An Investigation of Web Presence and Business Intelligence of Australian Game Development Companies Using Webometrics Indicators

Siu Man LUI, School of Maths, Physics and Information Technology, James Cook University [HREF1], Cairns, . Email: carrie.lui@jcu.edu.au

Calvin Chun YU, Department of Information Systems and Management [HREF2], Business School [HREF3], The Hong Kong University of Science and Technology [HREF4], Email: calvinyu@ust.hk

Abstract

Web presence is an important intangible asset, especially for the innovative or creative industry. In order to investigate the web presence of Australian game development companies, a list of 80 Australian game development companies' websites was compiled. We developed an automated tool to execute the advanced search functions of the commercial search engine to obtain webometrics indicators for each website in the list. We collected webometrics indicators such as age, size (number of pages), visibility (number of external in-links) and hubness (number of external out-links). Evolution of the size of the companies' websites over the past 10 years was examined for studying the trend of web presence of the game development companies. The countries and languages of the web pages in-linked to and out-linked from the companies' websites were used to discover the international linkages and exposures of the companies. This study demonstrated a cost-effective method for discovering and monitoring business intelligence and trends of an industry using the Internet.

Introduction

Web presence is an intangible asset for a company. Web links on the company's website form the network that connects businesses, customer, suppliers and investors. The Web has provided a fertile ground for business intelligence research and there is a growing body of literature on this topic. Webometrics is the study of the quantitative aspects of the construction and use of information resources, structures and technologies on the Web (Björneborn & Ingwersen, 2004). Over the past few years there has been considerable interest in whether the techniques of citation analysis could be directly applied to the Web for extracting useful information. Researchers have studied the relationships of webometrics indicators with scientific research activities among universities (Thelwall, 2005), ranking of electronic government websites (Petricek, Escher, Cox, & Margetts, 2006), as well as performance and competitions among companies (Vaughan & Wu, 2004). Prior studies suggested that useful information can be uncovered from the link count and link topology. For example, link counts of university website were found to correlate with the universities' research ratings (Thelwall, 2005); links to commercial web sites were found to correlate with business performance measures of companies (Vaughan & Wu, 2004).

This paper aims to investigate the web presence and impacts of the websites of Australian game development companies. The electronic game industry is a significant contributor to the Australian economy - the amount Australians spent on electronic game has outstripped the amount spent on film. Local game development companies now employ more than 1600 people around the country, generating exports worth more than $110 million each year (House of Representatives, 2004) . Web presence is considered as an intangible asset for a company because it allows a company to market itself to potential customers and investors (Kandunias, 2000). This is especially true for game development companies whose ultimate customers are mostly members of the Internet community. It is more likely for their investors to approach websites as the first step to learn about the companies. Game development companies develop products that entertain people and websites are naturally their first stage of demonstrating their entertaining capabilities. Given the special characteristics of this industry, understanding the status and impact of their web presence is important for these companies in assessing their web strategies. Hence, the results of this study are of high interest and importance for the industry. In this study, the age, size (number of pages), visibility (Aguillo, Granadino, Ortega, & Prieto, 2006; Thelwall, 2005) (number of external inlinks) and hubness (Petricek et al., 2006) (number of external outlinks) of the websites of the Australian game development companies are measured. Evolution of the size of the companies' websites over the past 10 years was examined for understanding the trend of web presence of the game development companies. The countries and languages of the web pages in-linked to and out-linked from the companies' websites were used to discover the international linkages and exposures of the companies. Additional data related to the inter-linkages among the websites of the companies and the worldwide top 20 game publishers (Gamasuta, 2006) were collected for exploring the collaborations among these parties.

Methods

We considered a list of 80 Australian game development companies' websites compiled from the directory of the Game Developer Association of Australia (E3, 2006) and a major online game developer community website [HREF5]. The age of the websites was determined by searching for the Internet Archive [HREF6]. The earliest data that the website appeared in the Archive was used as the estimated age of the website. In the sample, 4 domains were not indexed in the Internet Archive, therefore the creation dates of these domains determined by Whois [HREF7] were used instead. These dates were then converted to the number of years from 1 March 2007, the date that data collection took place, and used as the age of the website for correlation analysis in the later section.

In order to collect the data for the size and link counts of the websites, we developed a software to retrieve the required data from multiple search engines. To ensure the reliability of the data collected, 4 search engines were used in this study. They are AltaVista, Google, MSN, and Yahoo. The use of multiple search engines can avoid the possible bias of a particular search engine. Moreover, the three search engines Google, MSN and Yahoo covered 85% of search shares (Danny Sullivan, 2006), this ensures the validity of this study. Also, two rounds of data were collected by executing the same search at the same search engine on two different dates with a 2 week interval. Data from the two rounds were combined and averaged. This was shown to be beneficial in a previous study (Vaughau, 2004).

Size of the websites refers to the number of pages indexed by the search engine for the domains of the game development companies. The search strategies used to obtain the size of the websites are presented in Table 1. Size of the website reflects the extent of web presence.

Table 1 Examples of Search Strategies

Note: xx: domain name of the game development companies. yy: domain name of the Top 20 game publishers.

There are two types of in-links, the external in-links and total in-links. The former are those originated from websites outside the domain in question while the latter include all links that point to the domain in question regardless of the origin of the links. Because the total in-links include navigational links of the website in question, which do not represent quality or impact of the domain being linked to, it has been suggested that external in-link count is a better measure than total in-link count for visibility of a website (Vaughan & Thelwall, 2005). Similarly, only external out-link count is considered in this study to determine the hubness of the website.

The 10 search strategies shown in Table 1 involved 67 searches for each domain. In addition, the external in-links and external out-links searches on MSN and Yahoo were also carried out using additional constraints to determine the distributions of the languages and top level domain of the web pages linked to the game development companies' websites. Since MSN and Yahoo supports the search for 37 languages and 40 languages respectively, another 71 searches were run for each domain to determine the languages distribution of the in-links. To determine the distribution of the top level domains of the web pages linked to the game development companies' domains, 250 searches were conducted for each domain. Therefore, there were total 388 searches carried out for each domain, and hence, a total of 31040 searches for the whole study. The online help websites of the search engines [HREF8], [HREF9], [HREF10] and [HREF11] provided details of how to use the advanced search functions applied in this study.

Results

Age of the sampled websites

While there are four companies' websites aged over 10 years, over half of the companies' websites are less than 5 years old. Over 25% of the websites have been established during the period from January 2005 to March 2007. Although the first game development company in Australian was founded for over 20 years (Game Developer Association of Australia, 2006) , and there are several in late 80s, the earliest web presence among these 80 game development companies was found in 1996. Victoria , New South Wales and Queensland are where more established game development companies can be found. These states are also where the highest number of game development companies can be found. Figure 1 shows the cumulative number of game development companies' websites found in each year from 1998 up to March 2007. The descriptive statistics of the age of the websites are shown in Table 2.

Figure 1: The numbers of game development companies' websites exist from
1998-2007. (Data source: www.who.is and www.archive.org)

Table 2. Age of the websites of the 80 Australian game development companies

Size of the sampled domains

Table 3. Size of the Websites of the 80 Australian Game Development Companies Determined by Different Search Engines.

Note: xx: domain name of the game development company.

The size of a website refers to the number of web pages indexed by the search engine for that domain. Game development companies in Queensland have the largest total number of web pages. The results are consistent across all search engines. Surprisingly, game development companies in Victoria have a relatively small number of pages per website even though it has the large number of game development companies and the 5 largest game development companies as indicated in a recent industry profile report (Game Developer Association of Australia, 2006). Further examinations of the websites of the game development companies in Victoria and Queensland found that one of the game development company in Queensland have a very large website. That website has 37,000, 8,698, 20,500 and 21,600 pages indexed by Google, MSN, Yahoo and AltaVista respectively. For the websites of the game development companies in Victoria, the most established company is a studio for a global company. The website of this game development company has a relative small number of pages, because detailed information about the products was hosted on the website of the global company. Size of the website will be used together with the external in-link count to determine the impacts factor (Ilan & Peritz, 2002) of the website.

Although the absolute numbers of pages of each domain indexed by each search engines are different, the results from different search engines are highly correlated. The Spearman correlation tests rather than the Pearson correlation test was used because the frequency distributions of the size, in-links and age of the websites were skewed. The correlation between the age of the websites and its size, in-link count and in-link to home page count are statistically significant. However, the correlation is relatively small.

Table 4. Spearman's rho Correlation Coefficient between the size and age of the websites.

** The correlation is statistically significant .01. * The correlation is statistically significant .05.

Figure 2: The numbers of game development companies' websites and number of pages
indexed by AltaVista from 1998-2007.

We further examined the number of page indexed by year by the data retrieved from AltaVista with data range constraints. Figure 2 presents the number of the sampled websites and number of pages of the sampled websites indexed by AltaVista from 1998 up to March 2007. It shows that, the number of pages in 2006 is ten-fold of that of 2005. Moreover, almost 2/3 of the websites are created within the first three months of 2007. This is a fascinating result. Further research should be conducted to explore how the game development companies use their websites, and what are the major components and functions of their websites.

Visibility (number of external in-links) of the sampled domains

The results shows that websites of the game development companies in Victoria have the largest number of external in-links linked to their websites, and also the largest number of in-links linked to home page of their websites. Table 5 shows the results of the in-link analysis, and the averaged Web Impact Factor of the websites. Web Impact Factor is defined as the number of pages with a link to the website and the number of pages in the website (Noruzi, 2006). The results of the averaged Web Impact Factors are consistent with the in-links counts. These indicated that the websites of the game development companies in Victoria has a relatively higher impact factor than those of other states. Prior studies found that the number of in-links linked to a company's website had a significant relationship to the business measures of the company, such as gross revenue, profit and R&D expense (Vaughan & Wu, 2004). While the current study found that the relatively high impact factor of the websites of the game development companies in Victoria, further investigation of the relationship of this to the business performance measurement is urged.

Table 5: In-link Analysis and Web Impact Factor of the Websites of the 80 Australian Game Development Companies

Note: xx: domain name of the game development company. yy: domain name of the Top 20 game publisher

Contribution of the in-links by different languages and different top level domain are shown in Table 6. Over 80% of the web pages in-linked to the game development companies' websites are English websites. Among the non-English websites in-linked to the game development companies' websites, German websites are the most common. The results are consistent when we examine the contribution of in-links by different top level domains. Besides .com and .au domains, websites from .de (country level domain of Germany) contributed the highest percentage of in-links. The distribution of in-links by different languages and top level domain provided useful information to understand the potential market of the game development companies. It also indicates the extent of exposure of the companies and their products in different countries. For example, if the products of the game development companies is reviewed by an overseas online newspaper or magazine, it is very likely that the online newspaper and magazine will contain a hyperlink, in-linked to the game development company websites.

Table 6: Languages and Top Level Domain Distribution of the In-links.

Hubness (number of external outlinks) of the sampled domains

The number of out-links from a site is related to its hubness. In political science terms, hubness could be seen as a measurement of the extent to which an organization 'collects' information from the outside world, by providing users with links to other sources of information (Petricek et al., 2006). In a commercial website, out-links could indicate collaboration relationships among companies. An out-link could also be the link to online media about the news of the company. Table 7 shows the out-links statistics of the sampled websites. The websites of the game development companies in Queensland showed the highest number of out-links. However, when we normalized the number of out-links against the number of companies in the state, game development companies in the Canberra Australian Capital Territory have the highest number. By looking at the external out-links from the game development companies websites to the top 20 game publishers' websites, we found that game development companies in Victoria have more out-link to the top 20 publishers, the results are consistent even after normalizing the out-link counts against the number of companies in the state.

Table 7: Out-link Analysis of the 80 Australian Game Development Companies

Note: xx: domain name of the game development company. yy: domain name of the Top 20 game publisher

Conclusions

This paper reported the results of a preliminary study of the Australian game development companies' websites. Web presence of the websites were studied using the webometrics indicators of the websites. This study demonstrated a non-obtrusive and cost-effective data collection method without transgressing ethical boundaries or seeking multiple permissions of web crawling. The next stage for future research will be to compare our measurement against results of business performance measurement of those companies. Given that the Web is commonly used by companies to disseminate and collect information, understanding the relationships between the webometrics indicators and business performance measurements will be useful for gathering business intelligence from the Web and Web mining.

References

Aguillo, I. F., Granadino, B., Ortega, J. L., & Prieto, J. A. (2006). Scientific Research Activity and Communication Measured With Cybermetrics Indicators. Journal of The American Society for Information Science and Technology, 57 (10), 1296-1302.

Australia, G. D. A. o. (2006). Australian Electronic Game Industry Profile November 2006. Available Online [HREF12].

Björneborn, L., & Ingwersen, P. (2004). Toward a Basic Framework for Webometrics. Journal of The American Society for Information Science and Technology, 55 (14), 1216-1227.

Danny Sullivan. (2006). comScore Media Metrix Search Engine Ratings. Available Online [HREF13].

E3, G. A. (2006). 100% Creative Juice. Available Online [HREF14].

Gamasuta. (2006). Game Developer's Top 20 Publishers. Available Online [HREF15].

Ilan, J. B., & Peritz, B. C. (2002). Informetric Theories and Methods for Exploring the Internet: An Analytical Survey of Recent Research Literature. Library Trends, 50 (3), 371-392.

Kandunias, C. (2000). Accounting for Intangible Assets - Web Presence: The Birth of a New Asset. Retrieved 1 March 2007, Research Paper Series of Flinders Business School. Available Online [HREF16].

Noruzi, A. (2006). The Web Impact Factor: A Critical Review. The Electronic Library, 24 (4), 490-500.

Petricek, V., Escher, T., Cox, I. J., & Margetts, H. (2006). The Web Structure of E-Government Developing a Methodology for Quantitative Evaluation. Paper presented at the The 15th international conference on World Wide Web, Edinburgh, Scotland.

Representatives, H. o. (2004). From Reel to Unreal: Future Opportunities of Australia's Film, Animation, Special Effects, and Electronic Game Industries.Available Online [HREF17].

Thelwall, M. (2005). Scientific Web Intelligence: Finding Relationships in University Web. Communications of The ACM, 48 (7), 93-96.

Vaughan, L., & Thelwall, M. (2005). A Modeling Approach to Uncover Hyperlink Patterns: The Case of Canadian Universities. Information Processing and Management, 41 (2), 347-359.

Vaughan, L., & Wu, G. (2004). Links to Commercial Websites as a Source of Business Information. Scientometrics, 60 (3), 487-496.

Vaughau, L. (2004). Exploring Website Features for Business Information. Scientometrics, 61 (3), 467-477.

Hypertext References

HREF1

http://www.jcu.edu.au

HREF2

http://www.bm.ust.hk/ismt

HREF3

http://www.bm.ust.hk

HREF4

http://www.ust.hk

HREF5

http://www.sumea.com.au/sdevelopers.asp

HREF6

http://www.archive.org

HREF7

http://www.who.is

HREF8

http://www.altavista.com/help/search/syntax

HREF9

http://www.google.com/help/operators.html

HREF10

http://search.msn.com/docs/help.aspx?t=SEARCH_REF_AdvSrchOperators.htm

HREF11

http://help.yahoo.com/help/us/ysearch/tips/tips-04.html

HREF12

http://www.gdaa.com.au/docs/Industry_Profile.pdf

HREF13

http://searchenginewatch.com/showPage.html?page=2156431

HREF14

http://www.gdaa.com.au/e3_2006/AustraliaGames_Directory.pdf

HREF15

http://www.gamasutra.com/features/20070108/wilson_01.shtml

HREF16

http://www.ssn.flinders.edu.au/business/research/papers/00-3.htm

HREF17

http://www.aph.gov.au/house/committee/cita/film/report/report.pdf

Copyright