Search engines are more different than people think

The conventional wisdom is that the major search engines serve up similar results. Our research suggests the opposite: the search engines are far more different than most people think.

For example, what do you suppose is the overlap between the top 10 results on Google and the top 10 results on Yahoo! for the average search term? Most people we asked thought about 70 percent. So did we, until we checked.

The actual overlap is roughly half of that. In tests we conducted using the 500 most popular search terms, on average, Google and Yahoo! shared only 3.8 of their top 10 results.

Even more striking is the distribution of results: fully 30% of the search terms had 2 or fewer overlapping terms, and only 17% had 6 or more overlapping results among the top 10.

The Overlap between Google and Yahoo!:

Source: Jux2 Analysis of 500 top search terms, April 2004

What are the main differences?

So far we've found that none of the major search engines is particularly similar to the others, with none of the pairings sharing more than 4 results in the top 10, on average. (We're working on testing MSN's new algorithmic search products, and will have more to report shortly.)

That said, results from Google and Yahoo! overlap more than results from Google and Ask Jeeves, which in turn have more overlap than Yahoo! and Ask Jeeves.

Source: Jux2 Analysis

The following chart shows another way to look at it. In every match up, over 80 percent of searches share fewer than 6 results of 10.

Source: Jux2 Analysis

We are focusing here only on the top 10 search results because we have found that most searchers either find what they are looking for in the first page of results, or they change their query and try again.

However, looking past the top 10 results confirms the thesis that the search engines are quite different from one another. In a separate test of 91 random searches, we found that Google and Yahoo! share only 23% of their top 100 results. Furthermore, only 4.8 of Google’s top 10 results even made Yahoo’s top 100 (conversely, only 5.4 of Yahoo’s top 10 made Google’s top 100).

Why does this matter?

If the search engines are providing top results that are very different from each other, then by using only one search engine, Internet searchers are potentially missing relevant results.

It stands to reason that searchers will get a better assortment of highly relevant results by searching two engines at once with jux2. So, give jux2 a try and let us know what you think.