Last week, MediaPost reported on an Outsell report, "How Information Providers Can Keep Pace With User Demands For Time-Saving Solutions," which showed Google increasing its market share dominance but also reportedly increasing dissatisfaction with the quality of search results. To get a better look at these provocative findings, we obtained a copy of the report.
The analysis compared surveys conducted in early 2005 and 2006 among U.S. "knowledge workers" in four sectors, Corporate, Government, Health Care and Academia. It found increasing amounts of time spent on "information tasks" (gathering and analyzing information). The two primary culprits were the increasing proliferation of information and increasing inefficiency in information retrieval ? in particular of web search.
The report cites two statistics to draw the inference that there's increasing dissatisfaction with web search. First it cites an increase in the use of corporate intranets as a starting point for information (up from 13% to 19%). It also cites "search failure rates" (defined by users as unsuccessful searches for desired information) as a serious problem. Failure rates across the four categories were 31% on average. This is indeed striking.
Unfortunately, the report neglects to further drill down and discuss whether and how these failure rates are specifically affecting knowledge workers' behavior now and attitudes toward usage of search engines in the future. (It does reflect usage of a range of information resources, which are basically flat across the two surveys, except for the rise in intranet usage).
Also striking was the market share data reflected in the report. The sample size was 5,740 respondents in the early 2006 survey. And while these "knowledge workers" are not perfectly representative of the general U.S. population the sample size is large enough to merit some generalizations to that larger population.
Danny last week compared the market share numbers for April from comScore, Nielsen and Hitwise. Of the three, Hitwise had Google with the largest market share at 59%. Outsell's report shows and even larger margin (again, not necessarily representative of the population as a whole). But the Outsell distribution is as follows:
During the survey interval Google gained share at the expense of all its competitors.
Another finding of the report is that wireless data usage has now reached "critical mass," with almost 70% of users accessing some sort of content over wireless devices.
In the MediaPost article Outsell analyst Roger Strouse muses on the reasons for failed searches:
"[A] variety of factors are to blame for search failure, including deliberate manipulation of Google's semantic system: 'From early on there's been a problem with what people used to call spamdexing, where people would apply tags to their Web sites that actually had no relation to the content at all. And that problem is still there, and is definitely going to have to be resolved some time soon.'"
Right now there are few effective alternatives to search engines for consumers as a starting point for research when they don't know where to find information. One quick and facile argument is that vertical search threatens Google et al. However I would argue that the relationship between vertical and general search is complex and the emergence of vertical search doesn't negate general search usage in those categories automatically.
The report's findings are interesting and certainly point out that everyone can do better in more efficiently providing relevant results to users. However, the report confirms that until something better and more relevant comes along, for now "Search, They Name Is Google."