Home
 > Research and Reports > TEC Blog > Will Max Get Mad When He Surfs Your Website?

Will Max Get Mad When He Surfs Your Website?

Written By: D. Geller
Published On: April 24 2000

Will Max Get Mad When He Surfs Your Website?
D. Geller - April 24th, 2000

Event Summary

It's more than axiomatic that surfers have low tolerance for sites that load slowly or make it hard to find information - it's actually true. Companies need to know whether their websites suffer from this kind of problem. Most traffic analysis tools tell what pages have been visited and how long the pages were viewed. By looking at the last page that a surfer visits a company can deduce what parts of a site surfers don't like, but this requires that the company first lose those surfers (and potential sales).

WebCriteria takes a different approach. It relies on determining the behavior of real surfers against a variety of web site styles, and then rolling the results into a model of a "typical" surfer. This model, named Max, examines the pages of a site and reacts to slow loading graphics, placement of links, the amount of a page that is actually visible on a screen at one time, the language style in the text, and other factors. MAX is based on a psychological model called GOMS first developed by researchers at Xerox PARC; the model incorporates information about such human behaviors as reading comprehension and motor skills. It has been verified against real surfers and found to be statistically accurate by an independent testing laboratory.

The new study that is being undertaken by the company will expand the heterogeneity of the model. Max, like most research in psychology, is based on a population of college students. The new study will enable the company to build a family of Maxes (and Maxines) that cover other specific demographics, specifically age and browsing experience. The company could also, depending on the data collected, create goal-oriented Maxes, such as one that captures the behavior of surfers shopping for particular kinds of merchandise.

The reports from the product show detailed information about Max's browsing experience on a URL-by-URL basis, so that a company can zero in on problem pages and understand why they have problems. The product also provides comparative analyses of competing sites. The company publishes aggregated statistics for various vertical segments for free on its website.

Market Impact

WebCriteria's approach should interest website marketers. As a result, the products should do well. This will not be an easy product to replicate, although there will probably be some look-alikes if WebCriteria seems to be catching on. And, especially if they can build interesting models from their new research, we expect them to make significant penetration. However, we do not believe that they will replace existing traffic analysis tools. These will still be necessary on a daily basis even after a site has been analyzed by Max and optimized. Traditional traffic analysis attempts to answer questions about what happens when people look at a site; Max wants to explain why it happens.

There are other ways to model human behavior besides the one taken by WebCriteria, so don't be surprised to see articles comparing the validity of different psychological modeling theories appearing in the Internet trade press before too long. Such academic discussions may be a breath of fresh air compared with the usual news and analyses.

User Recommendations

About half the company's sales are made to large corporations, which seems reasonable. Max is an easy purchase for a large company that needs every advantage that it can gain in a competitive web market, and may be unsure about the quality of its Web development. Smaller dot-coms (a quarter of current sales) probably have more confidence in their abilities, but if they are entering a competitive market they will also find this a useful tool. The remaining sales are to consultants and agencies that use the product for their customers or resell it. We think that any website that is in a tough competitive situation could find a friend in Max.

 
comments powered by Disqus

Recent Searches
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z Others

©2014 Technology Evaluation Centers Inc. All rights reserved.