13) How do you adjust your prices? *
a) Manually
b) Automatically, offline
c) Automatically, in real-time
Other:
13a) How many products do you compare regularly at competitors e-shops?
13b) How many products would you like to compare regularly at competitors e-shops?
13c) How many competitors e-shops do you observe?

PUBLIC
Page 168 of 255
E-COMPASS
13d) How many competitors e-shops would you like to observe?
13e) Do you use a service or an electronic tool to observe and compare the prices of competitors?
Selection:
a) Own electronic tool
b) Standard electronic tool
c) Electronic interface to a platform or service provider
d) None
If select b). Which electronic tool do you use?
14) Are you interested in a service which compares your products prices, sends alerts to you when prices
exceed certain price limits, and supports your price adjustments either manually or automatically? *
a) No
b) Yes
c) Yes, if the service fulfils the following requirements:
15) Is fraud a concern which prevents you from: *
a) selling your products/services online?

PUBLIC
Page 169 of 255
E-COMPASS
b) expanding your online market access to the entire EU?
c) using online payment transactions or e-card systems?
16) What is the typical proportion of fraudulent cases in your total volume of transactions? *
a) < 0.1 %
b) 0.2 % – 1%
c) 1.1% – 3%
d) 3.1% – 5%
e) >5.1%
17) How do you deal with online payment fraud? *
a) I do not deal with online payment fraud
b) Manual review of incoming transactions
c) Automatic offline fraud-assessment tool
d) Automatic real-time fraud-assessment tool
e) By transferring risk to a third party (i.e.“I do NOT explicitly deal with it“)
f) Combination of (b) & (c)
g) Combination of (b), (c) & (d)
Other:
18) Do you use a specific software to deal with online payment fraud? *
a) No
b) Yes
If YES please, which one is it?
19) Do you use your own software or assessment method to deal with online payment fraud?*
a) No
b) Yes
If YES please, which one is it?

PUBLIC
Page 170 of 255
E-COMPASS
20) What type of transaction-specific information does your antifraud personnel or antifraud tool take into
account? *
a) Device Identity
b) Device Country
c) Device City
d) Bank Country
e) Address Country
f) Address city
g) Telephone country
h) e-mail address
i) 3D secured transaction
j) Bank name provided by the user
k) Suspicious or dubious behavior
Other:
21) What do you monitor, plan to monitor or do not monitor on your Website/e-shop?: Information of the
visitor's origin: *
do monitor
plan to monitor
do not monitor
Websites which refer the
visitors to the e-shop
Common Key words which
are used in search engines by
the visitors of an e-shop
Most common entry pages
Common search phrases
which are used in search
engines by the visitors of an
e-shop
Geographical origin of the
visitors (e.g. country, region,
town)
Websites which refer the visitors to the e-shop
I wish to have that information
I wish to anonymously receive that information from comparable e-shops
Common Key words which are used in search engines by the visitors of an e-shop
Most common entry pages
Common search phrases which are used in search engines by the visitors of an e-shop

PUBLIC
Page 171 of 255
E-COMPASS
Geographical origin of the visitors (e.g. country, region, town)
Other information of the visitors's origin
22) What do you monitor, plan to monitor or do not monitor on your Website/e-shop?: Information of
visitors's attributes *
do monitor
plan to monitor
do not monitor
Number of visitors per week
Number of new visitors
Number of frequent visitors
Technical equipment of the
visitors (e.g. browser-version)
Number of visits per visitor
Number of visitors per week
Number of new visitors
Number of frequent visitors
Technical equipment of the visitors (e.g. browser-version)
Number of visits per visitor
Other information of visitors' attributes

PUBLIC
Page 172 of 255
E-COMPASS
visitors's behaviour: *
do monitor
plan to monitor
do not monitor
Pages which are most often
viewed
Number of page views per
visit (depth of visits)
Time of stay per visit
(duration of visits)
Applied key words within the
own e-shop search
Most common exit pages
Most common page view
sequence (click paths)
Pages which are most often viewed
Number of page views per visit (depth of visits)
Time of stay per visit (duration of visits)
Applied key words within the own e-shop search
Most common exit pages
Most common page view sequence (click paths)
Other information of visitors' behaviour

PUBLIC
Page 173 of 255
E-COMPASS
purchasing behaviour *
do monitor
plan to monitor
do not monitor
Number of visitors who did a
purchase
Average value of a shopping
cart of the e-shop
Number of visitors who put a
product into the basket
Number of visitors who break
up the check out (purchasing)
process
Average time of stay in the eshop until purchasing
products
Number of visitors who did a purchase
Average value of a shopping cart of the e-shop
Number of visitors who put a product into the basket
Number of visitors who break up the check out (purchasing) process
Average time of stay in the e-shop until purchasing products
Other information of purchasing behaviour

PUBLIC
Page 174 of 255
E-COMPASS
25) What do you monitor, plan to monitor or do not monitor on your Website/e-shop?: New function for
monitoring *
do monitor
plan to monitor
do not monitor
Analysis of the access of
mobile devices
Categorization of user groups
(visitors segmentation)
Qualitative user survey
Page-oriented user feedback
Form field analysis
Comparative tests (e.g. A/Btests)
Mouse-tracking
Analysis of the access of mobile devices
Categorization of user groups (visitors segmentation)
Qualitative user survey
Page-oriented user feedback
Form field analysis
Comparative tests (e.g. A/B-tests)
Mouse-tracking
Other information of new function for monitoring

PUBLIC
Page 175 of 255
E-COMPASS
26) How much do you invest in your web activities?
a) < 10K€
b) 10K€ - 50K€
c) 50K€ - 500K€
d) 500K€ - 1M€
e) > 1M€
Please, send me the results of the survey
OK
I am interested in participating as a pilot user in E-COMPASS. I am aware that not every e-Shop can be
considered within the project.
OK
Additional Information
In case you are interested in the results and/or participation of the project E-COMPASS, we need the
following information
Your eMail address:
Your telephone number (optional):
URL of the e-Shop (optional):

PUBLIC
Page 176 of 255
E-COMPASS
Figure 130. Data Base gathering of responses from online questionnaire, extended English version

PUBLIC
Page 177 of 255
E-COMPASS
8.1.4 Focus Group questionnaire guide for fraud in e-commerce

PUBLIC
Page 178 of 255
E-COMPASS

PUBLIC
Page 179 of 255
E-COMPASS

PUBLIC
Page 180 of 255
E-COMPASS

PUBLIC
Page 181 of 255
E-COMPASS

PUBLIC
Page 182 of 255
E-COMPASS

PUBLIC
Page 183 of 255
E-COMPASS

PUBLIC
Page 184 of 255
E-COMPASS

PUBLIC
Page 185 of 255
E-COMPASS

PUBLIC
Page 186 of 255
E-COMPASS
8.1.5 Presentation slides: Corinth Infoday Project Presentation

PUBLIC
Page 187 of 255
E-COMPASS

PUBLIC
Page 188 of 255
E-COMPASS

PUBLIC
Page 189 of 255
E-COMPASS

PUBLIC
Page 190 of 255
E-COMPASS

PUBLIC
Page 191 of 255
E-COMPASS

PUBLIC
Page 192 of 255
E-COMPASS
8.1.6 Presentation slides: Corinth Infoday Press Release

PUBLIC
Page 193 of 255
E-COMPASS

PUBLIC
Page 194 of 255
E-COMPASS
8.1.7
Presentation slides: Corinth Infoday Express of Interest Form for SMEs

PUBLIC
Page 195 of 255
E-COMPASS
8.1.8
Presentation slides: Corinth Infoday Anti-Fraud Technologies Presentation

PUBLIC
Page 196 of 255
E-COMPASS

PUBLIC
Page 197 of 255
E-COMPASS

PUBLIC
Page 198 of 255
E-COMPASS

PUBLIC
Page 199 of 255
E-COMPASS

PUBLIC
Page 200 of 255
E-COMPASS

PUBLIC
Page 201 of 255
E-COMPASS

PUBLIC
Page 202 of 255
E-COMPASS

PUBLIC
Page 203 of 255
E-COMPASS

PUBLIC
Page 204 of 255
E-COMPASS

PUBLIC
Page 205 of 255
E-COMPASS

PUBLIC
Page 206 of 255
E-COMPASS
8.1.9 Presentation slides: Corinth Infoday Press Release

PUBLIC
Page 207 of 255
E-COMPASS

PUBLIC
Page 208 of 255
E-COMPASS

PUBLIC
Page 209 of 255
E-COMPASS

PUBLIC
Page 210 of 255
E-COMPASS

PUBLIC
Page 211 of 255
E-COMPASS

PUBLIC
Page 212 of 255
E-COMPASS

PUBLIC
Page 213 of 255
E-COMPASS
8.1.10 Presentation slides: Halton Infoday/Interviews Online Data Mining Info Collection

PUBLIC
Page 214 of 255
E-COMPASS

PUBLIC
Page 215 of 255
E-COMPASS

PUBLIC
Page 216 of 255
E-COMPASS

PUBLIC
Page 217 of 255
E-COMPASS

PUBLIC
Page 218 of 255
E-COMPASS

PUBLIC
Page 219 of 255
E-COMPASS
8.1.11 Extended questionnaire: GAIA - CIC Infoday/Interviews Info Collection

PUBLIC
Page 220 of 255
E-COMPASS

PUBLIC
Page 221 of 255
E-COMPASS

PUBLIC
Page 222 of 255
E-COMPASS

PUBLIC
Page 223 of 255
E-COMPASS

PUBLIC
Page 224 of 255
E-COMPASS

PUBLIC
Page 225 of 255
E-COMPASS

PUBLIC
Page 226 of 255
E-COMPASS

PUBLIC
Page 227 of 255
E-COMPASS

PUBLIC
Page 228 of 255
E-COMPASS

PUBLIC
Page 229 of 255
E-COMPASS

PUBLIC
Page 230 of 255
E-COMPASS

PUBLIC
Page 231 of 255
E-COMPASS

PUBLIC
Page 232 of 255
E-COMPASS

PUBLIC
Page 233 of 255
E-COMPASS

PUBLIC
Page 234 of 255
E-COMPASS

PUBLIC
Page 235 of 255
E-COMPASS
8.1.12 Data Base scheme: Core version of products diagram (CIC)

PUBLIC
Page 236 of 255
E-COMPASS
8.1.13 Presentation slides: ATEVAL Infoday/Interviews Online Data Mining Info Collection

PUBLIC
Page 237 of 255
E-COMPASS

PUBLIC
Page 238 of 255
E-COMPASS

PUBLIC
Page 239 of 255
E-COMPASS

PUBLIC
Page 240 of 255
E-COMPASS

PUBLIC
Page 241 of 255
E-COMPASS

PUBLIC
Page 242 of 255
E-COMPASS

PUBLIC
Page 243 of 255
E-COMPASS
8.1.14 Presentation slides: ATEVAL Infoday/Interviews Anti-fraud Application Info Collection

PUBLIC
Page 244 of 255
E-COMPASS

PUBLIC
Page 245 of 255
E-COMPASS

PUBLIC
Page 246 of 255
E-COMPASS

PUBLIC
Page 247 of 255
E-COMPASS

PUBLIC
Page 248 of 255
E-COMPASS

PUBLIC
Page 249 of 255
E-COMPASS

PUBLIC
Page 250 of 255
E-COMPASS

PUBLIC
Page 251 of 255
E-COMPASS

PUBLIC
Page 252 of 255
E-COMPASS

PUBLIC
Page 253 of 255
E-COMPASS

PUBLIC
Page 254 of 255
E-COMPASS

PUBLIC
Page 255 of 255

Report Template - SME E

Transcription

Similar documents

business performance management in e-commerce

TEXT ANALYTICS

This website was created for Bachendorf`s, a designer jewelry

Grand Opening poster - the Town of Halton Hills

Public overview CP - ANIMA Investment Network

openness and learning analytics - Universidad del Sagrado Corazón

presentation

July 25, 2013 – FRAUD ALERT-Harbor Freight Tools LSC Card

PDF

July 31, 2013 – FRAUD ALERT-Harbor Freight Tools Update