Webometrics
Encyclopedia : W : WE : WEB : Webometrics
The science of webometrics (also cybermetrics, web metrics) tries to measure the World Wide Web to get knowledge about the number and types of hyperlinks, structure of the World Wide Web and usage patterns. According to Björneborn and Ingwersen (2004), the definition of webometrics is "the study of the quantitative aspects of the construction and use of information resources, structures and technologies on the Web drawing on bibliometric and informetric approaches." The term webometrics was first coined by Almind and Ingwersen (1997).
Similar sciences are Bibliometrics, Informetrics, Scientometrics, Virtual ethnography, Web mining and Webology.
See also
References
- Mike Thelwall, Liwen Vaughan, Lennart Björneborn (2005). 'Webometrics'. Annual Review of Information Science and Technology, 39, pp. 81-135.
...Web Metrics Continued, By: Radek Wojtal
The Purpose of Web Metrics:
After the boom and bust of the infamous 1990’s tech bubble, corporations began to assign more tangible methods to evaluate web-portion of their business plans. It was no longer acceptable to value a web-site based on its potential alone. Investors and managers alike demanded that resources be meticulously accounted for; all elements of the site began to be quantified as would any other part of the business. The field of web-metrics exploded from this inopportune event. Corporations now enlist an entire industry dedicated to solely assessing whether websites are accomplishing what they were designed to do.
In the beginning, simple metrics were applied to sites. Marketers soon felt pressured however to deliver more sophisticated metrics to their clients. Metrics, meaning measurement, entail setting realistic performance goals that will, most notably and for the purpose of this illustration, most often reflect the marketing objectives of an organization – and then measuring to ensure they are being realized. Where a marketing objective may be to “increase awareness” of a product by 20%, the corresponding web-metric may in fact be to increase “hits” on that product by 80% as a retention coefficient must be factored in. With such realizations, we will observe how the metrics, and in turn their corresponding tools had to become more sophisticated. (Nowell 2001:19)
We will explore the original purpose of each metric and the tools used to assess them, the realization of their weaknesses, and the new metrics and tools developed accordingly in response to those weaknesses.
Tools:
Traditional & Well Developed Web-Metrics Tools
Submissions/Interaction
The first tools devised to measure online performance were the simple feedback instruments. The most popular of these came in the form of contest-entry mechanisms. The web opened up an inexpensive avenue for direct-marketers to receive responses. Consumers no longer had to “snail-mail” their responses to organizations. The success-rate of campaigns of this nature immediately increased. Managers justified expenditures by saying that their metric of “increasing submissions by 15%” had been realized.
The problem with this approach was that the derived findings were very one-dimensional and much qualitative analysis had to be conducted to extract any rich and meaningful data. Moreover, with the increased volume of submissions/interaction, this often meant that organizations were swamped with responses. Having to farm out such qualitative analyses to external organizations, establishing patterns often came at unreasonable costs. (Nowell 2001:14)
Click-Throughs/Clickstreams
As corporations grew more aware of the limitations of one-dimensional analyses, they began to look for more robust and relationship-rich derivations from their sites. The first and obvious approach was to make a simple tally of hits made to each page and cross-reference them to efforts being made to drive traffic to that particular page. The traffic counter became a fixture in convincing other organizations to advertise on any given site.
Though this approach could be cross-referenced with meaningful efforts such as an ad-campaign at a point in time, it still lacked the qualitative insight as to what each hit meant. Questions emerged as designers found that there were inequitable distributions of hits amongst the pages on their site. Some received vastly more, where the logical sequence of a web experience would dictate that they should have the same amount of hits as others. These variances in the distribution of hits indicated design flaws which would result in lost hits to subsequent, but logically more important pages. (Nowell 2001:37)
The more advanced statistical approach was born in charting the clickstreams of sites. This tool simply put, charts the path of each experience with the site, determining amongst other things, where each visitor would stay the longest and from which page they would leave prematurely. This now gave designers the ability to assign more sophisticated metrics to their sites. They could now measure the “experience” one had on the site.
It was not enough to count when a visitor had seen bits and portions of a site; now, only strings of meaningful experiences could be counted as successful hits and the organization had a better understanding of what this site was doing for their bottom line. The issue remained however, that the clickstreams revealed very little about the browsers themselves. A new tool had to be devised so that even more sophisticated metrics could be stated like: “increase submissions by 15% by young adults aged 18-24.” (Nowell 2001:37)
Cookies
With the creation of cookies, a tool which allowed web-authors to leave small strings of code on visitor’s computers, organizations now had an even better idea of how valuable any given browser’s experience was with their website. Though detested by the public, these cookies recorded information about the browser and their online activities.
Enter the privacy issues. Many individuals feel that cookies invade their privacy and sophisticated anti-cookie solutions continue to be developed. Many individuals’ computers are now completely cookie-free. This is an inconvenience to web metrics specialists who have not evolved with this trend. They can no longer tell who is an original visitor to their site, and who has been there hundreds of times before. Organizations now had to develop a new way of obtaining this information. (Nowell 2001:17)
Emerging Web-Metrics Tools
Google/Firefox Extension
In an attempt to curb the decrease in intelligence on browser activity created by multitudes of anti-cookie solutions being introduced, developing newer, more sophisticated, and most importantly less invasive metrics tools became vital. One such tool is the union of Google, the most popular search engine on the internet and Firefox.
The synthesis of these two services is the newest tool in web metrics, synchronizing browsing data across multiple computers as the owner moves from one to the other. As browsers enlist in this service, the feeling of it being invasive (as cookies are when gathering such data) is eliminated. The major problem impeding its usefulness at the moment are both its limited distribution at the moment, and some minor details which it will still not be able to provide to web metrics professionals.
One of those details is the geographic location from which place the hit occurred. This is becoming an increasingly difficult thing to gauge as remote-access laptops become more prevalent. It is important for web metrics professionals to know if they are targeting the right geographic market with their sites. (Nowell 2001:142)
Webtrends ODBC Driver/Geogaphical Mapping
Queries about the geographic whereabouts of browsers have been difficult to answer in the past. Web metrics specialists working with marketers have found it frustrating when they could only give rough outlines as to which area a site’s browsers live or work in. Though it has long been possible to identify each computer’s identification numbers and where they are registered to, till now there has not been a user-friendly interface that would allow web metrics specialists to easily communicate this data to marketers.
With a Webtrends driver, and corresponding geographical mapping software, it is now possible to visually plot the distributions of hits on maps. These can be as specific as area codes or certain neighbourhoods. This ability is particularly helpful to websites such as Hotels.com, where they often advertise in specific areas of cities where business travelers are known to frequent. It allows them to later look at the areas which received the highest level of hits during a testing period and choose which ads did well and why. (Nowell 2001:143)
The Progression Continues…
As demonstrated, web metrics tools evolved from simple response mechanisms, voluntarily used by browsers, to today’s advanced data extraction methods. Web metrics specialists are now able to set objectives as advanced as “a 10% increase in 18-24 year old male traffic, with a known affinity for a related product, living in a particular neighbourhood, without violating their privacy!” This is possible thanks to the progression of the tools listed above.
There are other tools, but this article focused on the main ones having had the most notable impact on the sophistication of web metrics. In the future, the largest opponent to the development of newer, better tools will be the issue of privacy. With new legislation, and a growing distaste for anything big-brotheresque, the sites who find the most creative and tasteful access to this data from their browsers will remain the most successful.
References
Nowell, David. (2001) Marketing on the Web 1st Edition. Toronto: McGRaw-Hill Ryerson Limited. ISBN 0-07-089038-2
From Wikipedia, the Free Encyclopedia. Original article here. Support Wikipedia by contributing or donating.
All text is available under the terms of the GNU Free Documentation License See Wikipedia Copyrights for details.