Museum Analytics

Transcription

Museum Analytics

Version: 1.0
Date: Jan 26th, 2014
Rui Guerra and
Fransje Pansters
INTK
Museum Analytics
Action Research Project (ARP)
Suppor ted by:
INTK
Neude 5
3512AD Utrecht
The Netherlands
www.intk.com
Rui Guerra
T +31613719204
[email protected]
TABLE OF CONTENTS
1.
Introduction
.................................................................................................................................................................................................................3
2.
Project Description
....................................................................................................................................................................................................4
3.
Online Analytics Maturity
.......................................................................................................................................................................................13
4.
Website Analytics
......................................................................................................................................................................................................20
5.
Social Media Analytics
..............................................................................................................................................................................................29
6.
Participants Feedback
...............................................................................................................................................................................................35
7.
Conclusion
.................................................................................................................................................................................................................38
INTK
Neude 5
3512AD Utrecht
The Netherlands
www.intk.com
|
Rui Guerra
T +31613719204
[email protected]
1. INTRODUCTION
Museums and cultural organization have a public duty to present their stories and collections to a wide public. It is a fact that there
is a growing online audience potentially interesting for museums. Not surprising, cultural organizations have been investing a considerable amount of resources on online infrastructures, such as, institutional websites, e-newsletters and social media. One of the
most pressing challenges that organizations face in regard to these technologies is how to define, measure and evaluate their
effectiveness. The 'Museum Analytics - Action Research Project' tackles these challenges by installing a truly analytic culture in museums, by empowering museum professionals with expertise and tools to measure the impact of their online endeavors. Moreover,
it seeks to help gain insights that lead to a deeper understanding of online audiences. The project follows the footsteps of a similar
project developed by Culture24 in the UK started in 2011 and still continuing today.
The Museum Analytics - Action Research Project took place over a period of approximately 10 months and it included three workshops, two online evaluation surveys and one evaluation meeting. Several museums professionals representing 16 Dutch museums
participated in all activities. The followed approach required the active participation of all involved organizations to share their experiences in developing and understanding the impact of online services with the wider project group. The present report is divided into several chapters following the same order of which different topics were discussed in the several
workshops. The chapter 'Project Description' presents a detailed account on how the project was developed. It includes the project
goals and methodologies, as well as some background information on how the project was initiated. This chapter has been written
with the intention to help others that might be interested in launching a similar project. For readers that are focused on how implementing the outcomes of the project in their own organization this chapter can be skipped. The chapter 'Online Analytics Maturity' introduces a model developed by analytics expert Stéphane Hamel and explains how it can be used by museum professionals
to assess their current situation in terms of online analytics. Furthermore, it provides an actionable path towards improving competences and leveraging data analyze and decision-making. This chapter includes participants' findings from using the model at the beginning and at the end of the project. The focus of the chapter 'Website Analytics' is to understand whether the participating museums were achieving the goals proposed for their museum/individual institutional websites. The chapter includes the outcome of an
online survey conducted to investigate the motivation of online visitors; instruction on how museums can use Google Analytics to
measure the performance of their websites and a benchmark of the participants' institutional websites.
The participants’ social media presence is analyzed in the chapter 'Social Media Analytics'. The chapter includes a social media benchmark of all participants as well as social media publishing best practices based on the findings of participants as well as
based on information supplied by Facebook and Twitter. The chapter also includes practical information on how museum professionals can use social media to increase website visits and museum attendance. A chapter is included with the participants' feedback
about the project. The final chapter includes the projects conclusions and list of future work.
The present report aims at inspiring others to develop similar projects in the cultural sector or any other professional sector to
which the learnings of the project might be useful. In order to achieve that goal, the report has been intentionally made as detailed
as possible both about the methodology used to implement the project, the tools used and its outcomes. Finally, this report is released into the Public Domain. To the extent possible under law, the authors waive all copyright and related or neighboring rights to
this report.
Museum Analytics - Introduction
Page 3 of 39
2.
PROJECT DESCRIPTION
2.1.
PROJECT AIM
The Museum Analytics Action Research Project (ARPNL) aims to install a truly analytic culture in the Dutch museum community, by
empowering museum professionals with the expertise and tools in order to understand online audiences and effectively measure
the impact of online endeavors. This report is the first step towards answering the pressing challenges that museums face in regard
to online infrastructures, which is: how to define, measure and evaluate the effectiveness of online activities?
2.1.1.
Project Focus
The focus of the project has been defined based on the assumption that online infrastructures are developed in order to help
organizations to fulfill their mission. The relation between the online endeavors and the organization’s mission is usually outlined in a
document called “online strategy”. For many organizations, this document is the point of departure when it comes to reporting on
their online strategy, as it often includes metrics to measure the effectiveness of such activities. Once metrics have been defined, it is
possible to use free software to setup periodic reports also known as digital dashboards. By continuously analyzing the outcomes of
online activities, organizations can adjust their online tactics in accordance with their overall mission. The research focus was to:
•
Map overall organizational goals to online activities by creating an online strategy;
•
Setup and use free software to measure and evaluate online activities and online audiences;
•
Analyze and interpret weekly and monthly reports about online activities and online audiences;
•
Understand how to integrate the use of online tactics into the organization overall strategic mission.
The focus of the project was defined in a project-outline before the project started. It was presented and agreed on by the
participants in a project launch meeting.
2.1.2.
Project Scope
The specific scope of the project was informed by the input and ideas of the project group. The following topics reflected the
participants’ interests and defined the project scope:
•
Online analytics maturity (including an analysis of goals, metrics, tools, etc.)
•
Website analytics
•
Social media analytics
These priorities set the agenda for several face-to-face meetings and workshops that in turn formed the structure of the research
project. Some of the interests expressed by the participants were:
•
Time to reflect on wider organizational goals and the role that online activities can play in fulfilling these (or not).
•
Understanding online user motivation: Why do people visit museum websites?
•
Learn how to set up Google Analytics in accordance with the sector’s current best practices.
•
Deeper insights into the use of Google Analytics for web reporting.
•
Gaining insights about the current use of mobile devices to visit museum websites.
•
Improve internal reporting in order to engage staff members with the organization online activities.
•
Being part of a professional network that actively shares knowledge and informs each other about practices.
Museum Analytics - Project Description
Page 4 of 39
While the project was still in premature state, several participants mentioned topics that, although relevant, unfortunately did not fit
the scope of the current research project. Some of the topics that came up were e-newsletters and e-commerce. In respect to enewsletters, participants were interested in sharing information about specific metrics and tools used by other organizations. The
discussion about e-commerce focused on existing business models and frameworks used to sell tickets online. Although the
participants had the opportunity to discuss these and various other topics, the research project could not include them in its main
focus. It was agreed by the project group that such topics are relevant and interesting to be addressed in a possible follow-up
research project.
2.2.
METHODOLOGY
The project methodology has been based on the interests of participants, in order to achieve the active sharing of knowledge. We
agreed that the project would be framed around a collaborative approach with several face-to-face meetings, in which participants
would take an active role. Between those meetings, an online collaboration tool was used to continue the active exchange of
information. The project consisted of two meetings, three workshops and two evaluation surveys over a period of eight months. The
workshops were the core of the research project. In order to get the best of the project, it was vital that participants attended all of
the meetings and workshops. Different tools were used to define, measure and evaluate the participants’ online endeavors. An
evaluation survey was used to collect participants’ feedback on the overall project. All participants contributed to the project’s
funding with a fee. Below, the project methodology is further described.
2.2.1.
Collaborative Approach
The project was developed in a participatory and collaborative style, together with sixteen Dutch museums. Each participating
museum put forward a member of their staff to act as the liaison point for the research. These individuals were directly involved in
the delivery of the organizations digital offer, either at an implementation or strategic level. Furthermore, they were responsible for
any other digital staff, freelancers or technical contractors, which allowed them to access specific data and to apply the project’s
findings immediately. The collaborative project approach demanded the active involvement of each participating organization to
share their experiences in developing and understanding the impact of online endeavors with the project group. In practice, this
approach meant an investment to being open and honest about online successes and failures, as well as the sharing of internal
analytical processes and cooperation with the INTK team to share analytics data from several platforms.
The collaborative and active project approach was achieved by requesting each participant to attend all meetings well prepared.
Participants were instructed on the necessary preparation well on time. Most preparation assignments were covered over 80% by
participants. These assignments stimulated the project group’s motivation and dedication. By way of benchmarking, the results of one
individual participant would always benefit the results of others, as they were put into context. Furthermore, questions that had
arisen during the preparation could directly be discussed during the following meeting or workshop.
2.2.2.
Meetings
In total, the project group gathered for five meetings (project launch, 3 workshops, and evaluation) over a period of nine months,
from February 2013 until October 2013. Three workshop meetings formed the core of the project. Each workshop dealt with one
of the agreed focal points and collaboratively formulated priorities, as we agreed on during the first workshop. During two
workshops, an international expert was invited to share his/her knowledge on museums and the evaluation of online activities and
audiences. More information about these experts can be found in the appendix.
All meetings and workshops have been held in the Netherlands and were hosted by one of the participating museums. More
museums were willing to host a meeting or workshop than there were opportunities to do so. The final meeting locations were
chosen according to geographical location. As a result, the project group gathered in Amsterdam, Otterlo, Den Haag and Utrecht.
Most participating museums were located in the Randstad area (Amsterdam, Rotterdam, Den Haag and Utrecht). By spreading the
meetings over several cities, it was aimed to encourage participants to visit colleague institutions outside of their own city and to
possibly divide traveling time among the group.
2.2.3.
Guided Museum Tours
All meetings and workshops were hosted by one of the participating museums. The hosting institutions offered a guided tour as part
of the workshop program. These tours differed in character and style, as some described the extraordinary museum architecture,
while others exemplified the unique aspects of the presentation of their collection. The tours gave a special touch to each workshop
or meeting, as they were instructive and enjoyable. Not only did we get to know each other’s working environment, it was also very
Page 5 of 39
interesting and inspiring to compare different collection presentation styles and approaches. The last meeting was not hosted at a
museum, but at the Dutch Game Garden (where the INTK office is located). The Dutch Game Garden is a non-profit organization
that supports starting game developers and other creative companies. In order to take advantage of the technological creative
environment at the Dutch Game Garden, the tour was replaced by several short presentations by startup companies.
2.2.4.
Online Collaboration
In between each meeting or workshop, almost two months passed. Therefore, an online collaboration tool called Basecamp was set
up in order to ensure a continuous flow of sharing knowledge, information and experiences among the participants. Most topics
discussed on Basecamp were related to one of the workshops, but also issues that fell outside the scope of this research project
were shared. Basecamp was thus used as the central platform for communication. As the project further developed and participants
got to know each other better, the online platform was used more intensively.
Activity on Basecamp can be summarized as follows:
•
49 people were registered;
•
37 files were shared;
•
46 discussions were started;
•
176 comments were placed.
Thanks to the online platform, communication among participants remained open and transparent to the entire group. Files and
other information that was shared in the platform, made it a real knowledge hub for all the participants. All in all the online platform
truly encouraged the collaboration among participants, as demonstrated by the open and honest sharing of information and
experiences among the group.
2.3.
ACTIVITIES OVERVIEW
From February to September 2013, the following activities took place: a project launch meeting, three workshops and an evaluation
meeting. The workshops formed the core of the research project, as each workshop covered one of the topics defined in the
project scope. In summary, the first workshop dealt with Online Analytics Maturity (including the analysis of goals, metrics and
tools), the second with Website Analytics and the third with Social Media Analytics.
2.3.1.
Project Launch Meeting
The Project Launch Meeting was hosted by the EYE
Filmmuseum in Amsterdam on February 6th, 2013.
The meeting brought together more than twenty
museums that had shown interest in participating in
the research project. Hein Wils (at that time working
as project manager at the Netherlands Institute for
Sound and Vision) explained the background of
ARPNL, while Fransje Pansters (project manager,
INTK), described the ARPNL methodology. This
meeting offered an opportunity for interested
participants to clarify any possible doubts about the
project and it’s methodology. After the project launch
meeting, all attendants were asked to give a final
answer about their commitment to the project.
From the twenty-seven museums that attended the
project launch meeting, sixteen decided to
participate. An overview of all museums invited to
participate can be found in the appendix. The most
important reasons against participating were related
to a lack of time to contribute actively to the
project, a tight museum staff capacity and/or financial
issues regarding the participation fee.
EYE Filmmuseum, Amsterdam. Photo by Maarten Noordijk.
Page 6 of 39
2.3.2.
Workshop 1: Online Analytics Maturity
The first workshop was hosted by Science
Centre NEMO in Amsterdam, on March
15, 2013. The workshop was divided into
two parts. In the first part of the
workshop, Jane Finnis (Chief Executive of
Culture24) presented the key findings
from the Let's Get Real project. She talked
about how we understand online success
and how we could relate it to the overall
organizational mission.
of museums.
Furthermore, the scope and research
priorities of the research project were
further refined.
During the second part of the workshop,
facilitated by Rui Guerra (co-founder of
INTK), participants discussed their
organizational strengths and weaknesses in
online analytics, following the online
analytics maturity model defined by
Group picture at the rooftop of the Science Centre NEMO.
Stéphane Hamel. By the end of the
workshop, each participant had identified
their main points of improvement, that were tackled in subsequent workshops. More information about the online analytics maturity
model and it’s outcomes can be found in the chapter Online Analytics Maturity.
2.3.3.
Workshop 2: Website Analytics
The second workshop was hosted by the
Kröller-Müller Museum in Otterlo, on May
15, 2013. Sebastian Chan (Director of
Digital & Emerging Media, Smithsonian
Cooper-Hewitt National Design Museum,
New York) was leading the workshop. He
expounded on how analytic reports have
changed and are changing many of the
cultural organizations he has worked with
over the past years. Together with
Sebastian, Google Analytics (GA) was
introduced and discussed. Special attention
was paid to how GA can be used to keep
track of and report about museum goals.
While the workshop was highly
informative, it was not easy to deliver
information tailored to the needs of each
and every one of the participants. Some
Group picture at the Kröller-Müller Museum.
would have liked more specific and in
depth information, while others preferred
more basic guidance. This devision was to the result of a great difference in expertise. To concede to the different needs of all
participants, a blogpost was used to share in depth information about Google Analytics after the workshop. In some specific cases,
participants received individual guidance on how to install and configure their Google Analytics account. More information about
the lessons learned from Google Analytics with Sebastian Chan can be found in the chapter Website Analytics.
Page 7 of 39
2.3.4.
Workshop 3: Social Media Analytics
The third workshop took place on July 4th,
2013 and was hosted collaboratively by the
Gemeentemuseum and Museon, which are
located next to each other in The Hague.
The workshop dealt with social media
activities and focussed on measuring
performances and evaluating outcomes.
Similar to the first workshop, the
workshop was divided into two parts. In
the first part, Rui Guerra presented and
discussed the current state of social media
among the project group. Results were
contextualized by comparing them with
the overall social media usage in the
Netherlands. Rui described best practices
to publish content on social networks
based on statistics collected from Museum
Analytics. The second part of the
workshop focussed on sharing
Gemeentemuseum, The Hague
experiences. Charlotte Bosman, Online
Media Consultant at Van Gogh Museum,
presented her approach to social media. Both presentations were regarded as highly informative and inspirational by the project
group. More information about the social media results can be found in the chapter Social Media Analytics.
From the previous Website Analytics
workshop we learned that participants
were eager to discover more about
custom Google Analytics reports. Such
reports could be used to compose a
dashboard that shows analytic
information, tailored to the needs of a
specific department or in line with of the
organizations goals. Therefore, the second
part of the workshop was also directed at
a practical exercise that taught participants
how to compose a customized Google
Analytics dashboard. Starting point was to
formulate a dashboard goal to determine
what the dashboard should report about. A
dashboard can be composed out of many
Participant composing a dashboard by cutting and pasting widgets.
different metrics (widgets). Those have to
be chosen carefully, preferably according a goal, because only 12 widgets can be inserted in a dashboard. The 17 most common
widgets reporting on topics such as visits, traffic sources and most popular content, were selected to use during the assignment.
These widgets were printed on paper, as cutting and pasting with paper made it easy to select, shift them around and to keep an
overview. In the end, three dashboards were created:
•
Social Media Dashboard: focussing on metrics that report on social media activities in relation to the institutional website.
•
Educational Colleague Dashboard: focussing on metrics that report on online activities related to education.
•
Institutional Goals Dashboard: focussing on metrics that report on institutional goals measurable with the website. For
example, a dashboard reporting on e-ticket sales if the organizations goal is to increase e-ticket sales.
More information about the results of the assignment can be found in the chapter Website Analytics and Social Media Analytics.
Page 8 of 39
2.3.5.
Evaluation Meeting
The evaluation meeting took half a day and was
hosted by the Dutch Game Garden (DGG) in
Utrecht, on September 5th, 2013. The meeting
dealt with discussing the progress of the project
group in terms of online analytics, based on the
online analytics maturity model by Stéphane
Hamel. Participants compared the outcome of
their online maturity model before and after the
project took place. This comparison allowed
participants to indicate the impact that the
research project had had on their own
organization, as well as to compare their progress
to each other. The final results are being discussed
in the chapter Online Analytics Maturity.
Furthermore, the evaluation meeting was directed
to evaluating the research project itself, as
participants were asked for their feedback on the
project in general.
Group picture at the Dutch Game Garden, Utrecht.
The DGG serves as a creative incubator for
technological innovation and the development of all kinds of games.Therefore, instead of a museum tour, five creative agencies were
invited to give a brief talk about some of their innovative projects interesting that were of interest to the participating museums.
2.3.6.
Evaluation Surveys
Two evaluation surveys were part of the research project. First, a survey was used to evaluate the progress of participants in terms
of their online analytics. This was referred to as the Online Analytics Maturity Model by Stéphane Hamel. See chapter Online
Analytics Maturity for more information about the results of this survey. Second, the research project itself was evaluated by the set
up of an online survey. This survey collected feedback about the impact that the project had had on the participating organizations
and how the project had changed the museums’ representatives on a personal level. The results of the survey are presented in the
chapter Participants Feedback.
2.4.
TOOLS
Various tools have been used to collect information on website analytics, social media analytics and online audiences. All data was
collected automatically. Google Analytics was used as the main tool for the analysis of web traffic. Museum Analytics was used as the
tool to collect, visualize and share social media data among participants. Qualaroo was used as a tool to shed light on user behavior
and motivation. Furthermore, a blog was created to share some first project results with an interested audience. To measure the
participating organizations’ online analytics maturity, a model developed by Stéphane Hamel was used. As the chapter Online
Analytics Maturity goes deeper into the use and results of this model, it will not be further discussed here.
2.4.1.
Google Analytics
Google Analytics (GA) is a service offered free of charge by Google. It collects and visualizes information about the traffic on a
website. The research project used GA as a main tool for analyzing web traffic among the participants’ websites. GA is one of the
most common tools used for this purpose in the cultural sector and it has been recommended by similar research projects such as
the Let’s Get Real action research project organized by Culture24.
Page 9 of 39
2.4.2.
Museum Analytics
Museum Analytics (MA) is an online platform dedicated
to share and discuss information about museums and
their audiences. It includes daily updated information
about social media usage by museums, as well as
information about online and onsite visits updated on
an annual bases. Currently MA generates daily, weekly
and monthly reports for museums. The platform
includes more than 3200 museums worldwide, of
which 194 are Dutch museums. During the Museums
and the Web conference in 2012, Museum Analytics has
been awarded two prices: People’s Choice and Best
Museum Professional Website.
2.4.3.
Qualaroo
Qualaroo is a tool to run online surveys. During the
research project, Qualaroo was used to get a better
understanding of user motivation, focussing on the
question: why do people visit a museum website? The
results can be found in the chapter Website Analytics.
2.4.4.
ARPNL Blog
Many museum professionals had shown interest in
following the outcomes of the ARPNL. Therefore a blog
was created with the purpose of publishing and
discussing some of the main results the project with a
Museum Analytics
national, as well as an international audience. In line
with the nature of the internet, frequent blogposts
were made. The topics were related to issues discussed during the workshops. All blogposts were first shared on Basecamp, to ask
the participants for their feedback. The ARPNL blog can be found online.
In conclusion, besides the tools that were used by the research project, many other tools exist that could be used to for online
analytics. Participants shared and discussed their favorite tools on Basecamp. Some of these were very useful for a day-to-day
operation, yet the tools applied in this research project have been chosen because they were most suitable to the project’s goals.
2.5.
FUNDING MODEL
The funding model of the research project was characterized by a collaborative approach. The participants covered approximately
half of the projects costs, whereas the other half had been covered with funding form the Mondriaanfonds and the Rijksdienst voor
het Cultureel Erfgoed. All participants paid a €1000 participation fee. We were aware of the different budgets of the participating
museums. Therefore, the participation fee might have been relatively cheap for some museums, whereas it might have been
considerably expensive for others. Few museums had to cancel their intention to participate, due to the fee. Nevertheless, it was
difficult to come up with an approach that was fair to all participants. By means of the ARPNL blog, organizations that could initially
not afford to participate, still have access to the project outcomes and methods. Furthermore, they are encouraged to follow the
guidelines that this report provides.
In conclusion, the project methodology was framed by a collaborative approach in which participants took an active role. All
participating organizations had at least one representative whom attended all meetings. In general, every two months a meeting
took place, over a total period of eight months. The workshop meetings lasted one full day and were hosted by one of the
participating museums. Every workshop was held at another museum. Some workshops were so popular that the participating
organizations requested to attended with more than one representative. To make sure that the workshops stayed at a manageable
size, a maximum of 30 participants was agreed on for each workshop.
Page 10 of 39
Participants’ feedback has played an important role throughout the project, as their interests formed the basis of each workshop.
Whenever possible, workshop programs have been adjusted to newly raised questions or interests. At the end of the project, the
participants’ improvement on online analytics was evaluated, as well as the project itself was evaluated. To collect analytic
information, several tools have been used. The tools have been chosen based on their common usage in the cultural sector and the
degree to which they answered to the project goals. The collaborative approach of the project was reflected in the projects funding,
which was partly covered by the participating museums and partly by the Mondriaanfonds and Rijksdienst Cultureel Erfgoed.
2.6.
BACKGROUND
ARPNL followed the footsteps of a similar project that was carried out by Culture24 in the UK in 2011 and 2012. The outcomes of
the Culture24 Let’s Get Real action research project have been presented in cultural heritage conferences, such as, DISH2011 and
Museums and the Web 2012. The frankness of the Culture24 report was its openness to speak about failure in the sector to really
capture the attention of online audiences. The results of this research have been positively received. Several museum professionals in
the Netherlands responded to this research with great interest, especially because at that time, a research of this nature had not
been carried out yet in the Netherlands. During DISH2011, the possibility was therefore discussed to develop a similar project
involving Dutch museums. ARPNL was initiated by Rui Guerra (co-founder of INTK) in April 2011.
In order to make a benchmark possible, the Dutch research project focused on website analytics and social media analytics, similar
to the Culture24 Let’s Get Real action research project. Furthermore, both projects took a similar approach to collaborative
learning of participants.
2.6.1.
Participants
ARPNL was a joint project from INTK, the Mondriaanfonds, Rijksdienst Cultureel Erfgoed and sixteen participating museums.
INTK’s role was to lead and coordinate the project in close collaboration with the Netherlands Institute for Sound and Vision, who
acted as formal project coordinator.
Participants and sponsor logo’s
Page 11 of 39
2.6.2.
Related Projects In The Netherlands
In the past years, several projects related to the ARPNL subject matter have been initiated in the Netherlands. We were aware of
these projects, such as De Digitale Feiten (The Digital Facts, DDF) and Museumkompas.
De Digitale Feiten research (The Digital Facts, DDF) was carried out in 2008/09 by Digitaal Erfgoed Nederland (DEN). DDF gives an
overview of the state of affairs in the field of digitization of cultural heritage, by creating an overall picture of the size and costs of
the digitization process in the Netherlands. The report provides a good insight into the order and progress of this digitization
process, as well as a statistical framework for measuring it. Different from DDF, the ARPNL focusses on the online presence of
participants and the relation with online audiences. How far organizations are in the digitization process was not covered in the
research project.
Museumkompas was a two-year project implemented by Crossmedialab (Kenniscentrum Communicatie & Journalistiek, Hogeschool
Utrecht), Erfgoed Nederland, DEN and several museums. The aim of the project was to support museum professionals in the
development of robust new crossmedia services by providing the professional with the knowledge and skills to develop and
implement solutions within the context of their own organization. The starting point of Museumkompas is related to ARPNL, but
takes a different approach in practice. By closely following the Culture24 research structure, ARPNL could benchmark results in an
international context.
We have been in contact with Museumkompas to discuss our joint interest and the possibility to join forces. However, at that time,
Museumkompas was already about to develop the Crossmedia Monitor Musea, a platform that collects and presents information
about social media of over 500 museums in the Netherlands. This platform is very similar to Museum Analytics (that was already in
existence then). Unfortunately, both Museumkompas, and ARPNL turned out to be too far ahead in the project’s process, in order
to find common ground for collaboration. By using Museum Analytics as a platform to evaluate social media, ARPNL could make an
international benchmark, as 3000 museums worldwide are part of the platform.
In conclusion, ARPNL differs both from DDF and Museumkompas in its emphasis on collaborative learning, which was first
implemented by Culture24. Without the active participation that was required from all participants, the project could not have been
realized. ARPNL has made its participants aware that many colleagues from different organizations face the same challenges, and that
they can be solved easier by joining forces. ARPNL has started a culture of exchange of knowledge and discussion among museum
professionals, that has the potential to continue after the end of the project.
Page 12 of 39
3.
ONLINE ANALYTICS MATURITY
Museum professionals have recognized the importance of online endeavors to empower their organization to fulfill it’s mission. As
the museums’ online presence becomes a fundamental part of the daily museum activities, it is essential to evaluate and report on
online performances. There are many tools and metrics available to measure the impact of online activities. In general, online
analytics are used to bring knowledge about how a website is used and to study the behavior of online audiences. For example,
online analytics can be used to find out what visitors search for on the website, or which artworks of the online collection are
viewed most.
However, as participants mentioned, it is not always easy to make sense of analytical numbers that report on online performances.
There are so many metrics available that the amount of information can become overwhelming, without a clear focus or plan.
Indeed, metrics are most useful if it is well understood what they should stand for. With a clear focus, online analytics can be used
for sustainable development of the online presence.
Many different models can be used to better understand the value of online analytics. To define the participants’ analytics situation
and approach, this project has used a model developed by analytics expert Stéphane Hamel (2009) This model is called: the Online
Analytics Maturity Model (OAMM). The OAMM was applied two times throughout the project, once at the beginning of the
research project in March 2013 and once towards the end in September 2013, in order to measure and analyze the participants’
improvement in online analytics.
This chapter further describes the OAMM and the approach that was taken to help participants improve on their online analytics
situation. Furthermore, an analysis is made of the OAMM results of March and September 2013. Finally, the OAMM outcomes of the
project are put in an international context.
3.1.
ONLINE ANALYTICS MATURITY MODEL (OAMM)
The OAMM is a framework that helps organizations to
assess their current analytics situation and provides a structured and actionable path towards improving online analytics. Stéphane Hamel proposes six key process areas to take
into account when it comes to being successful online. Each
key process area is graded by maturity levels, scaling from 0
(damaged analytics) to 5 (analytically addicted). The OAMM
was first introduced in the cultural sector by Tate Gallery, as
part of a research project developed by Tijana Tasich and
Elena Villaespesa. A paper with the outcomes of this research, as well as the slides presented at the Museums and
the Web 2012 conference, are available online. As Tate explains in its paper, the OAMM provided a critical framework
to visualize the state of analytics approach. Likewise, for this
research project, the model clearly demarcated and reflected on the analytics areas of the participants, especially
where they were in need of improvement.
To stimulate improvement, the six key process areas of the
OAMM were transformed into clear actions. Also, the
Museum Analytics - Online Analytics Maturity
Online Analytics Maturity Model, clear format.
Page 13 of 39
project’s workshops were tailored to some of the areas. Below, the six key process areas are described, along with the improvement
actions that participants were recommended to achieve.
3.1.1.
Management, Governance And Adoption
The area Management, Governance and Adoption defines the degree to which online analytics are being spread within an
organization. Ideally, members of the management team are accountable for promoting online performance analysis among the
organization, in order to ensure that analytics find more solid ground. It would be best if one management member is responsible
for understanding correlations between metrics, evaluating online performances against the museum’s mission and raising awareness
among the organization about the growing importance of online visitors. This member is called the web-analytics champion.
“Metrics are just numbers and indicators without meaning and significance.”
(Stéphane Hamel)
The issue of spreading analytics throughout the organization was defined as one of the key interests of the ARPNL project.
Therefore, the participating museums were recommended to select a web analytics champion within their organization. Most of the
individuals that had attended the project workshops appointed themselves as web analytics champion. As a possible first step
towards raising awareness about online visitors, the champion was advised to create a simple graph that compared the number of
onsite visitors to the number of onsite visits, over the past couple of years. This would be a first step towards raising awareness
about the online visitors within the organization, as visualizing the online visitors shows their dimension and makes them less
abstract.
3.1.2.
Objectives
Online analytics data are easiest to interpret in comparison with objectives. These can be defined in accordance with the museum’s
mission in an online strategy. By defining objectives, participants were encouraged to look only at the metrics that answered their
questions, instead of trying to find the objectives by looking at all the answers available (not an uncommon approach towards online
analytics).
Few participants had defined an online strategy at the start of the project, as most where in the process of developing one. An
effective online strategy does not necessarily have to be very elaborate, but should include goals that take the organizational mission
as a starting point. Participants were assigned to define a succinct online strategy by creating a one-page document with at least
three goals, referring to their museum’s overall mission. Questions raised to help participants define goals were: what is the mission
of the museum? How can the internet empower the museum staff and the museum audience to fulfill that mission? Also, several
existing museum online strategies and social media strategies were distributed to use as inspiration. Participants’ most common
goals were:
•
Increase online visits and engagement to online content;
•
Increase visits to the museum;
•
Increase e-mail newsletter subscriptions;
•
Increase e-ticket sales.
We made sure that these goals were in accordance with the SMART criteria: Specific, Measurable, Attainable, Relevant and Timesensitive. Goals defined based on the SMART criteria lead to actionable strategies focused on results. As for these goals, they can all
be measured with online analytic tools. Based on the analytic results, clear actions could be defined for improvement. For example:
the effectiveness of a campaign aimed at increasing e-ticket sales could easily be measured by comparing e-ticket sale numbers
before and after the campaign. Instead of an ad-hoc approach towards increasing e-ticket sales, a possible next step could be defined
based on the analytic results. Participants who already developed an online strategy shared it among the group, so feedback could be
given and others could use it as a starting point.
3.1.3.
Scope
The scope defines the areas of the online strategy where the analysis is focused on. It is not unusual for cultural organizations to
have a scattered and undocumented online presence. Therefore it was recommended to start inventorying the current situation for
Page 14 of 39
those participants who had not already done so. Most
participants used a simple spreadsheet to document all the
websites their museum had created, along with the social media
platforms, blogs, apps and other publishing platforms that were
being used online. Some possible fields advised to include in the
spreadsheet were: platform name, date of creation, current
statistic numbers of the specific platform, name of the person
responsible for the platform and the URL. Once the participants’
online presence was mapped, it was advised to choose a couple
of platforms to focus the analysis on. Participants were eager to
share their spreadsheets, so other participants who did not have
one yet, could use their framework as starting point.
Example online presence spreadsheet
3.1.4.
The Analysis Team And Expertise
The area Analytics Team and Expertise describes the degree of analytics knowledge and expertise within an organization. Ideally,
each organization has an online analytics expert, who is to advise colleagues on how to translate the organizations’ mission into
web analytics requirements. Analytic numbers can be hard to understand for people, unless they are being put into context by an
expert. For many people, the stories behind the numbers may be even more valuable than the numbers itself. For example, the
expert could help the curatorial team by creating a report that indicates the most popular webpages of the museum’s online
collection, as well as suggestions on how to improve the less popular pages.
Rather than overwhelming the entire organization with customized weekly or monthly reports, it is more effective if simple action
can be derived from insights that the reports may provide. In order to improve on the Analysis Team and Expertise, participants
were recommended to ask their colleagues what analytical information they could use in their decision making process.
3.1.5.
Continuous Improvement Process And Analysis Methodology
The area Continuous Improvement Process and Analysis Methodology determines the level to which a structured approach to a
continuous learning process is developed. Information from analytics brings knowledge and insights that could initiate new questions
and possibilities. Therefore, it was recommended to define a strategy that ensures continuous learning. Much analytical data about
online exists, that can be useful to test and analyze in the context of each organization. Rather than tackle all possible questions at
once, it was advised to focus on one improvement point at a time.
Participants were recommended to organize a monthly meeting dedicated to a specific goal and to measure it’s progress, in order
to improve it. Take for example the goal of increasing e-ticket sales. During such a monthly meeting, the museum staff can discuss
how many people actually buy tickets online? Can people find the page on the website easily? Can it be made easier for visitors to
buy tickets online? Do online ticket sales benefit the queuing time at the museum, or the workload for colleagues at the cash desk?
The answers to these questions subsequently help defining the progress of the specific goal, so new actions for improvement can be
initiated.
3.1.6.
Tools, Technology And Data Integration
The area Tools, Technology and Data Integration specifies the level to which online analytics are used by an organization. The tools
become most valuable when they are put into context and when results are being translated into actions by the web analytics
champion. Still, the usage of appropriate tools and technology was an important element of the research project. Google Analytics
(GA) is one of the most-applied tools to analyze website statistics. During the research project, most participants had Google
Analytics installed and configured, following Seb Chan’s GA health checklist. Once GA was properly setup, user segments were
installed that made sense for the participant’s organizations. Seb Chan recommended a set of basic user segments and advanced
user segments. More information about GA can be found in the chapter Website Analytics.
Apart from Google Analytics, other tools have been used in the research project in order to measure and evaluate participants‘
online endeavors. These were: Basecamp for online collaboration, the OAMM for defining participants’ analytical situation and
improvement, Museum Analytics for social media analytics and Qualaroo for a qualitative online audience investigation. Of course
there are many more tools available and in fact, used by participants. Discussion about the tools participants use on a day-to-day
basis was encouraged online. The five tools used most, alongside the ones covered in the research project, were: Facebook Insights,
YouTube Analytics, Hootsuite, Twitter Counter and Bit.ly.
Page 15 of 39
In conclusion, to improve participants’ online analytics maturity, several actions (as described above), were recommended. The areas
Management, Governance and Adoption, Objectives and Tools, and Technology and Data Integration, have been discussed during one
of the full day project workshops. Recommendations for other areas can only be realized within the scope of an individual
organization. Overall, participants were eager to accomplish the suggested actions and they were willing to help each other as is
proven by the information, tips and tools that were shared online.
3.2.
OAMM RESULTS
Participants’ OAMM results were collected using Hamel’s Online Analytics Maturity Self Assessment Tool. The outcomes of the
model were visualized in a graph, that shows at a glance how well balanced an organizations’ analytics situation is. A balanced
OAMM result demonstrates that an organization is distributing it’s limited resources as equally as possible to secure a sustainable
development of its online presence. Therefore, an ideal model is well balanced across the several key process areas.
“To be successful, executives must recognize web analytics is
more than a reporting system and represents an effective way
to identify weak points and improvement opportunities.”
(Stéphane Hamel)
Based on the first OAMM outcomes of March 2013, participants were encouraged to define a strategy for improvement by
discussing their organizational strengths and points of improvement during the first workshop of the research project. Based on the
final OAMM results of September 2013, an analysis of the analytics improvement was made.
3.2.1.
Before The Research Action Research Project
The project started by requesting participants to benchmarking their current online analytics situation, using the OAMM. The result
of participants’ average analytics situation in March 2013 was visualized in the following graph. The project started by requesting
participants to benchmark their current online analytics situation, using the OAMM. The result of participants’ average analytics
situation in March 2013 was visualized in the following graph.
Participants average benchmark, March 2013
Page 16 of 39
The OAMM average of the ARPNL participants showed that some development existed in most areas. However, improvement was
to be gained on ‘Management, Governance and Adoption’, ‘Objectives’ and 'Continuous Improvement Process and Analysis
Methodology'. This showed that very few participants had analytics spread within their organization, that no clear goals were defined
to report on online endeavors and that hardly any participant had a strategy to ensure analytics improvement on the long term. In
short, the OAMM result made clear that most participants had no structured approach towards online analytics. Although some of
the participants had expertise using analytics, their organization did not necessarily discuss the results of analytic reports on a
regular basis. In order to spread analytics among the organization, participants were recommended to setup a monthly meeting for
the discussion of analytics. Ideally, several museum staff members to whom the results of online analytics could influence their daily
work, would take part in such a meeting. The importance and definition of objectives was dealt with in the first workshop.
Most participants have a very broad scope of online platforms on which they are active, often including multiple websites, several
social media platforms, video platforms and publishing platforms such as blogs. This broad scope reflects the open and inclusive
policies that most museums follow. However, to aim at a broad scope might not necessarily represent an organizational strength. In
fact, a too broad scope might be detrimental to the museum online strategy. Instead, it was recommended to try to focus on
specific platforms, by which more value can be brought to the online visitors active on these platforms, instead of the online
audience in general.
The OAMM results were discussed during the first project workshop. Participants decided to further develop their continuous
improvement process and to better define online objectives. Most participants agreed that the scope area was too broad although it
was not completely clear how the scope could become more focused without affecting the museum vision of openness and
integration. Overall, the aim was to move towards a more balanced model where resources were equally distributed among all
areas.
3.2.2.
September 2013: After The Project Research
During the evaluation phase of the project, participants were requested to answer the same OAMM once again. The rational behind
this approach was to be able to compare the online analytics improvement throughout with the research project. The average
OAMM of all participants showed the progress from March to September 2013 and was visualized in the graph below.
Participants progress from March to September, 2013
Page 17 of 39
As can be seen by the OAMM, the project group has made a significant improvement in the 'Continuous Improvement' area and in
‘Objectives’. It should be no surprise that some improvement was realized in defining better objectives, as an entire workshop was
dedicated to that purpose. Participants were encouraged to define at least three goals to optimize their online marketing and
business strategy. Ideally, these goals had to be monitored on a frequent basis, using tools like Google Analytics.
The most striking development was the improvement found in the 'Continuous Improvement Process and Methodology' area.
Participants have evolved from no methodology to a methodology shared by an entire department. In short, scheduling weekly or
monthly meetings to discuss analytics have clearly helped participants to evolve towards a process of continuous analytic
improvement. During the project evaluation, participants declared to feel more confident as they have learned which tools to use
and how to interpret analytic results, not only within their own organization, but also in comparison to other museums. This can be
regarded as the first and most important step into spreading analytics, eventually, throughout the entire organization.
In terms of ‘Management, Governance and Adoption’, the project group showed great differences. Some museums were represented
in the workshops with a member of the management team, however, the vast majority of participating museums were represented
by project managers interested in using analytics. Still, the individuals attending the several project workshops, regarded themselves
as analytics champion towards the end of the project. They felt more comfortable dealing with online analytics data, tools and
reports, thanks to the research project.
Some individual participants have evolved in more areas, but the results of the project group in general remained stable, as the
OAMM results show. The variety in experience and knowledge among participants is a reasonable explanation for this result. The
workshops and the exchange of information and knowledge online, gave the opportunity for organizations to level up in terms of
expertise and the tools used. In March 2013, the museums were requested to reduce the scope of their online approach. However,
due to many museums’ policies aimed at openness and a broad reach of audiences, most participants were not able to reduce the
scope of their online activities.
In conclusion, between March and September 2013, the online analytics situation of the participants has improved. The model is
better balanced as the core strength and point of improvement have shifted since March 2013. However, as Hamel explains, the
model is continuously altering and improvement in all areas is always possible. Apart from using the model to define the current
state of analytics, the model is most useful to set priorities or define a strategy for where to go next. ARPNL has offered the tools,
resources and benchmarking to support future improvement for participants.
3.3.
INTERNATIONAL CONTEXT
The international context of the OAMM results can be considered in comparison to the OAMM results of Tate Gallery, United
Kingdom. Tate first started using the model in 2010, as part of a research project and the museum has known great improvement
since then. In 2010, Tate had no analytics strategy defined and no one was responsible for managing the analytics (blue graph). A big
step forward was made after participating in the Culture24 Let’s get real action research project in 2012 (yellow graph). With the
tools and insights provided in that research, Tate refined each area and developed an analytics strategy to define where they want to
go next (pink graph).
The international benchmark confirms that
museums worldwide are facing the same
challenges in regard to measuring and evaluating
their online success. Similar to Tate, the Dutch
project group had no methodology for a
continuous improvement of analytics process
when they first started using the model.
Furthermore, Tate’s scope in 2012 is identical to
the Dutch projects’ scope in both March and
September 2013. A possible explanation is that
museums worldwide share similar visions of
openness and accessibility of heritage to a wide
audience.
In short, from the Tate OAMM progress and
Tate’s web analytics maturity chart 2010-2013
Page 18 of 39
research, it was learned that apart from knowledge and tools, time and effort are needed to keep up with the analytics situation and
to make analytics, eventually, part of the entire organization. Both Tate and the Dutch projects’ OAMM results, show a similarity in
growth, moving towards a stable and well balanced model.
The Culture24 research in the UK served as a kick start for analytic awareness and improvement, that Tate continues to carry out.
Based on the OAMM improvement of the Dutch research between March and September 2013, it is likely that the Museum
Analytics Action Research Project serves a similar purpose in the Netherlands. Several participants mentioned that towards the end
of the project, the OAMM was used within teams and sometimes within different departments. Furthermore, some participants
stated that the results were being compared and discussed on a regular basis.
3.4.
CONCLUSION
The OAMM was a useful tool to visualize the state of analytics among participants and to measure analytical improvement. As evaluating online endeavors is a continuously shifting process, it should be continuously structured and evaluated. Thanks to the OAMM,
the areas that needed most attention and improvement were defined. Next, participants were provided with tools and actions to
improve on these areas. The progress in OAMM results between March and September 2013 showed that online analytics have
found more solid ground within all participating organizations, as is in accordance with the project aim.
Page 19 of 39
4.
WEBSITE ANALYTICS
4.1.
INTRODUCTION
4.2.
ONLINE VISITORS MOTIVATION
The website analytics part of this action research project focused on analyzing and communicating whether museums are achieving
the proposed goals for their websites. The research was developed from two different perspectives: the visitors needs and the goals
defined by the organization. In terms of visitors needs an online survey was conducted in order to answer one fundamental question: what are visitors looking for when they visit museum websites? Google Analytics (GA) was used to analyze the participants
websites performance against the goals previously defined. Participants were requested to follow a health checklist to guarantee
that GA has been properly installed and that it delivers consistent information across all organizations. Participants were requested
to implement their organizational goals using GA in order to receive frequent reports not only about standard metrics, such as
visits, average time on site but also other metrics like e-newsletter subscriptions and educational material downloaded. In order to
spread an analytics culture throughout the entire organization, several different reports were created and shared with all participants. For example, special reports were created targeted at the needs of curators, reporting information about the online usage of
collections. Finally, the participants' websites were benchmarked and online metrics were contextualized within the group.
Museum professionals have access to a myriad of tools that can reveal useful information about online audiences. Software like
Google Analytics provides detailed information about website visitors and their behavior during their visit. Although this information
can be useful to understand online audience behavior, it often does not answer fundamental questions, such as, what is the motivation behind a museum website visit?
The Indianapolis Museum of Art conducted a series of studies to answer that exact question. The results of their studies are
published online. In short, the museum team collected feedback from website visitors in order to identifying online motivational
categories. With a list of those categories at hand, the team conducted another online survey asking visitors why they were visiting
the museum website. The question, the possible answers and respective responses are included bellow: Today, I am visiting the website to:
•
Plan a visit to the museum (50%);
•
Find specific information for research or professional purposes (16%);
•
Find specific information for personal interest (21%);
•
Engage in casual browsing without looking for something specific (10%);
•
Make a transaction on the website. (2,7%).
The Indianapolis Museum of Art published their results with hope that they would provide a reference dataset and a replicable
model for other museums that are interested in conducting a similar study.
That is exactly what the 15 Dutch participants did. The museums used a bi-lingual survey in Dutch and in English which was
delivered depending on the language of the webpage being visited. The results were similar to those found by the Indianapolis
Museum of Art. A noteworthy exception was that in the present project the number of casual browsing was significantly higher. A
possible reason for that discrepancy is the fact that in the Dutch survey the option 'Engage in casual browsing without looking for
something specific' was presented as the first option. Some of the key findings are resumed bellow.
Museum Analytics - Website Analytics
Page 20 of 39
Approximately half of online museum visitors are planning a museum visit.
It should be no surprise that people visit museum websites while preparing a visit to the museum’s physical campus. From this
research, 47% of online museum visitors are planning a visit to the museum. The numbers of answers for each individual museum
ranged from 35% to 57%.
Most museum websites include clear information for those that plan a visit although often that information is spread across several
webpages. Nowadays, Google also supplies most necessary information, such as, address, telephone, opening hours and even a
museum rating based on visitors reviews. Since planning a visit is the main motivation to visit museum websites, how can museums
take a step further in facilitating that process? A possible approach could be to send an automated email including all necessary
information for the visit: opening hours, admission fee, direction from visitors location to the museum based on user preferred
transportation mode and an overview of the programme available on the day visitors have chosen.
There are more online museum visitors searching information for personal reasons than for professional
reasons. For the 15 museums that participated in the project, in average 16% of the visitors claimed to be searching information for personal
reasons whereas only 13% of the visits had a professional motivation. This is a relevant finding as often museums struggle with the
decision of customizing their online information for scholars and museum professionals or for a general public.
Also relevant to notice is that there is a considerable large online audience that is not necessarily planning a visit neither searching
for information for professional reasons. Indeed 37% of the online visitors were either casually browsing or searching information
for personal reasons. Just like people are interested to visit a museum physical campus, there seems to be an audience interested in
spending some of their leisure time to visit museum websites. What do museum websites have to offer to these audiences? What
Page 21 of 39
can online visitors do and learn on museum websites? The Tate website displays prominently in the front-page a large image from
their collection which is not necessarily related with the events at their physical location and instead it is targeted to online
audiences. The Rijksmuseum website makes three clear propositions to online visitors: 'Plan a visit', 'Collection', 'About the
museum'. In the collection section, visitors can explore or search both the collection and the library catalogue, as well as create and
download their your own collection using the Rijksstudio. These are two examples of interaction scenarios targeted at online
audience that are not necessarily professionals nor people planning a visit. Many museums produce audiovisual content that is
interesting for online audiences at large, however, this content is not always clear accessible on their websites. We believe this is a
missed opportunity as there is a large number of people interested in spending some of their leisure time visiting and exploring
museum websites.
Online museum visitors are interested in shopping online.
While planning the online survey, participating museums considered to remove the answer 'To book for an exhibition or event, or
buy something'. The main reason being that 8 from the 15 museum websites did not include e-commerce functionality. The group
decided to keep the answer for matters of consistency with the study done at Indianapolis Museum of Art. Despite the fact
that more than half of the museums did not offer e-commerce in their websites, it was surprising to notice that 4% of the online
visitors still intended to book an event or buy a product.
Museums have been integrating e-commerce in their websites for several years. Today museums sell online tickets for events and
exhibitions, request donations or memberships and create online shops to sell books and other products often selected by the
museum staff. There are several examples of museums successfully using e-commerce. The Metropolitan Museum of Art reported
to have generated $4.5 million on online membership sales alone, in the fiscal year of 2012. In 2009/2010, Victoria and Albert
Museum reported an overall net turnover for online retail of £614,862 which represented 9.4% of their overall retail
turnover. Their website had a conversion rate of 1.63% which means that every 3 in 200 online visitors completes a successful
transaction. These are two examples of museums successfully using their websites as an extra source of income. As museums gift
shops are established revenue sources for museums, it is logical that online shops will follow a similar path.
In the spirit of open data, the raw data collected during the online survey have been made accessible online. The survey was
installed on several museum websites using an online survey tool called Quaraloo. The survey was available online on the website of
15 participating museums from June 17th until July 17th, 2013 and more than 41,000 answers were registered. The study can be
easily reproduced by other museums and the results can be contextualized with the ones published in this report.
4.3.
GOOGLE ANALYTICS
Google Analytics (GA) is a popular free tool to analyze how visitors find and use websites. Once GA is properly installed in a website it is possible to make informed decisions about design and content, improve the website to convert more online visitors into
onsite visitors, track the performance of keywords and other marketing campaigns and track metrics such as revenue, average order
value, and e-commerce conversion rates.
Most participants had Google Analytics already installed prior to the project. The participants representing the Museon museum did
not install Google Analytics in their website as at the time of the project the website used a CMS that was no longer maintained.
The Museon found it relevant to get involved in the project as they were involved in developing a new website and the focus of the
project was relevant for their own online ambitions.
4.3.1.
Google Analytics health checks
Similar to the project Culture24, we have recommended a health checklist to the participants GA installation. This health check is
important in order to retrieve consistent information from GA as well as to have access to more detailed information that
otherwise would no be available. Participants were requested to follow the Culture24 health checklist. In addition we have either
updated or added new recommendations to the existing list. A document has been created containing information that can be passed to an IT department with instructions on how to properly
setup GA. The instructions can be accomplished within 1 to 2 hours. A list with a non-technical explanation about the proposed
improvements is included bellow.
Page 22 of 39
1.
'New' Google Analytics code: In order to use GA, a small piece of code needs to be installed in every page of a
website. In December 2009, Google has launched a new tracking code that offers an improved way to track website
visitors. The document includes instructions on how to upgrade to the latest GA tracking code.
2.
Monitor file downloads, clicks in external links and clicks in e-mail addresses: The default GA
installation does not track the number of people that download files from a website, such as, PDFs or images. Also, it is
often useful to know when online visitors click in links to external websites and e-mail addresses. The instructions are
included in the document and the new information is available at GA/Content/Events/Overview. 3.
Monitor shares with social media buttons: Modern websites include buttons that allow visitors to share
content on social media platforms like Facebook and Twitter. Recently Google has included the possibility to
track content sharing, both on your own website and on external social sites. The new reports can be found on GA/
Traffic Sources/Social/Plugins. 4.
Improve visitor interaction on your website: Most websites include multiple paths to reach the same
content. For example, the opening hours of a museum might be accessible on the top menu as well as in a link in the
footer. With the default GA installation it is not possible to collect information about which of the two links is most
used. The GA health checklist includes instructions on how to track multiple links to the same content. The new
information is available at GA/Content/In-Page Analytics.
5.
Monitor multiple websites in a single report: Many organizations have multiple websites, often accessible at
different domain names. A common question is how many visits all these websites attract per week or month. It is
possible to unify multiple websites in a single GA report. The document includes information on when it is advisable to
do it and how to properly setup the multiple websites.
The Google Analytics health checklist with clear instructions for IT professionals is available to be downloaded at: http://
www.intk.com/action-research-project/google-analytics-health-checklist/GAhealthchecklistforIT042013.pdf
4.3.2.
Google Analytics Dashboards for museums
A proper installation of Google Analytics can provide museum professionals with a myriad of information about their website usage
and respective online visitors. It is therefore important to be able to glean the useful snippets of information and deliver them to
those inside of the organization that can use the information to take informed decisions.
"Not everything that counts can be counted, and not everything
that can be counted counts." William Bruce Cameron 1963
Information collected by GA can be visualized using widgets and multiple widgets can be organized in dashboards. During the
research project three special dashboards have been created focused on the information needs of museums professionals: overview
report, social Media report and Collection report.
Define goals to be included in reports
The reports or dashboards can only truly measure success in case clear goals have been defined using GA. Bellow are a selection of
goals previously mentioned on chapter ‘Online Analytics Maturity’:
•
Reach more people online: GA goals are typically used to measure how well a website fulfills a certain target. GA
goals are typically not used to measure increase on online traffic as online traffic is already one of the metrics presented
on GA. This goal can be easily measured using the Audience report existing in GA.
•
Increase e-mail newsletter subscriptions: A website can be an optimal tool to increase e-mail newsletter
subscriptions. For example, a call for action can be include in the front-page or next to a blog entry inviting visitors to
subscriber to the newsletter in order to receive announcements of new content or events. GA goals can be used to
measure the performance of the website in terms of new subscriptions. Typically newsletter subscriptions goals can be
setup by associating the goal with the ‘thanks for subscribing page’ or with an event once a visitor clicks subscribe.
Page 23 of 39
•
Increase engagement to online content: For many museums their mission is to increase understanding of a
certain topic and with that goal in mind they have made information about part of their collection accessible online. In
order to measure whether a website is successfully delivering that content to visitors it is possible to associate a GA
goal to a page that contain such content. Depending on the nature of the content, it might be useful to consider that the
goal is only achieved once a visitor has remained a certain time on the page. Furthermore, some museums have found it
useful to associate a monetary value with such goals. The logic is that, online visitors that access information online
saved money to the museum as opposed to people that call requesting similar information.
•
Download special content: Many museums produce digital material (educational, press, etc) to be distributed via
their website. GA goals can be used to measure how many times such material has been downloaded. This can be
achieved by associating an event with the download of files, as described in the GA health checklist. The GA goal can be
based on the events triggered when files are downloaded.
•
Increase visits to the museum: Museums that sell e-tickets online can use GA e-commerce functionality to
report on metrics that might include the sales of museums tickets. Museums that do not have e-commerce functionality
implemented can use an approach that, although not exact, can lead to a good approximation about the impact that the
website has on museum visitors. For many museums, there is a co-relation between the number of people that visit the
opening times page and the number of people that visit the museum. Let’s say that for a given museum, 30% of online
visitors that visit the opening times page do visit the museum and that the average ticket entrance for the museum is 10
euros. This means that approximately for each 100 online visitors, a total of 300 euros is generated on ticket sales at the
museum. This information can be included in GA by associating a goal to every unique visitor that reaches the page
about opening times and adding a goal value of 3 euros.
Overview report
The overview report is a dashboard based on three metrics: Visits, Traffic sources, Most popular content. The graph (or widget)
with the visits per day provides a fast overview on the best and worst days of the month in terms of number of online visits. Traffic
sources includes the top 6 sources that have directed the most traffic to a website in the last month. Some of the sources that can
expected to be seen in this metric are google, social networks and possibly newsletters (in case UTM codes are being used). The
third metric 'most popular content' shows the top 10 most visited webpages. Usually the front-page is the most visited page,
followed by pages included in the website navigation menu. Other pages that can be expected in the top 10 are popular blog posts
or any other pages that have attracted the attention of online visitors. On the right column of the dashboard, it is possible to find
several other metrics: Total number of visits, Percentage of new visits, Traffic type Overview, Average Visit Duration, Bounce
Rate, Page per visit, Keywords. In order to keep the dashboard generic, the most common metrics have been included. It is
recommended that each organization customizes the dashboard to their individual needs. For example, goals have not been
included in this report in order to keep the difficulty level as low as possible. In case goals have been defined on GA, it is possible to
add them to the dashboard. The Dashboard is called '1. Overview Report V1.0 (by intk.com)' The first '1' is there in order to keep it
as the first Dashboard in the Dashboards list, we have also included a version number. The dashboard can be installed by following
the link: https://www.google.com/analytics/web/template?uid=tygIIVuYQRO4HWPajY4w6A
Social Media report
The Social media report is a dashboard focused on the impact that social networks have on the website. The dashboard is based on
three metrics: Visits, Most popular content and Social network. These metrics are restricted to visits that had its origins in social
networks. For example, the most popular content only shows the number of visits that had origins in social media. Any museum can
automatically instal the dashboard in their GA by following the link: https://www.google.com/analytics/web/template?
uid=km4u1uMKRe6h-eamM3o-tg
Participants that have social media sharing buttons installed in their website and have values applied to goals and followed the GA
health checklist for IT can have access to a more advanced social media dashboard. The dashboard can include information about
the content that is being most shared using the website sharing buttons. Every time a visitor uses a social media button it counts as
a social action and there are widgets that show the relationship between the number of action and the number of visits generated
via those actions. The 'data hub' is another interesting feature to be included in the advanced social media dashboard. It includes
information about what content people engage with, share, and discuss on sites other than the museum site itself. For example, if a
message is posted on Google+ with a link to the website it will appear on the 'data hub' report. Currently 'data hub' includes
Page 24 of 39
activities from more than 17 social networks. Unfortunately popular social networks like Facebook and Twitter are not included in
the list of partners. An advanced dashboard that includes the referred widgets can be found on the dashboardjunkie.com website.
The social media dashboards are specially interesting for museum staff members that are actively involved in social media. These
reports are useful to inform staff members about the impact that their efforts have on the website. Due to the often high activity on
social media, it is recommend to send social media reports on a weekly basis.
Collection report
Many museums have made available a digital version of their collection online. A dashboard can be used to continuously evaluate
some of the impact that digital collection have online. The collection dashboard can be downloaded at: https://www.google.com/
analytics/web/template?uid=fbi2trteQeWsDItohc6HsA
Some of the questions that the dashboard answer are :
•
How do visitors find the collection?
•
How many visitors browse the collection? •
What percentage of visits include the collection?
•
How long do visitors stay in collection pages?
•
Which search terms lead to the collection?
•
What devices do visitors use?
The dashboard needs to be customized for each individual museum. In order to customize it, each individual widget needs to be
edited. Widgets include one or more filters. One of the filters only shows pages containing the word 'collection'. The filter assumes
that URLs to the collection include the word 'collection'. To customize the filter, first it is necessary to inspect the construction of
the URLs that lead to collection pages. A pattern needs to be identified and added to each individual filter. The collection dashboard
is relevant for staff that are involved with the collection and have access to change information available online about each individual
object.
4.4.
PARTICIPANTS WEBSITE BENCHMARK
GA is a useful tool to measure and report on web analytics. The information that GA reports is useful for several departments
within cultural organizations. However, for museum professionals that are new to GA, some of its results can be abstract and appear
to be meaningless. For example, what does it mean to have 1,000 visits per month? Is that a lot or little? A possible way to better
understand these values is to contextualize them with values from other organizations. Due to the collaborative aspect of the research project, it was possible to benchmark the web analytics of the several participants.
4.4.1.
Web visits
During the research project, the web metrics of
15 participants have been collected and
presented in a comparative fashion. The data
used have been collected with GA from April
1st until April 30th. The number of visits per
month was the first metric to be analyzed. The
figure bellow includes the results of the top 10
museums in terms of visits.
The number of online visits across the
participating museums varied from a few
thousand visits per month to a quarter of a
million visits per month. As expected, the most
popular museum websites correspond to the
museums that have the largest onsite
audiences. From the participants the Van Gogh
museum is the most visited one and its website
Page 25 of 39
is also the most popular in terms of visits. More than half of the participants registered twice or more online visits than onsite visits.
The relationship between the number of onsite and online visits can be useful to contextualize the number of online visits. So to
answer our previous questions: a website that registered 1,000 visits per month can be considered successful for a museum that has
less than 500 onsite visits whereas it might be considered a modest result for museums with more than 500 visits.
4.4.2.
Average time on site and bounce rate
Average time on site and bounce rate are two common metrics measured by most web analytics tools. The meaning of these two
metrics is not always obvious. Do visitors spend more time on a website because they have difficulties in finding the information
they are searching for or because they did find the information and spend time engaging with it? Nevertheless, it is interesting to
benchmark these metrics in order to contextualize the results. With such benchmarks, we can answer the question: what is a
common average time on site and bounce rate? For the participants the average time on site ranged from 1,5 minutes to
approximately 3 minutes whereas the bounce rate ranged from 35% to 50%. These metrics are also useful to analyze individual
pages where it might be easier to understand visitors behavior. For example, websites might have a big discrepancy between bounce
rate for mobile and desktop users. This might be a sign that their website is not ready for mobile phones. In the spirit of open data,
participants have agreed to share the mentioned online metrics. At the Museum Analytics website it is possible to see monthly
updated data from participants about website visits, average time on site and bounce rate. The data is collected automatically using
GA API.
4.4.3.
Geographic segmentation
The geographic location of online visitors is an important metric for museums. Some museums are targeted at local audiences
whereas others are oriented towards national or international visitors.
Although geographic location is often not a barrier to visit a website, the language that websites are available in do tend to be.
Naturally, dutch museums that are mainly oriented towards a national audience tend to have their websites available only in Dutch.
Page 26 of 39
In order to better understand online audiences, the number of visits for all participants have been segmented in 3 groups: local city,
national (excluding local city) and international. The results are presented on a normalized stack bar chart on the figure bellow. Most
participants have a larger national online audience than an international audience with the exception of the Van Gogh museum. In
their case, the majority of online visits was based in Amsterdam, London and Paris, making the Van Gogh Museum the most
internationally oriented museums from all participants. The most visited websites by local audiences are the websites of the Eye
Filmmuseum, Amsterdam museum and Stedelijk museum. All three museums are based in Amsterdam which is the most populous
city of the Netherlands. The Kröller-Müller museum has a relative low percentage of local visits and a quite large percentage of
international visits which is an expected result as the museum is located in a rural area and it presents a collection targeted to
national and international visitors. Geographically segmenting online audiences is useful to understand whether audiences are in
accordance with the aimed target audiences.
4.4.4.
Visits using mobile devices
Several participants were interested in knowing more about how visitors access their website with a special emphasis whether the
devices used were a desktop or a mobile device. According to a study published by Statistics Netherlands six in ten Dutch Internet
users used a mobile device to access the internet in 2012. Since 2010 smartphones have surpassed computers in terms of internet
access. The tablet computer, which was first measured in 2012, was used by nearly 20 percent of mobile internet users in that year.
The web visits from all participants in the month of April have been analyzed in terms of used devices. The percentage of mobile
visits ranged from 19% to 32%. Naturalis Biodiversity Center and Science Center NEMO where the museum websites with the
largest percentage of mobile visitors both passing the 30% whereas the Museum Rotterdam registered approximately 19% mobile
visitors. Besides the actual mobile visits it is also important to analyze the growth across the years. From April 2012 to April 2013,
the Naturalis mobile visits almost doubled increasing from 7,926 to 14,612. The data collected from the participants websites as well
as the information published by the Statistics Netherlands indicate that there is a growing online audience using mobile
devices. Also worth noticing is that the most popular mobile devices in this research were the iPad flowing by the iPhone. This is a
relevant finding, as most participating museum websites are not ready for mobile devices. Participants did mention that adapting
their website for mobile technologies is a high priority in the organization.
4.4.5.
Visits by source
The web visits have also been analyzed in terms of their source. This is a particularly relevant study in order to understand where
online visits have their origin.
Page 27 of 39
Web traffic has been categorized by source: Search Traffic, Direct Traffic, Referral Traffic and Campaigns. Search engines such as
Google and Bing are the sources that drive more traffic to website. In average across all organization 52% of their traffic starts on a
search engine. These results are similar to the ones published by the Culture24 research project, where participating organizations
registered 56% of search traffic. Most organizations registered a considerably low percentage of visits from online campaigns. These
results do not necessarily mean that online campaigns are ineffective. Instead it reveals that newsletters and other online campaigns
are not being properly configured. It is possible to measure the impact that online campaigns have on websites by including UTM
parameters in all links to the organization website. Once this parameters are included the number of visits is reported in the
'campaign' source.Very few museums are using UTM codes in their newsletters which indicates that their current impact in terms of
website visits cannot be measured. The study of e-newsletters is one of the topics that participants found relevant for a future
research project.
4.5.
CONCLUSIONS
Museum websites play a fundamental role in the organization online strategy, therefore it is important that museum professionals
are able to analyze their performance against organizational goals. Online surveys and Google Analytics are two essential tools to
deliver such analyses.
Online surveys are a useful method to collect qualitative information and answer fundamental questions about websites and
organizations. It was surprisingly easy to run a survey in 15 simultaneous websites. The large number of answers collected reinsures
the quality of the results obtained. The outcomes of the survey have already been discussed in this chapter, therefore this conclusion
focuses on possible future work. An interesting question to be asked in a future survey is after asking the motivation, ask whether
such motivation has been fulfilled. This survey would answer the question of whether visitors that were planning a visit found the
website useful and had access to all the information that was needed. Additionally, the survey could directly request the opinion of
online visitors about what could be improved. Although online surveys can be useful to answer specific questions, one should take
into account that visitors might not have the answer to the question and might not be interested in seeing their online experience
interrupted by online surveys.
Most participants were using GA before the start of the project. However, not all participants were acquainted with it, neither knew
how to create reports that answered their specific questions. In order to answer some of the questions requested by the
participants changes had to be made to the default GA installation. Most participants did not had clear goals defined on GA.
Without those goals, it is not possible to measure the performance against the organizational goals. A list of goals was defined with
instructions on how to implement them on GA. During a workshop, participants were able to define reports using GA dashboards
about information that might be interesting for different departments within the organization. These reports can be configured to
be sent periodically to the right people inside the organization. Period reports can play a fundamental role not only in improving the
website but also at spreading an analytics culture inside organizations.
Page 28 of 39
5.
SOCIAL MEDIA ANALYTICS
5.1.
INTRODUCTION
The third and final workshop focused on social media activities and whether participants were achieving the goals proposed for the
medium. Museums have joined social media platforms in order to reach audiences in platforms where they are already active. Social
media platforms have a growing audience and this audience has demonstrated interest in interacting with content shared by museums. A proof of such world-wide social media interest is shown in the front-page of the website www.museum-analytics.org where
it is listed the most engaging content published by museums world-wide in the last 7 days. As it can be observed some of the most
engaging posts on Facebook can reach more than 5,000 likes and some of the most engaging tweets are often retweeted more than
500 times.
The research project focused on two social media platforms: Facebook and Twitter. The choice was based on the fact that these
were the two social platforms that all participants had already a large audience. In preparation for the workshop, participants where
asked to share the goals they have defined in their social media strategies. The three most common goals mentioned were: increase
fan base and engagement, increase website visits and increase museum visits. During the workshop, each goal was tackled
individually. In terms of fan base a benchmark was created between the participants. In order to improve engagement, increase
website traffic and museum footfall best practices were presented and discussed.
5.2.
PARTICIPANTS BENCHMARK
In order to benchmark the participants performance on social media, a special report has been create on Museum Analytics. The
report is automatically updated every day. It includes a graph that shows among the participants, the most popular museums on
Facebook and Twitter. As expected the museums with a larger onsite number of visitors are also typically the ones that have the
larger number of Facebook Fans and Twitter Followers. In the case of the participants, the most popular museums of Facebook and
Twitter were the Van Gogh Museum, the Stedelijk Museum and the EYE Film Institute.
The report also includes a a list of the most engaged content published on Facebook and Twitter in the last 7 days. Museum
Analytics measures engagement on Facebook by adding the number of likes and comments whereas on Twitter engagement is
measured by simply looking at the number of retweets. Typically the most engaging content is dominated by museums that have the
largest audience. However, it is interesting to notice that on Facebook the most engaging content often includes a picture. On
Twitter the most engaging content often includes information specific to day or season it was published. An example is the Tate
weather, a weekly feature when Tate social media manager publishes one image from their collection with a message about the
weather in London. These type of features allows both the communication department and the public in general to look at the
museum collection from a different angle.
The participants were also interested in understanding how their fans and followers base could be contextualized in the museum
sector worldwide. A world ranking was calculated for July 2013. In terms of fans, Van Gogh museum was among the top 50 most
populars museums whereas Stedelijk museum, the EYE Film Institute and the Gemeentemuseum were among the top 500 most
popular museums on Facebook. In terms of followers, Van Gogh museum is among the top 100 most popular museums whereas, the
EYE Film Institute, the Netherlands Institute for Sound and Vision, Amsterdam Museum, Centraal Museum and Gemeentemuseum
are among the top 500 most popular museums on Twitter. These rankings do not imply that there is a competition among
museums. Instead the ranking can be used to contextualize numbers that otherwise can be quite abstract.
Museum Analytics - Social Media Analytics
Page 29 of 39
Finally, the growth rate of Facebook fans and Twitter followers was analyzed. On Facebook the fastest growing museums where
Naturalis Biodiversity Center and Museum for communication. Naturalis tripled their Facebook fans when comparing the number
of fans in January 1st, 2012 with the number of fans in December 31st, 2012.
On Twitter, the growth was less accelerated as on Facebook. The fastest growing museums were Kröller-Müller Museum, Stedelijk
Museum Amsterdam and Scheepvaartmuseum. The followers of the Kröller-Müller Museum increased by 400% during 2012. Bellow
are two graphs with the growth on Facebook and on Twitter for all participants.
Page 30 of 39
5.3.
PARTICIPANTS BENCHMARK
Museums are popular on social media. Some of the most popular museums can gather millions of fans both on Facebook and Twitter. But how are museums doing in terms of engaging and how can their social media publish improve? Based on recommendations
from Facebook and Twitter, publishing best practices are presented bellow.
•
Be succinct. Facebook states that 'Posts between 100 and 250 characters (Less than 3 lines of text!) see about 60%
more likes, comments and shares than posts greater than 250 characters.' Posts don't necessarily have to include all the
information you want to communicate. Instead, posts work better as information teasers. Posts should include a URL to
a web page where fans can find further information.
•
Post regularly. Facebook recommends: 'Post at least five times a week to stay top of mind for fans. Consistency is
also key.' Publishing on Fridays, Saturday and Sundays is fundamental for museums as that is the time people plan their
weekend activities. Weekend posts can be planned and pre-schedule during the week and published during the weekend.
•
Post photos and videos. Posts including a picture generate about 120% more engagement than the average post.
Think twice before publishing a post without a photo or a video.
Twitter publishing best practices are not radically different from the ones mentioned by Facebook. Here are some of the best
practices recommended by twitter:
•
Keep Tweets short and sweet. Tweets cannot be longer than 140 characters but it seems that shorter is even
better: "There’s no magical length for a Tweet, but a recent report by Buddy Media revealed that Tweets shorter than
100 characters get a 17% higher engagement rate."
•
Make it real-time. 'Timing matters, especially for breaking news or live Tweets. But how about for everything else?
The short answer: it depends on the content of the Tweet, your objectives, your audience, their geography and
more.' There are a plethora of apps that estimate the best time to tweet based on your audiences
activity. Followerwonk is one of them.
•
Tweet often. 'Tweet frequency depends on a number of variables such as your audience, purpose and business
objectives. That said, a good basic rule is between three to five Tweets per day.' Whereas Facebook recommends 1 post
a day on Twitter you can make several tweets a day. Make sure they are distributed throughout the day.
Twitter and Facebook best practices can be found online: https://business.twitter.com/best-practices and http://fbrep.com/SMB/
Pages_Publishing_Best_Practices.pdf.
Managing the social media platforms of a museum can be extremely time consuming. The mentioned publishing best practices can
help to increase the productivity of the time invested on social media.
5.4.
WEBSITE VISITS VIA SOCIAL REFERRAL
The main reason why a museum joins social networks should not be to increasing website visits. Social media are best suited to
establish an interaction or dialogue with audiences. Starting a conversation with online audiences is therefore a more suited main
goal than to increase website visits. Nevertheless, increasing website visits is a plausible goal to include in a museum social media
strategy. Social media platforms can indeed be useful to drive traffic to websites. After all, some museums have thousands of fans
and followers and therefore it makes sense to use these platforms to increase awareness of the fact that a museum might have their
collection available online or to announce special activities such as workshops, guided tours, etc. Website visits are increasingly important for museums. For example, for some visitors it might be the first encounter with the museum which will encourage them to
visit the museum campus. For others, it’s a way to spend leisure time exploring all kinds of museum content, which may lead to a
close engagement between visitors and museum, that again might lead to donations (memberships) or for example a purchase in the
online shop. Based on these reasons many museums include as one of their social media goals to drive traffic to their website. Website visits that have their origins on social media platforms are commonly referred to as 'visits via social referral'. We have analyzed
the traffic of all participants to understand which participants were generating most traffic via social referrals. The outcome of the
investigation is shown in the graph.
Page 31 of 39
The percentage of visits via social referral varied among the participants from 1% to 7%. The EYE Film Institute and the Centraal
Museum were the participants that had a higher percentage of visits via social referral. There are many reasons why a museum might
have a higher or lower percentage of visits via social referral. Website traffic can have its origins in different channels, such as,
organic search, paid search, direct traffic, other websites, email campaigns, etc. When evaluating the source of a website traffic it is
important to take all the channels into account. Ideally, the number of visits from each channel should reflect the effort and
investment an organization has devoted to the channel. For example, in case a museum has a dedicated social media team and the
one of the goals of their strategy is to increase website visits then it would be expect that the traffic from social media networks to
be considerably high.
Social media managers can influence the number of people that navigate from social network platforms to the museum website. In a
previous research project, INTK collaborated with a museum in order to do exactly that: increase the traffic that social media
platforms where bringing to the museum website. It was decided to focus on the social networks for which the museum already
had an audience. The number of fans and followers on Facebook and Twitter was approximately 5,000 people for each network. The
organization’s staff was already posting and tweeting at a regular basis. The new strategy introduced two simple changes. First, the
staff members were instructed to include in most posts and tweets a deep link back to the website. Secondly, a custom weekly
report was created to inform the staff members about how many website visits each of their posts/tweets had generated. The
website traffic with its sources on social networks, increased from 200 visits a month to 2,000 visits a month. This represents an
increase of visits via social referral from 3% to 17% of the total online visits. As it is evident from the collected data the strategy had
a clear impact in the website traffic.
Page 32 of 39
5.5.
INCREASE MUSEUM VISITS WITH SOCIAL MEDIA
Among museum professionals there is a clear understanding that the internet in general and social media in particular cannot be
reduced to marketing tools to promote museum visits. Many museums have been successfully using social media to engage with
their public and to achieve institutional goals online. That said social media can be also be a powerful tool to encourage online audiences to visit the museum.
Parallel to the action research project, INTK has developed an experiment together with Naturalis in order to understand the
impact that social media can have on museum visits. The goal of the experiment was to create a campaign using a social media
platform and precisely measure how many people the campaign would reach and how many of those would end up visit the
museum. At the time of the experiment, Facebook included a feature called "Facebook offers". The feature allows business to offer
discounts to Facebook offers. To claim an offer, users have to click on a "Get offer" button visible on a post made by the museum.
See the post bellow:
Users that claim the offer receive an email with instructions on how to redeem their discount. For our experiment, we offered a
second ticket for free for visitors that claim the offered and bought one ticket. This is similar to offer 50% discount, however, for
marketing reasons to use the Dutch word Gratis (for free) often leads to a higher impact than to announce a percentage discount.
The email that the users received look as following:
Page 33 of 39
The post that contained the Facebook offer was boosted with advertisements. In the period of 22 days, the post was shown more
than 1 million times. More than 7,000 coupons were claimed online. And a total of 781 coupons were redeemed at the
museum. The campaign brought at least 1562 new visitors in 22 days, representing a 5,5% increase in visits for the mentioned
number of days. The redemption rate was 11% (redeem coupons/claimed coupons). The month that the campaign was launched
(February 2013) was the most popular month in terms of visits in the last 3 years. The experiment shows that social media managers can use the marketing features offered by Facebook to drive footfall to the
museum. At the time of writing, other social networks such as Twitter, do not include similar features. However, it is possible to
reproduce the offer feature using custom software and use Twitter advertisements to drive traffic to the offer. Finally, it would be
interesting to compare the costs and the impact of such marketing campaign with the costs and impact of traditional print
marketing campaigns.
5.6.
CONCLUSIONS
As he number of internet users in general keeps on growing and the number of social media users increases rapidly, museums will
most likely further invest in these platforms. One of the main reasons for doing so is that all cultural organizations aim at being
where their audiences are. Social media in particular has promised the possibility for museums to rethink the relationship with their
publics. Museum have been experimenting with different kinds of content and different kinds of engagement on social networks. We
have seen museums creating regular features that are must appreciate by their audiences. The activities that museums have developed using social media have definitively encouraged cultural organizations to be more open
and accessible. An example of such activities it the "Ask A Curator Day". The project brings together passionate experts from
museums and galleries around the world to answer the public questions on art, history and science. On an announced day a
museum expert is available for conversations with the audience via social media. Any museum or gallery is free to join and the
number of questions asked and answered has been growing throughout its multiple editions. This is a straightforward example on
how cultural organizations can be more open and accessible using social media.
Finally, the impact that social media had among the participating institutions in terms of website and museum visits has been
analyzed. The website of all participants has seen an increase of web visits via social referral. At the time of writing the percentage of
visits via social referral among the participants varied between 7% and 1%. It has been shown that it is possible to further increase
website visits via social referral to larger values such as 15% or even 20%. Furthermore, the potential that social media has to
increase museum visits, has also been researched Using Facebook offers it has been shown that it is possible to encourage social
media users to visit the museums. Campaigns launched on social media can often be tracked in all its steps. This means that
marketeers have access to precise information about how many people have seen the campaign, how many have claimed an offer
online and how many of them have redeemed the offer at the museums. These precise information affords marketeers the
possibility to understand how users engage with their campaigns at all stages and subsequently take decisions on how to improve
the overall campaign.
As previously mentioned, the number of social media users keeps on increasing and the nature of social media platforms itself keep
on changing rapidly. It is not possible to predict the future, however the time is ripe to collect the fruits that museums can grow
using social media.
Page 34 of 39
6.
PARTICIPANTS FEEDBACK
Charlotte Bosman - Van Gogh Museum
Thanks to the Museum Analytics Action Research Project, we were able to make better reports for the
different departments of our organization. Awareness of the importance of online measurement increased
within the organization during the project. Currently we are working on a new website. The learnings from the
project will be taken into account.
Rob Noordhoek - Museum Rotterdam
The Action Research Project enabled us to better implement change in our approach to online media and
understand previous choices by providing various ways to acquire and interpret data. It also introduced us to
possibilities and pitfalls through good (and bad) practices. It established an informal network that encourages
sharing of knowledge, feedback and learning.
Irene Haan - EYE Filmmuseum
Just when the project started, we were in the middle of defining goals and KPI's for our corporate website. The
timing of the project was perfect and made even more clear to us that we needed to define what exactly we
wanted to measure (and what not!), draw conclusions and act upon the results.
Astrid Hulsmann - Gemeentemuseum, Den Haag
The project has given us an insight in the possibilities, which at this point, regarding to Google Analytics, is a bit
overwhelming. As you know, we're following up on this project now, starting with a small but firm basis. The
social media workshop was a lot more familiar and I therefore was able to immediately make a few minor but
important adjustments.
Henny van Haelen - Museum voor Communicatie
To us, Museum voor Communicatie, the project has significantly contributed to the ability to collect and analyze
the results of our online activity. The program also gave us practical information: what kind of social media
posts (should) have better effects than others? It was inspiring to hear how other museums face the same
issues and how they handle these issues.
Museum Analytics - Participants Feedback
Page 35 of 39
Sanne Geraets - Het Scheepvaartmuseum
The Museum Analytics Action Research Project has given us (Het Scheepvaartmuseum) an insight of the
influence of a social media calendar and how to collect and manage the content. Also it has increased our
awareness of the need to monitor our online success carefully, and it has increased our knowledge of Google
Analytics and how to use and interpret the data.
Christine Gündisch - Museon
Using the insights from the ARP, we are now monitoring the statistics of our various online channels much
more intensively. We used the new knowledge for building our new website, which will be live within the next
month. When the new site is online, we can monitor the statistics with google analytics from day 1! :-)
Bonny Van Sighem - Centraal Museum
Participating in the Action Research Project gave us new possibilities to use web statistics in order to better
reach our visitors, both online and onsite. In the overwhelming possibilities of Google Analytics we now know
what works best for us. From 2014 we will use customized reports from GA as a new source for developing
museum strategies. The generous sharing of ideas, forms, doubts, methods and statistics by all participants was
also very useful and much appreciated.
Robin Holland - Stedelijk Museum Amsterdam
The project contributed to be more aware of measuring the museums' online success. Also management is
aware of the importance of online success, and that helped the museum a lot. I can truly say that the online
success and monitoring different questions is higher on the museum agenda.
Sander Pieterse - Naturalis Biodiversity Center
The project and workshop helped us to be more aware of the importance of measuring online success. We
formed a cross-departmental working group, which focuses on defining goals, improving our web analytics
setup, and delegating related tasks.
Radha Pleijsant - Netherlands Institute for Sound and Vision
Measuring online success is higher on the overall agenda of our institute, thanks to the workshops. Moreover,
we have a better insight in the questions we need to ask ourselves in order to meet the right objectives.
Currently we implement our learnings into a large project called: 'Collection online'.
Sylvia Gentenaar - Kröller-Müller Museum
Before the Action Research Project, we received statistics on our corporate website, project websites and
social media but we didn’t know how to interpret them or how to present them to our management team.
The project created awareness of the importance of figures and setting up clear objectives, and helped us to
find the right tools. Currently, we are working on our first reports.
Page 36 of 39
Anat Harel - Jewish Historical Museum
The Museum Analytics Action Research Project has definitely increased our awareness of the need to monitor
our online success carefully, and gave us an insight into what questions should be asked (and can be answered
using a variety of online tools).
Erik van Tuijn - Netherlands Institute for Sound and Vision
For both Beeld en Geluid and the Gemeentemuseum, it’s always been my goal to get smarter about serving our
online visitors. The Action Research Project has definitely served as a pressure cooker for this. With lot’s of
stimulating dialogue between participants and series of solid and useable tips, tricks and formats. And best of
all, the project has offered a roadmap that will keep us getting smarter.
Yvette Lardinois - Zeeuws Museum
The project has personally helped me in analyzing the statistics (digital and on-site data). It has also ensured that
we can better define the goals we want to pursue, and whether they are realistic as an organization. We plan to
use the new acquired knowledge to plan and implement our digital strategy and further develop our digital
annual report.
Hannah Goeman Borgesius - Amsterdam Museum
The Action Research Project has placed the importance of measuring our online activities higher on the
museum. A number of departments in our organization including the media and participation, marketing &
communications and public education departments have began to measure on a regular basis the impact of
their activities online. In the next year, we plan to further improve our reports in order to better inform our
decisions.
Page 37 of 39
7.
CONCLUSION
The Museum Analytics - Action Research Project that took place over a period of approximately 10 months was a unique pilot
project where 16 Dutch museums collaboratively tackled the challenges of defining, measuring and evaluating the effectiveness of
their investment on online technologies.
The project was structured based on a model defined by Stéphane Hamel and followed the footsteps of a similar project developed
by Culture24 in the UK. The Hamel model describes an actionable path towards improving competences in terms of analyzing
online performance. The model defines six areas were improvement is equality important: 'Management, Governance and
Adoption', Objectives, 'Scope', 'Team and Expertise', 'Continuous Improvement Process and Analysis Methodology' and 'Tools,
Technology and Data Integration'. All participants took a survey to benchmark their online maturity following the Hamel
model before and after the workshops took place. The results and detailed conclusions are included in the chapters of this report.
This chapter presents an overview of the participants findings for each area defined in the Hamel model.
The majority of the participants agreed that in order to install an analytic culture in their organization, it is essential that directors
and heads of departments have a deep understanding about the importance of data analysis for decision making. Analytics software
can visualizes online endeavors by which progress can be mapped and goals can be set. Therefore it is easy to act according to its
results. Involving directors and decision makers, is one of the biggest challenges and it has been noticed that once they understand
the relevance, the interest and usage of data analysis rapidly spreads throughout the entire organization.
Performance analysis can only be made once clear goals are defined. Most museums have an organizational mission. However, it is
not always obvious how to define goals or metrics, consistent with the mission of the organization. The metrics traditionally
reported by museums in an annual basis are museum visits, number of activities and financial overviews. For analyzing online
performance, it is important to define specific metrics, such as, number of online visits, number of newsletter subscribers and
number of transactions completed. By reporting on an monthly or weekly basis about specific metrics it is possible to inform
museum departments on action that can be taken in order to improve their overall performance. During the project such metrics
have been discussed and defined. Besides encouraging participants to analyze their own online progress, it was also possible to
benchmark endeavors with others and contextualize numbers that might otherwise appear abstract.
Museums often have a culture of inclusion where they aim to reach as many people as possible. This ambition often leads to a broad
scope in terms of online presence. Most participants were present in several social media platforms albeit not always active. The
Action Research Project focused on three platforms only: the museum institutional website, Facebook and Twitter. The choice of
platforms was based on the fact that these were the platforms in which all participants were most active. Other communication
tools and platforms that the participants were interested in analyzing but were left out of the scope of this project were newsletters
and e-commerce. These tools are great candidates for a future analyze in a possible second phase of the Action Research Project.
Most participants pointed out that the project had a profound impact in terms of developing expertise about data analysis. Several
participants have also created an internal task force that meets on a regular basis in order to discuss the progress of their online
efforts. Both the exercised developed during the project and the periodic meeting established a concrete methodology and have
lead to continuous progress of the participants online strategy.
The project also had an impact in terms of exchange of information across different organizations. As described in the report, the
project was based on a collaborative method were all participants could exchange information and learn from each others‘
experiences. This exchange took place in several forms were participants called each other for direct assistance and by openly
exchanged information within the group by using a dedicated online platform that is still in use today.
Museum Analytics - Conclusion
Page 38 of 39
In terms of tools, all participants are currently using free tools to support data analysis. During the project, best practices were
defined to setup Google Analytics. Recently, Google has launched a new version of Google Analytics known as Universal Analytics.
The new version encourages users to measure not only online metrics but also other relevant metrics that might not necessary be
associated with online activities. This represents a unique opportunity for museums in terms of data integration. For example,
museums can use Universal Analytics to measure and create reports about museums visits while cross analyzing that information
with the numbers of website visits. These new possibilities are yet to be explored and they represent a convincing argument to
launch a second phase of the Action Research Project.
Finally, it is important to understand that online technologies are changing at a rapid pace. It is therefore understandable that skills
and methods to evaluate online success are evolving too. This type of collaborative research encourages participants to keep up
with these changes. It is therefore not surprising that many of the participants have shown interest to continue this format of
collaboration.
This project would have not been possible without the support of the following individuals:
Frank Bergevoet (Rijksdienst Voor het Cultureel Erfgoed), Agnes van Veen (Beeld en Geluid), Jane Finnis (Culture 24, UK), Seb Chan
(Cooper-Hewitt National Design Museum, US), Astrid Hulsmann (Gemeentemuseum), Sander Pieterse (Naturalis), Pieter Vader
(Naturalis), Esther Herberts (Naturalis), Charlotte Bosman (Van Gogh Museum), Bonny van Sighem (Centraal Museum), Nanda
Broekhuizen (Centraal Museum), Irene Haan (EYE Filmmuseum), Martin Buytendijk (EYE Filmmuseum), Petra IJmker (NEMO),
Marjolein Schipper (NEMO), Radha Pleijsant (Beeld en Geluid), Erik van Tuijn (Beeld en Geluid), Hein Wils (at the time Beeld en
Geluid), Johan Oomen (Beeld en Geluid), Yvette Lardinois (Zeeuws Museum), Vera Bartels (Museum voor Communicatie), Henny
van Haelen (Museum voor Communicatie), Hester Gersonius (Amsterdam Museum), Anne-Mirthe Dieudonné (Amsterdam
Museum), Hannah Goeman Borgesius (Amsterdam Museum), Robin Holland (Stedelijk Museum), Matthijs van der Meulen (Q42 for
Stedelijk Museum), Sylvia Gentenaar (Kröller Müller Museum), Lies Boelrijk (Kröller Müller Museum), Sandra Boks (Kröller Müller
Museum), Ilonka Coenraad (Het Scheepvaartmuseum), Sanne Geraets (Het Scheepvaartmuseum), Sinja Bloeme (Het
Scheepvaartmuseum), Anat Harel (Joods Historisch Museum), Kasper van der Laan (Joods Historisch Museum), Maarten Okkersen
(Museon), Christine Gündisch (Museon), Rob Noordhoek (Museum Rotterdam), Doriene Reitsma (Museum Rotterdam), Fransje
Pansters (INTK), David Jonas (INTK) and Rui Guerra (INTK).
Intellectual property
To the extent possible under law, Rui Guerra and Fransje Pansters have waived all copyright and related or
neighboring rights to Museum Analytics - Action Research Project.
Museum Analytics - Conclusion
Page 39 of 39

Museum Analytics

Transcription

Similar documents

1994 - 2000 - . . 1846 . . . Location: http://maps.google.com/maps?ll

Baby Shower Packages - Children`s Museum of Phoenix

TEXT ANALYTICS

TAC Update 2/19/16

PIONEERS AND DRIFTERS, COWBOYS AND INDIANS

HP-U-505 MSI.indd

Memberships from Coast to Coast and Everywhere In Between

a Capitol adventure - The Jefferson, Washington DC

Arnhem Open Air Museum Holland in one day!

Saturday, June 16th 2:00 to 5:00 pm Tickets: $15