Effect of the Advent of the World Wide Web, the Internet on Business - Coursework Example

Data for Internet Research – A Report Introduction Information and Communication Technology (ICT) has made rapid advances in the last two decades. The unprecedented developments in ICT have enabled almost every sphere of business, work and life to take advantage of the technology. ICT has pervaded every sphere of life in the form of the now ubiquitous computer system and the Internet connectivity. All these have resulted in far-reaching changes in the way we live, the way we work and the way we conduct business. The advent of the World Wide Web and the Internet has revolutionized the way business is conducted bringing in the concept of what is known as e-business. E-business is defined as “the conduct of business transactions and activities using in large part electronic means and typically involving use of the Internet and the world-wide web” (Clegg, Icasati-Johanson & Bennett, 2002, p. 245). As more and more people get access to the Internet and World Wide Web on a global scale, more and more businesses naturally evolve from their former brick-and-mortar way of functioning to e-businesses. A dominant part of the global customer base is fast turning into people who do not physically visit stores to make their purchases, but do so from the comforts of their homes, Internet kiosks or cafes or even offices through the Internet. Even if they do physically visit the stores, they first make their buying decisions on the basis of the information they glean from the relevant web sites that they browse on the Internet. It is in the context of this vastly changed business scenario and the radically altered methods and means of buying and selling that the approach to market research needs to be reanalyzed and rethought of. The conventional way of conducting market research based on physical survey of customers and businesses can neither be justified nor effective in the digital age. Online marketing intelligence and strategic planning has to rely on online data or Internet data – the data that ICT applications can gather from the footprints of potential customers as they browse through multitudes of web sites and pages in search of the products and the product information that they require. Approach of the Paper Mullarkey (2004, pp. 42 – 58) reviews the data types on the World Wide Web for conducting audience analysis and making strategic plans for any business concern. The paper attempts to identify the appropriate web data type to use in different marketing research analysis. An overview of the paper reveals that Mullarkey approaches data relevancy from different angles. The primary concern is whether the data is actually reflective of the trends and preferences of the actual customer browsing the Internet; from a second angle, the approach is shifted to level of attraction of web sites and their content; finally, the angle of focus is shifted to what kind of a response a particular web site would be able to accord to marketing ads placed on its various pages. Mullarkey examines the technology and methods utilized in garnering the two different data types. A comparative analysis of the technology applications is attempted to bring out their respective advantages and disadvantages from different perspectives such as data objectivity, customer identification, extent of data details, etc. Mullarkey follows up the theoretical analysis with interviews with data providers of the two different data types in a bid to complement the theoretical findings of the paper from a practical point of view. In its conclusion, the paper comments on the appropriateness of the two different data types for different types of studies depending on the objectives and the different conditions involved. There are benchmarks in other media with respect to data for strategic planning: for example, analysis of television viewing use Peoplemeter data, while the print media is governed by circulation figures that are independently audited. The main objective of the paper is to evolve some kind of standardization in Internet data type selection for conducting online marketing research. Defining Web Data Types The paper reports essentially two different types of web data – site-centric data and user-centric data. Site-centric data is the data that is generated and captured at the web site under consideration. The collection of site-centric data is essentially done at the web site itself, or in other words, certain ICT software or applications capture the data as the user accesses the web site. User-centric data, on the other hand, concentrates on adopting an approach that is based on the users and not on the web site per se. Site-centric data is concentrated around a specific web site. The data is generated as users access the site. When a user accesses a particular web site, the access is recorded and tracked by the web server in which the web site is hosted. Such access noting and identification is done through a process known as logging in a web server file known as the ‘log file’. A ‘hit’ is recorded by the log file every time a user accesses the web site. The log file can record and track all the movements of the browser within the web site giving details as pages visited, ad banners accessed, time spent on specific contents, etc. This type of data will therefore provide invaluable information on the relevance of the particular website, the preferences and choices of the browsers who visit the site. Though limited user access information such as IP addresses can be captured in site-centric data, many sites employ a variety of methods such as user registration for more detailed identification and labeling of users. Only the web site under consideration and the browsers are involved in generation of site-centric data. There is no role for any third party. A single-website-multi-user is adopted in site-centric data. The data is generated for a single web site, but there is no limitation the user who accesses the web site. In user-centric data, the data is collected by a third party which identifies a sample of web users or browsers as representative of the entire population under consideration and tracks the browsing behavior of the sample across a range of web sites that they may choose to access. The approach in this case if turned around to a multi-website-sample-user one. Information about unique identified individuals is provided over time across multiple web sites The user-centric data provider informed that it conducted 500 phone interviews per month for empanelment of users in samples. The consents of the individual members in the sample are obtained, and appropriate software are installed in their personal computers at home to track their movements across the World Wide Web. The members are identified through unique registration usernames. In case of families, each member has his or her own access user name in the proprietary tracking software so that each member can be identified individually. User-centric data is therefore not restricted to a single web site but is restricted to a sample of users. Working out the Pros and Cons In the paper, Mullarkey has shown that both the data types come with their own advantages and disadvantages. Mullarkey establishes these advantages and disadvantages not only from analyses of the properties of the data and the technologies behind their generation and capture, but also from the practical experiences of the data providers which are collated in the form of interviews. Site-centric data is much less costly as the need for subscribing to any user-centric panel data service does not arise. Site-centric data can give much more details than user-centric data for individual web sites as the movement of the browser can be tracked click-by-click across the links and pages of the website. These details provide a high level of strategic value to the data. Such a level of details is however not possible commonly in user-centric data that is restricted to either domain-level details or to pre-identified specific pages or ad banners at the most. The level of details, as in any conventional market research, can be an inverse function of objectivity. Highly detailed web data may lack objective; and completely objective data may lack sufficient depth. User-centric data tends to retain objectivity as it is provided by independent audited third parties, whereas in site-centric data the web site management may conceal or refuse to provide data that is considered too sensitive. User-centric data providers are however trying to overcome the lack of details through the use of new applications such as the Bannertrack service provided by Nielsen/Netratings that reports monthly banner accesses in websites. In site-centric data, recording the number of hits or access could be a problem in itself. As a hit is recorded for every component of the web page, the number of hits recorded could be much more than the actual number. Moreover, there are non-human browsers in the form of robots, search software and spiders that could account for hits. This could be another reason for highly inflated hit records distorting the value of the data. Some users may simple not read the pages even after calling them through the browser. Conversely, whenever a website is once cached by the browser in the user’s system, hits are no longer registered even if the user accesses the page again as it is served up from the cache. This can be a significant drawback for site-centric data as the number of recorded hits would be much less than the actual number of hits. Even the problem can be overcome at the level of the computer system of the user by opting not to cache the pages, caching by the web server and proxy servers cannot be prevented. In user-centric data, the problem of such under or over recording of hits is absent as the sample users come in through a registration process. User identification, as it is, is an inherent problem with site-centric data. Though the IP addresses of the accessing machines can be acquired, the same person may not be accessing from the same machine all the time, or in the case of server-assigned IP addresses, many persons could be accessing from the same IP address. Site-centric data providers are trying to overcome the problem of user identification by using cookies and my making users register on the sites. However, the problem persists. User-centric data is, on the other hand, severely restricted in the sample of panelists. There is always the chance that the sample may not be representative of the user population under consideration. However, the sample can be composed of a particular demographic group according to the requirements of the study at hand. A big disadvantage of user-centric data is the fact that the selected panels do not include work place users who could comprise a significant proportion of the total number of users of any site under consideration. In the case of site-centric data, work place users are included naturally in the scheme of things. Even though user-centric data basically reflects user behavior across multiple web sites, a popular web site with its own niche followers may not be reflected in the data if it fails to attract the minimum number of empanelled users. Data Type Appropriateness In the light of the above discussions, Mullarkey concludes that selection of the data type will depend on the type of study or market research that is being conducted. Since user-centric data is more objective and involves multiple web sites it could be used in evaluating the popularity of multiple web sites on a comparative basis. If, however, a study is restricted to a single web site and does not involve a third party, and if the web site is under the control of the researcher, then site-centric data with its greater emphasis on details could be more suitable for the purpose. Mullarkey further recommends that a within-a-single-site study could use either site-centric or user-centric data whereas a between-site analysis would ideally user-centric data because of the higher objectivity and multiple website access by the same users providing a basis for comparative studies. The type of marketing study being conducted could also determine the type of data that could be used. Any study that examines user behavior would naturally be more inclined to use user-centric data due to the technical difficulties in identification of users in site-centric data. If site-centric data is used for such studies, then it will have to be matched site-centric data. Matched site-centric data would be required for conducting any conversion rate studies as page-specific or ad-specific data would have to be analyzed in such cases. Pure log-file based site-centric data can find use only in the case of evaluation of a website’s own performance without regard to user identification. The level of technical expertise of the research team could also be a determining factor in deciding the data type. Using matched site-centric data would require a much higher level of technical expertise than using user-centric data. Finally, the cost factor also has to be weighed in for selection of data types with use of site-centric data being less costly that use of user-centric data. A judicious combination of all the factors influencing the appropriateness of data types will finally decide the eventual selection of the Internet data types required for research studies and strategic planning for businesses. References 1. Clegg, C. W., Icasati-Johanson, B., Bennett, S., (2001), E-Business: boom or gloom?. Behaviour & Information Technology, Vol. 20, No.4, 293-298. 2. Mullarkey, G., W., (1997), Internet Measurement Data – Practical and Technical Uses, Marketing Intelligence & Planning; 2004; 22, 1; ABI/INFORM Global pp. 42 Read More