A new study appearing on the “Freakonomics” blog aggressively disputes the claim that people prefer Bing to Google and especially the statistical contention that they prefer the search engine over Google 2:1. In an article explaining the study, law and economics professor Ian Ayers argues that the Bing television and related media campaigns based on the Bing It On challenge constitute deceptive advertising.
Ayers says the 2:1 claim comes from a study of just 1,000 users. Using Mechanical Turk, he sought to replicate it with help from students at Yale Law School where he teaches:
The following chart shows the results of the Ayers study, which tested three categories of search queries:
The Ayers study found that Google was preferred most when study participants “used popular search terms or selected their own search terms.” Those who searched with the Bing-suggested queries “preferred Bing and Google in nearly equal numbers.”
There are two potential, contradictory reactions to the Ayers study:
- It either conclusively or largely disproves the Bing preference claims
- Putting aside the Bing advertising claims, the search engine performed relatively well vs. Google
Google won 53 percent of the time and Bing won 41 percent of the query tests, with a tie in 6 percent of instances. That suggests that Bing has the capacity to gain much more market share than it currently has (67 percent vs. 18 percent).
Ayers points out that the more assertive “prefer Bing 2:1” claim has been replaced on the Bing It On website with the more limited claim that “people prefer Bing.” A report with the full methodology and findings of the Yale study is available here.
Postscript: Microsoft provided the following comment on the Ayers/Yale study (attributed to Matt Wallaert, Behavioral Scientist, Bing):
Contributing authors are invited to create content for Search Engine Land and are chosen for their expertise and contribution to the search community. Our contributors work under the oversight of the editorial staff and contributions are checked for quality and relevance to our readers. The opinions they express are their own.
