Asia Art Museum on Google Art Project: A Web Log Analysis of the National Palace Museum’s Participation

Shao-Chun Wu, Taiwan


In 2011, Google introduced an art museum-oriented online platform called Google Art Project. Since then, Google Art Project has drawn attention and discussion in museum communities. In 2012, after evaluation and internal discussion, the National Palace Museum (NPM) decided to join Google Art Project in order to enhance the museum's international visibility, meet marketing needs, and provide visitors with new learning resources and other aims. Subsequently, NPM studied and verified the exact effectiveness in marketing, education, and other functions of joining Google Art Project.

Keywords: Google Art Project, National Palace Museum, Web log analysis, Google Analytics

1. Introduction

In February 2011, Google launched the art museum-oriented Google Art Project (GAP), which immediately triggered deep discussion and drew wide attention. It also expanded Google’s business from the text-centric Google Book Search and Google Scholar platforms to the brand-new arena of museums. GAP features innovations that are unusual in traditional museum websites, and it has been very rare in the past that such an information giant has collaborated with global museums. Although it has been quite a while since the launch of GAP, it is still an eye-catching project receiving a lot of attention, and the number of museums participating in the project, as well as the platform’s content, have been growing ever since, along with constant updates of its functionality and screen layouts.

Recognizing that participation in GAP would enhance the National Palace Museum’s visibility and marketing strength, increase the Museum’s website traffic, offer visitors novel access channels to the Museum’s art information, and reinforce the Museum’s educational activities, the Museum in 2012 undertook careful consideration through a number of meetings and discussions, and decided to join GAP’s Phase II, becoming a GAP partner. After joining GAP, however, the Museum needed to analyze the impact of its participation and evaluate the effectiveness of its marketing and educational outreach, so as to make appropriate enhancements to the Museum’s content on the GAP website and better integrate GAP into the Museum’s educational activities and Internet resource planning. As a result, this study takes a Web log analysis approach and uses Google Analytics to analyze Web traffic to the Museum’s content on the GAP website. Web traffic on the Museum’s official website is also included in the log analysis. In addition, visitors’ comments on the Museum’s participation in GAP and the statistics of the Museum’s media exposure are surveyed for complementary data, in an effort to highlight the project’s outcomes and issues after the Museum joined the ranks of GAP. Hopefully, the study’s results can help build a foundation for the Museum’s future in GAP.

To begin, this paper briefly introduces the background of GAP. By way of introducing opinions and views on GAP from various points of view, this paper then analyzes the possible advantages GAP can bring to museums, as well as its potential restrictions, followed by a description of our research methods. The third part of this paper is the analysis of the value of the Museum’s joining the GAP. The final part includes a conclusion on the impact and issues of the Museum’s participation in GAP, with suggestions to other museums for their own involvement.

2. Google Art Project and the museums

The Google Art Project originates from a well-known Google “20-percent project” in which Google’s employees can use 20 percent of their weekly work hours to explore and conceive new projects or ideas of their own initiative (Berwick, 2011). GAP was exactly the product of a new project created by a group of art-loving Google employees who used 20 percent of their weekly work hours to apply the technologies of Picasa, Street View, YouTube, and Google Maps to let museums all over the world cross geographical boundaries and time restrictions to display their art collections to the world through online queries anytime and anywhere (Mediati, 2011). With enormous effort by Google, this concept was finally realized in 2011. Phase I included 17 large museums, such as the Museum of Modern Art, the Metropolitan Museum of Art, and Tate, which provided images of their collections to GAP and allowed GAP staff into their galleries to create interior “Street Views” and high-resolution scans of selected art works. After two years of constant growth, the number of museums participating in GAP exceeded 260 in 2013, and the number of images of museum collections has reached over 46,559. During this period, GAP’s functions and interface were improved and updated many times, and the Art Talk platform, connecting to the Google Cultural Institute, was also established for art exchange and communications among participating museums.

Since its inception, GAP has attracted attention and discussion from the museum circle and other fields, and the opinions and views on GAP vary, with both positive and negative comments. This study has collected and compiled journal articles, newspapers, and online comments reflecting various points of view on GAP, and found that people in various fields have mostly positive views of GAP, though some insist that GAP has its limits. Following are the positive views on GAP and its possible limits:

(1) Positive or affirmative comments

1. From the standpoint of fulfilling museums’ missions and improving museums’ performance:

    1.1 Some people engaged in the museum business argue that the advantages of GAP are that it is free of charge and offers easy-to-use features, which can support museums’ public engagement and fulfill museums’ public service missions (Gordon, 2013). In the meanwhile, museums in GAP can help more people gain access to art information, and in terms of information availability, it stands for a more democratic approach (America, 2012).

    1.2 GAP offers an outstanding Web-page browsing mechanism that is unparalleled by other websites, and its innovative connotations make some museum experts believe that the emergence of GAP indicates a future trend in museum website design: a transformation from the traditional content-centric approach to a context-focused emphasis, whose innovative features will inspire museum Web designers with new ideas (Proctor, 2011). Also, the staff of museums participating in GAP and some museum experts contend that joining GAP does increase the museums’ website traffic and that there have been no signs showing any negative impact on the frequency of visitors to physical museums (Proctor, 2011; Wermuth, 2011; Bickersteth, 2011).

2. From the standpoint of educational and art information:

    2.1 According to many reports and comments, museums participating in GAP have gained one major advantage: a platform that allows Internet users to not only enjoy seeing museum collections from all over the world, but also benefit from educational opportunities that cross geographical boundaries and time limits; they can enjoy seeing the world’s museum collections without having to spend colossal amounts of money and time, and without having to endure crowding in the physical museums (Smith, 2011a ; Pack, 2011; Allan, 2012).

    2.2 In its educational features, GAP also has strengths: take fine art courses, for instance. GAP’s ultra-high-resolution images—gigapixel images—allow visitors to observe every delicate touch of the art work and canvas texture in full detail. Teachers can use the interactive whiteboard to link to GAP for lectures. Such great facilities substantially help fine art education (Grasso, 2011; Smith, 2011a; Allan, 2012). Moreover, GAP content can be used as reference data complementary to traditional art databases such as ARTstor and Williams Visual Resources (TechFest, 2011; Berwick, 2011).

3. From the standpoint of the viewer’s experience and user friendliness:

    3.1 Google has huge number of technical staff and service points around the globe, plus its global influence, which make GAP capable of including museums in every corner of the world in its GAP collaboration—in effect forming a “United Nations of Art” that contains versatile categories and collection themes. This creates one direct benefit: under this huge GAP technological umbrella, Internet viewers can query and view museum collections from all over the world on a single website instead of visiting different museum websites as they used to (Proctor, 2011; Smith, 2011b, 2012).

    3.2 GAP offers cross-border, cross-museum, and cross-material mechanisms that enable users to look for information about desired art objects and then use GAP’s personalized collection functions to build their own personal collections. Although in the past many museums have offered a type of Personal Digital Collection functionality (Marty, 2010), GAP is distinguished in that it allows users to build collections composed of images from a huge number of museums all over the world in a single platform, rather than building separate collections on separate museum Web platforms.

    3.3 Since GAP includes many Google services, such as Google Maps and YouTube, users only need their Google or Gmail accounts to use GAP’s personalization functions, as well as other Google services, in one integrated, consistent user interface. In addition, Google services already play an important role in Internet users’ daily lives, and that makes it much easier for users to connect the GAP experience with their daily Internet activities, without having to learn unfamiliar functions or interfaces of different websites (Proctor, 2011, Allan, 2012). This advantage will make museums contemplate whether they need to have a consistent style and user interface for their websites, so that their users can have easier access to museums’ online resources.

    3.4 GAP provides a vast number of high-resolution images of art works and allows users to zoom in and out on the art works’ images. In addition, GAP staff has taken high-resolution photos for some museums to create ultra-high-resolution gigapixel images that allow users to observe every detail of the art works online. Notably, these details may not be easily, or may never be, observed when you actually stand in front of the objects in the physical museums (Smith, 2011b; Wermuth, 2011). Moreover, GAP integrates Street View functions, allowing users to experience the magnificent, gorgeous views inside the great museums, and this was rarely available in the past (Smith, 2011a). In combination, GAP’s gigapixel and Street View features offer users brand-new experiences not available on traditional museum websites and art image databases.

    3.5 Some critics point out that today’s youth, especially those born after the year 2000, get to know the world in their own way, which is different from previous generations. Young people rely heavily on the Internet, smartphones, and computers, and they download digital content to digital devices (Grasso, 2011). For people engaged in art education, it is necessary to understand the channels that the new generation uses for accessing art information; perhaps the emergence of GAP is a much better opportunity than others for offering art learning resources to the new generation of students. In addition, the virtual experience of viewing of art works online versus the physical experience of visiting the actual museum should not be understood as standing in opposition to each another; in fact, in the world of the new generation, these two experiences may be complementary and may reinforce one another (Grasso, 2011).

 (2) Opinions that GAP has its limits

1. From the standpoint of GAP:

    1.1 Many critics point out that GAP can be restricted by copyright issues; thus, many twentieth-century modern art works and contemporary art works are absent from GAP, and post-war and abstract expressionist art works are also inadequately represented (Proctor, 2011; Smith, 2012; Gordon, 2013). Such a shortage could mislead Internet users in their understanding of art history.

    1.2 As of now, many major museums, such as the Louvre and most Swiss museums, are still adopting a wait-and-see attitude towards GAP, giving GAP’s contents a substantial deficit (Smith, 2011b, 2012).

    1.3 The collections in GAP vary enormously from museum to museum: some museums provide only a dozen object images, while others provide more than three thousand images of art works (Smith, 2011a, 2012).

2. From the standpoint of the GAP experience:

    2.1 Some critics who are more favorable to the traditional ways of viewing art works do not quite appreciate gigapixel and Street View, and even criticize these technological tools. They argue that the gigapixel images enlarge the art works to such an extent that they becomes meaningless in terms of appreciating them as art works: that’s just like a doctor examining a patient’s skin—after all, when the art works were created, they were not intended for such close observation. And then, Street View may distort the spatial sense inside museums due to the filming angles, plus the art works may be blurred from time to time, making the scenes from Street View a bit awkward and strange (Perl, 2011; Ballard, 2012; Visser, 2011). Simply put, in the eyes of some art lovers and museum-goers, the art appreciation experience of physical museums is never replaceable with modern alternatives like GAP; to them GAP only offers a kind of mechanical aesthetics, without the feelings or the thrill that happens only when you stand right in front of the actual art work (Ballard, 2012; Grasso, 2011).

    2.2 Due to its abundant content and huge image collections, it is inevitable that GAP’s system will have bugs or incorrect information (Smith, 2011a). Some GAP users argue that the museum floor maps and navigation are not quite practical in use (Proctor, 2011).

    2.3 Although GAP has successfully gathered huge image collections from many large museums and established the mechanisms for viewing, query, and retrieval, some users point out that GAP actually does not have curatorial direction (TechFest, 2011).

3. From the standpoint of GAP’s operations:

    3.1 Due to Google’s global influence, museums believe that participation in GAP can bring a positive impact to their marketing and open up a more accessible channel to their collection information; however, some critics argue that museums handing their art information over to an international enterprise that is globally dominant like Google may not be a wise way to steward the arts (Visser, 2011).

    3.2 After all, Google is a commercial corporation with pressures to make profits for survival and growth. As a result, some people doubt that Google can provide long-lasting GAP operations, keep investing in GAP’s resources, and expand its contents without commercial considerations. This is something that really matters and deserves a great deal of attention (Proctor, 2011).

    3.3 In the past, Google had legal litigations against organizations like the Authors Guild in setting forth its Google Books search. Some people in museum circles are concerned that they’ll be involved in legal issues if they participate in GAP. In the meanwhile, some also point out that Google currently only contracts with individual museums directly, without the intervention of artist copyright agents such as the Artist Right Society (ARS) for negotiations and authorizations; as a result, each museum has to try on its own to acquire complete rights to art works’ images before posting them on GAP (Gordon, 2012).

Through a compilation and analysis of the positive and negative comments on museums participating in the GAP, we’ve learned the positive side is that GAP fulfills museums’ missions, increases accessibility to art information, promotes traffic to museum websites, brings marketing impact, provides educational resources and tools, offers visitors a novel Internet browsing experience, allows the sharing of art information across museums’ ‘borders,’ and is congruent with younger generations’ modern way of living and learning, which is heavily reliant on the Internet. However, museums should also consider the reality that GAP’s content may not be complete or adequate, and that some art lovers and museum goers prefer the traditional way of appreciating art works: GAP is not in their thoughts.

Of course, the impact of joining GAP has been recognized by many museums worldwide, and that’s why more and more museums are joining its ranks. Nevertheless, museums after joining GAP cannot just leave the images of their collections on the GAP platform without assessing the outcomes of joining the project so they can take necessary measures to make GAP more helpful in the promotion of the museum’s educational mission. Evaluation can only bring positive value to museums’ participation in GAP, and help them benefit from the positive results discussed here. By using the Web log as the primary material for analysis, aided by other complementary materials, this study explores the outcomes of the National Palace Museum’s participation in the GAP and provides the study results as a reference for future planning.

3. Research methods

For a complete exploration of the National Palace Museum’s participation in GAP and its impact, various research methods were used to analyze data from various sources. This study takes Web log analysis as its main research method, combined with analyses of other complementary data, described as follows:

(1) The main research method

For the Web log analysis, Google Analytics was used to analyze the Web traffic on the National Palace Museum’s GAP site in order to obtain the number of visits, visitors’ behaviors, Web page dwell times, visit sources, etc. This analysis allows very direct observation of the impact of the Museum’s participation in GAP. Also included in this study is the Web traffic to the Museum’s own official website for cross-reference and comparison with the GAP platform traffic.

In fact, Web log analysis is regarded as one of the Transaction Log Analysis (TLA) methods, which analyze log file contents generated by the analytics system in order to understand user behaviors, the interaction between the system and its users, and the system’s performance (Jamali, Nicholas, & Huntington, 2005). The TLA concept has a long history: it was developed in the 1960s, but not until the 1990s when websites flourished and website log files became the mainstream of computer records did more people start to analyze Web log file contents. As a result, Web log analysis became one of the common methods for studying Web users’ behaviors in order to improve website design and management (Lin & Hong, 2010; Jamali, Nicholas, & Huntington, 2005).

There are more than one hundred different Web log analysis software packages available on the market. They can be divided into two categories: the ones making direct analysis of log files already generated, and the others that use a page-tagging approach to generate log files for analysis (Voorbij, 2010). The Google Analytics tool is a page-tagging Web log analysis system. In summary, Web log analysis has the following merits and drawbacks:

    1. It can quickly collect a vast volume of multidirectional data for analysis (Fang, 2007), suitable for the study of system performance, user behaviors, and highly evidence-oriented research (Jamali, Nicholas, & Huntington, 2005).

    2. Compared with surveys and interviews, Web log analysis offers a more economical route to data entry and collection, due to the automatic generation of log files. Also, it is not restricted by the study’s subjects and geographical boundaries (Fang, 2007).

    3. Web log analysis is a non-invasive approach, meaning it will not be misled or interfered with by the study’s subjects, and auto-generated data is more objective (Lin & Hong, 2010; Jamali, Nicholas, & Huntington, 2005).

   4. Theoretically, if only the information system keeps running, the log data can be generated non-stop. Therefore, Web log analysis is very suitable for studies that need to collect longitudinal data (Lin & Hong, 2010; Jamali, Nicholas, & Huntington, 2005).

   5. Web request statistics can be interfered with by the cache, thus compromising data accuracy. Spiders, robots, and crawlers can also affect the accuracy of the Web log data (Jamali, Nicholas, & Huntington, 2005).

   6. Web log analysis has no way to determine the users’ motives, emotions, and attitudes. As a result, additional approaches may be needed to catch those intangible features.

  7. On the same website, using different Web log analysis software can get different statistics. In addition, in some circumstances it is very difficult to authenticate the user’s identity with Web log analysis. For example, the same user can log in from different computers, or multiple users can share a single account (Voorbij, 2010).

(2) Auxiliary methods

To have a profound understanding of the opinions of the National Palace Museum’s participation in GAP from people in various sectors in order to make a complete assessment of the impact of the Museum’s participation in GAP, this study collects visitors’ comments from e-mail and Facebook channels, as well as comments from other museums and Internet blogs. The media exposure rates and media participation in the Museum’s activities are also taken into account for the analysis.

In summary, this study takes Web log analysis as its main research approach. Google Analytics is used to analyze website traffic on the Museum’s GAP platform and on the Museum’s official website. As for the auxiliary methods, this study extracts the visitors’ opinions expressed via e-mails and Facebook, and considers the media exposure rate after the Museum’s participation in the GAP, along with the number of media representatives participating in the Museum’s press conferences and activities. The aim is to produce the most complete possible exploration of the impact of the Museum’s participation in GAP. The research methods, analysis tools and subjects, and data collection periods are compiled and listed in Table 1.

Research Method

Analysis Tool and Subject

Data Collection Period

Web Log Analysis Google Analytics 2013/4/21–2013/9/4
NPM’s office website 2011/4/1–2013/4/1
Analysis of relevant materials Viewer opinions expressed via FB, E-mail, newspapers 2012/4/10–2012/4/24
Opinions from staff of other museums 2012/4/10–2012/4/24
The number of media participating in NPM activities and NPM media exposure rate 2012/4/10–2012/9/15

Table 1: Research methods, analysis tools and objects, data collect on periods

4. Study results

 (1) The status quo of the website traffic

On April 21, 2013, National Palace Museum started to use Google Analytics to analyze its GAP platform website traffic. Up to September 4 of the same year, there had been 9,341 visits for a total of 64,366 browsing pages, with 6,401 visitors, 32.6 percent of which were new visitors. Repeat visitors represented 67.4 percent of the total visitors.

The detailed usage report shows that the average visit dwell time is 5 minutes and 21 seconds, and each visitor goes through an average of 6.89 pages, with an average bounce rate of 57.33 percent. According to the global statistics provided by the GAP, in the first season of last year, the visits to the Museum’s content on GAP represented about 1 percent of all GAP visits. Nevertheless, at least up to the end of last year, the National Palace Museum was the most represented museum on GAP from the greater region of China and was the only museum in the region offering Street View and gigapixel services. However, the fact that NPM attained merely 1 percent of the total global GAP visits indicate that there is large room for the Museum to increase its content and improve its offerings on the GAP platform.

Visits Page Views Visitors New Visitors Returning Visitors Bounce Rate
NPM on GAP 9,341 64,366 6,401 32.6% 67.4% 57.33%

Table 2: Overall usage of the NPM GAP platform

 (2) Visit sources

In the analysis of visit sources on the Museum’s GAP platform, visitors were found to have some identifiable characteristics. According to the statistics generated from Google Analytics, Americans make up the largest portion of visitors (20.6 percent of the total), followed by 18.4 percent from Taiwan, 15.1 percent from mainland China, 5.3 percent from the United Kingdom, and 3.8 percent from Japan. Although Americans make up the largest portion of visitors, the most used language is Mandarin: the combination of traditional and simplified Chinese speakers makes up 16.6 percent of total visitors, while speakers of English are only 6.8 percent.

Country Visit Rate
U.S. 20.6 %
Taiwan 18.4 %
China 15.1 %
United Kingdom 5.3 %
Japan 3.8 %

Table 3: Top five countries of origin of visitors to the NPM’s GAP platform

A further analysis of the cities of the visitors shows that the city of Taipei sends the largest portion of visitors (6.2 percent), followed by London (4.2 percent), Portland, Oregon (3.1 percent), and Beijing (2.7 percent). Notably, the American city of Portland unexpectedly has a much higher visitor volume to the Museum’s GAP content than the major U.S. cities of New York and Los Angeles, and this is something that really deserves further exploration. To this end, this study makes a further analysis of the traffic sources of the visitors, along with a comparison with the cities of the visitors. The result reveals that direct links to the Museum’s GAP website provide the majority of sources, followed by links from Google search results, and thirdly referring links to the NPM website from “” (Portland Oregon Public School System, kindergarten to twelfth grade section). Other sources of website referrers include Mobile01, Baidu, Douban, Facebook, Google+, etc., among which Mobile01, Baidu, and Douban are the most popular social websites in Chinese cultural circles.

After a comparison of the cities and visitors, it was found that the No. 3 visitor source (the “” website) is located exactly in the No. 3 visitor city of Portland. This website is educational with abundant resources and content. The top five cities and visitor sources are listed in Table 4. From these results we’ve learned that linking the Museum’s content on the GAP platform with educational websites can greatly increase the Museum’s online visit rate, and this was one of the aims of the Museum’s participation in the GAP. Unfortunately, the statistics reveal that currently there is only one educational website ( that connects to the NPM’s GAP platform.

City Visit Source Visit Rate
London direct 2.3%
Taipei direct 2.2%
Portland 2%
London Google 1.7%
Taipei Google 1.5%

Table 4: Top five cities and visitor sources

(3) The visitors’ behaviors

How do visitors access the Museum’s content on the GAP platform? The visitors’ behaviors are one of the main indicators in our assessment of the impact of the Museum’s participation in GAP and in fulfilling the actual demands of online visitors.

The first part of the behavior is the dwell time and number of Web pages visitors go through. This study found that most visitors stay on the website for a short time: that 62.2 percent of the visits last less than 10 seconds, and only 23.6 percent of the visits last more than 60 seconds (Table 5). Apparently, most visitors just take a glimpse at the Museum’s GAP page and then bounce out. In terms of the number of Web pages browsed, 57.33 percent of visitors view only one page, and only 9.83 percent browse more than ten pages (Table 6). Since most visitors stay for a very short time and browse just a single page, it is reasonable to assume that most visitors jumping to the website come just for curiosity or are temporarily attracted by the news jointly published by the Museum and Google. Few visitors are immersed in the information about Chinese art or absorb it.

Less than 10 Seconds More than 60 Seconds
Proportion to Total Visits 62.2% 23.6%

Table 5: Visit duration

One Page More than Ten pages
Proportion to Total Visits 57.33% 9.83%

Table 6: Pages browsed by visits

When visitors log onto the Museum’s GAP homepage, what Web pages do they browse, and what artifacts are of most interest? This study uses Google Analytics to identify the top fifty most frequently browsed Web pages of the Museum’s GAP website and divides the pages into two categories according to the titles of the Web pages: pages with specific artifacts and the rest (i.e., portal pages or search results). The page views of these two categories were then surveyed, and statistics show that 67 percent of browsing is of the top fifty most frequently viewed pages, while 33 percent is to portal or query pages (Table 7). Apparently, the artifact data including images and text is still the main reason visitors engage in deeper browsing.

Pages with Artifacts Information Pages of Portals or for Queries
Proportion of Top Fifty Most Frequently Browsed Web Pages 67% 33%

Table 7: Page views by categories

The most frequently viewed artifacts are the Curio Box, Picture of the New Year, and Jadeite Cabbage. There are many possible reasons behind the high browsing volumes of these masterpieces. First, the Picture of the New Year was not well known previously, but after National Palace Museum chose it for the GAP gigapixel shooting, its photos were published in newspapers and brought in vast amounts of visitor interest. The Jadeite Cabbage and Curio Box were already highly popular antiquities before GAP; therefore, their high browsing volumes are not surprising. The top five most frequently browsed artifacts are listed in Table 8. These show that the key to increasing the dwell time and page views of art information through GAP is posting interesting artifacts.

Name of Artifact Proportion of Top Fifty Most Frequently Visited Web Pages Rank
Curio Box 20.31% 1
Picture of the New Year 14.86% 2
Jadeite Cabbage 6.6% 3
Traveler Among Mountains and Streams 3.43% 4
Oblation Zun 2.64% 5

Table 8: Top five most frequently viewed NPM GAP Web pages

To find out whether visitors’ identities and backgrounds affect their browsing behaviors on the Museum’s GAP website, this study compares the visits to these two Web page categories with the languages used by the visitors. The result shows that visitors using English have 8.28 percent of browsing pointing to the portal and search pages, and 11.92 percent pointing to pages with specific artifacts, so the difference is minor. On the other hand, visitors using Chinese have 24.76 percent of browsing pointing to the portal or search pages, and 55.04 percent pointing to pages with specific artifacts. The difference is substantial. This fact reveals that language and cultural backgrounds are critical factors affecting visitors’ willingness to go further in browsing the artifacts. The results of this cross-reference analysis of visitor languages versus Web page categories are shown in Table 9.

Proportion of Visits to Portal or Query Pages Proportion of Visits to Artifact Pages
Visitors with English 8.28% 11.92%
Visitors with Chinese 24.76% 55.04%

Table 9: Cross-reference analysis of visitor languages vs. Web page categories

(4) Visitors’ technologies and tools

According to the statistics, visitors to the Museum’s GAP website use various technologies and tools. As many as 91 percent of the visitors use desktop computers, 5.7 percent use tablets, and 3 percent use mobile phones. This demonstrates that most art-loving people and museum goers tend to use desktop computers with larger screens for better experience in viewing the art works; however, the GAP has yet to launch its official app or mobile website, and that could be one reason why tablet PCs and mobile phones are not popular yet in the GAP arena. The most popular operating systems used in the tablet PCs and mobile phones with the GAP is iOS, which makes up 74.8 percent, followed by Android with 24 percent. The visitors’ GAP devices and bounce rates are listed in Table 10.

Device Visit Rate Bounce Rate
Desktop 91.2% 56.6%
Tablet 5.7% 60.9%
Mobile 3% 70.1%

Table 10: Visits to NPM on GAP website devices and bounce rates

The most frequently used browser is Chrome, which makes up 53.8 percent of all the browsers used for the GAP, followed by 14.8 percent for Safari, 13.4 percent for Firefox, 13.3 percent for IE, and 1.2 percent for Android. Notably, visitors with mobile devices have an average bounce rate of 64.47 percent, higher than the bounce rates of 57.33 percent for overall visitors and 56.6 percent for visitors with desktop computers. But visitors with mobile phones have a bounce rate as high as 70.1 percent—evidence that mobile devices may not be ideal tools for viewing art works on GAP, and the display of art works as well as search results on a small screen may not satisfy the demands of either art-loving people or museum goers.

(5) Comments on National Palace Museum’s participation in the GAP and the number of people engaged in relevant events

The National Palace Museum is one of the most representative art museums in Taiwan and the greater China region. It has a huge number of visitors to its physical site every day, and its website is also heavily browsed. Consequently, its every move draws great attention from the public. Since the Museum’s participation in GAP, there have been lots of comments from its website visitors, physical site visitors, and staff of other museums via the Museum’s official Facebook page, e-mail, seminars, and newspapers. These comments will be helpful to the Museum in the course of improving its GAP content and understanding its visitors’ needs. Both positive and negative comments are described as follows:

  5.1 Positive and supportive comments

    5.1.1 People believe that the Museum’s participation in the GAP offers another channel to the art works, without having to squeeze in among the crowds at the physical site.

   5.1.2 Some experts and scholars say in meetings that the Museum’s participation in  GAP stands for the collaboration between the cultural and scientific industries, and offers a model for pioneering new technologies.

   5.1.3 Staff from other museums believe that the Museum’s participation in the GAP is worth emulation by other Taiwan museums.

   5.1.4 Some website visitors are curious about and interested in knowing how the paintings are captured with gigapixel and the effect of Street View.

   5.1.5 A few website visitors just press the “Like” button on Facebook; the total count is  about 225 clicks.

  5.2 Criticism and suggestions on improvement

   5.2.1 Some website visitors complain that the quality of images the Museum provides to the GAP are not good enough.

   5.2.2 Most people are positive about the Museum’s participation in the GAP, but the most frequent criticism is the very low quantity of images the Museum provides to GAP: there are only eighteen pieces.

Regardless of the comments, National Palace Museum’s participation in the GAP has brought about amazing media exposure and marketing impact both in Taiwan and mainland China. According to some statistics, the press conference announcing National Palace Museum’s participation in GAP drew a total of fifty media companies to attend, including all large TV stations, newspapers, network news, and radio stations, and there were 113 news reports detected in 48 hours. In contrast, the Museum’s other press conferences announcing other digital projects in the same year drew only a total of twenty media companies to attend.

GAP Press Conference Press Conferences for Other Digital Projects
Number of Attending Media Companies 50 20 (estimated)

Table 11: Number of media representatives attending NPM press conferences – Google Art Project vs. other digital projects in the same year

(6) Analysis of visitors volumes on National Palace Museum’s official website

One of the main purposes of the Museum’s participation in GAP was to increase the accessibility of art information to the public and attract foreign visitors to the Museum’s official site by stimulating their interests in the Museum’s artifacts while they browse the GAP website and read about the Museum’s art works. The Museum aims to achieve better recognition overseas. Therefore, this study also compares visits generated before and after the Museum’s participation in GAP. The National Palace Museum joined GAP in April 2012. One year before joining its ranks, the Museum had 2,622,964 visits, of which 12.11 percent were from the United States. One year after joining the ranks, however, the Museum has already reached a total of 2,710,664 visits (0.03 percent growth), and about 22.28 percent came from the United States. It is evident that after joining GAP, the National Palace Museum has seen enhanced international visibility, increased official website visits, and stronger visibility for its art information. The number of visits a year before and after the Museum’s participation in the GAP are shown in Table 12.

Total Visits to NPM Official Site Percentage of Total Visits by Visitors from U.S.
2011/4/1-2012/3/31 2,622,964 12.11%
2012/4/1-2013/3/31 2,710,664 22.28%
Growth Rate 0.03% 83%

Table 12: Visits to NPM official website before and after NPM joining the GAP

In addition, researchers have discovered that a year after the Museum had joined the GAP, links from the Museum’s official website to the GAP website were clicked 3,012 times. However, we only have incomplete data on website traffic to GAP over the same period; there are only about five months of data for the period from April 21, 2013, to September 4, 2013. We estimate 22,418 visits to NPM’s GAP platform after a full year of joining the GAP. This particular result indicated that not only NPM website’s traffic had increased after NPM joined the GAP, both of GAP link on NPM website and the NPM content on GAP had contributed to the overall GAP global traffic. Thus, NPM joining the GAP is a win-win situation for both parties in the web-traffic aspect, GAP had brought considerable traffic to NPM and NPM had also made certain contribution to the GAP’s overall traffic.

5. Conclusion

The purposes and expectations of NPM joining GAP were diverse: by joining GAP, NPM hoped to promote its art content, provide educational resources, increase NPM Web traffic, market the Museum’s image, elevate its international exposure, understand the behavior of Web visitors, and various other goals, as well as inspire future website design to be more innovative and develop novel ideas for information services. From the above Web log analysis and various other complementary methods, this study found numerous benefits and trends worth noting after NPM joined GAP:

1. The marketing results of NPM joining GAP were superb and had high public support.

Local Taiwanese citizens showed great support for NPM joining GAP, and the domestic news value and exposure were also high; therefore, the Museum’s marketing results were overwhelming positive.

2. From the standpoint of art promotion, the Web traffic was relatively low, and users did not go in-depth during their visits.

Analysis from various data also showed that, despite being the museum of the Greater China Region, Web traffic to NPM on GAP did not seem to be high. Furthermore, the majority of the visit sessions were short with only a few pages visited. Thus, there is still room for improvement in the promotion and traffic increase of GAP for NPM.

3. Cultural background is still a key element in users’ decision to visit the NPM’s Web pages on GAP.

Despite the fact that the majority of visitors were from North America, the most-used language of the visitors was actually Chinese. With the help of data cross-analysis, we also learned that Chinese-speaking visitors had visited certain artifacts’ pages more frequently than English-language users. It’s clear that cultural background determines whether or not visitors will visit NPM Web pages on GAP and engage in in-depth browsing of NPM artifacts.

4. It was a win-win situation for NPM to join GAP given that the Web traffic of the NPM official website has grown steadily, and NPM has also contributed to the overall GAP web traffic.

The study indicated that not only did NPM Web traffic grow continuously one year after the museum joined GAP, but a near 83 percent increase was also seen from the United States. The NPM website also accounted for approximately 3,359 visits to the overall GAP traffic.

5. Mobile devices may not be the best vehicle for viewing art.

Data showed that tablet and mobile phone users accounted for a very small percentage of NPM web views on GAP, and the bounce rate is somewhat high. A possible reason may be that while viewing art pieces on smaller screens, users did not experience the kind of comfort and aesthetic that viewing art pieces should provide. Another reason may be that art and museum lovers are more accustomed to viewing art with bigger screens on their desktops.

6. The key to drawing visitors to NPM on GAP lies in the popularity of the artifacts provided to GAP. For instance, among the artifacts provided to GAP, most of the popular ones are already very well-known among NPM artifacts.

7. The results of NPM joining GAP also indicate that, as museums joined GAP, Web traffic could receive a dramatic boost if it were integrated with other educational resources online and other e-learning websites. For example, a considerable number of visits to NPM on GAP were from a Portland learning website ( This showed that in order to achieve the educational aims of museums joining GAP, their content should also be integrated with existing educational resources online, or educational offerings utilizing information on GAP should be created.

8. Visitors’ feedback also revealed that some of the NPM content on GAP suffered from poor image quality, and a majority of criticisms acknowledged that the NPM content provided to the GAP was far from sufficient.

In summary, based on the analysis of NPM joining GAP, the results have been extremely favorable in the areas of museum marketing and news visibility. The project was also beneficial to the NPM’s official website traffic, in particular seeing an increase in visits from the United States. From observation of the visitors’ behavior, analysis was done of visitors’ backgrounds, languages, and browsing tendencies; it was discovered that cultural background is still the dominant factor in whether users visit NPM pages on GAP. Moreover, the study also found that visit volume on mobile devices is relatively low, and artifact popularity was key to visitors’ browsing habits. Improved results can be made by integrating with educational resources online. Furthermore, NPM on GAP can increase Web traffic in the future and respond to various problems and shortcomings by connecting the content on GAP with related domestic or overseas online educational resources, or by encouraging teachers to utilize resources on GAP to enhance their teaching materials. Lastly, NPM should also attempt to replace sub-par quality images in the future, as well as expand NPM content on GAP.


Allan, R. (2012). “Google Art Project : A worldwide art jaunt at your fingertips.” The Education Forum, 38(2), 36–39.

America. (2012). “Googling a masterpiece.” America, 206(15), 4.

Ballard, S. (2012). “The (new) accident of art.” Consulted July 12, 2013.

Berwick, C. (2011). “Up close and personal with Google Art Project.” Art in America, 99(4), 23–24.

Bickersteth, J. (2011). “Museums and the web–In praise of Google Art.” Consulted May 16, 2013.

Fang, W. (2007). “Using Google Analytics for improving library website content and design: A case study.” Library Philosophy and Practice, 9(3) (June), 1–17.

Gordon, L. (2013). “It’s Google, but is it art? Museums wonder whether they should open their galleries to digitizing.” ABA Journal. Consulted May 16, 2013.

Grasso, M. (2011). “Is the virtual gallery robbing us of our real experiences or simply bringing the real closer to us?” Art Write, 45, 16–18.

Jamali, H. R., D. Nicholas, & P. Huntington. (2005). “The use and users of scholarly e-journals: A review of log analysis studies.” Aslib Proceedings: New Information Perspectives, 57(6), 554–571.

Lin, S. C., & M. C. Hong. (2010). “A web metrics study on Taiwan baseball wiki using Google Analytics.” Journal of Educational Media & Library Sciences, 47(3), 343–370.

Marty, P. F. (2010). “My lost museum: User expectations and motivations for creating personal digital collections on museum websites.” Library and Information Science Research, 33(3), 211–219.

Mediati, N. (2011). “Google’s Art Project: The tech behind the scenes.” PC World, 29(4), 16.

Pack, T. (2011). “The Google Art Project is a sight to behold.” Information Today, 28(5) (May), 42.

Perl, J. (2011). “What the Google Art Project does-and doesn’t-mean for art.” New Republic. Consulted May 16, 2013.

Proctor, N. (2011). “The Google Art Project: A new generation of museums on the web.” Curator: The Museum Journal, 54(2), 215–221.

Smith, R. (2011a). “New way to ogle great art: Google it.” The New York Times, p. 15.

Smith, R. (2011b). “The work of art in the age of Google.” The New York Times, p. C1. February.

Smith, R. (2012). “An online art collection grows out of infancy.” The New York Times, p. C4.

TechFest. (2011). Google Art Project. Consulted April 10, 2012.

Visser, J. (2011). “How the Google Art Project might revolutionize the physical museum experience.” Consulted May 17, 2012.

Voorbij, H. (2010). “The use of web statistics in cultural heritage institutions.” Performance Measurement and Metrics, 11(3), 266–279.

Wermuth, S. (2011). “Google a great painting.” The Guardian Weekly (February), p. 36.

Cite as:
S. Wu, Asia Art Museum on Google Art Project: A Web Log Analysis of the National Palace Museum’s Participation. In , N. Proctor & R. Cherry (eds). Silver Spring, MD: Museums and the Web. Published October 4, 2013. Consulted .

Leave a Reply