Wagging its tail as it chews on a meaty bone of statistics, meta-search engine Dogpile.com conducted a study in conjunction with the University of Pennsylvania and Pennsylvania State University that shows Dogpile.com is better than everybody else. If you’re interested, I did a study of my own showing that Einstein was an imbecile when compared to the electrifying brainpower emitting from my gargantuan cranium.
On average, searchers, in a sense, use their own meta-search technique by querying 2.8 search engines per month. Dogpile touts its own search service’s ability to produce a better variety of results, taking the top results and compiling one list.
Based upon their statistics, searchers are not always finding what they’re looking for.
Conducted in July 2005, the study measured the overlap of first page results from the top four search engines and found that only 1.1% of 485,460 first page results matched up, meaning that most results were unique to each search engine. In all, 84.9% of search results were unique to one search engine.
Dogpile says that by only searching MSN, a search can miss 72% of the Web’s best first page results. Only Google-70.8%; Yahoo!, 69.4%; AskJeeves, 67.9%.
On average, 66.4% of Google first page results were unique to Google; 71.2% were unique to Yahoo!; 70.8% were unique to MSN; 73.9% were unique to AskJeeves.
Twenty percent of the time, the top search engines completely disagreed on the top five non-sponsored search results and only 7% of the time did they agree across the board.
Dogpile claims to provide the best results across the board by aggregating all top results into one. They have a nifty flash demonstration of the concept, too.
