For anyone who hasn’t been following this week/last week’s whole Google and Bing copying and not copying saga, let me fill you in.
Google noticed late last year that Bing were showing the same result as Google for a misspelled query (torsorophy which should have been spelt, tarsorrhaphy). But Bing showed no spelling correction in their result. The news was broken via Danny Sullivan in another great investigative journalistic pieces.
This tipped Google off to the fact that Bing may have been copying their search results and began an investigation. Google gave 20 engineers, 20 new laptops kitted out with Windows and IE8 with the Microsoft Toolbar installed and privacy settings allowing Bing to collect their clicks and general behavior.
Over a study of many rare and misspelled phrases Google showed that after a short gap of a couple of weeks Bing were getting the same results as Google. But what makes it really interesting is that Google went above their algorithm and inputted totally unrelated results and still Bing after a short delay were showing the same results.
Here are some of the sample pictures Google posted on their official blog post on the Bing Sting:
And a couple of days later Bing showed this result:
The conclusion in Google’s opinion was that “Bing was copying their search results” and Google felt very strongly about it saying “Bing’s results are merely a cheap imitation of Google’s”.
But is this the right conclusion?
I believe not for a couple of reasons:
The data that was being sent to Bing was sent through the Microsoft Toolbar which was only collecting click data.
That basically means that Bing tracked what a user searched for, and which pages they thought to be most relevant to that search, which were in turn chosen by Google and then clicked on by the user.
Now let’s look at that from another angle.
Bing sees you on a webpage and Bing determines that, that webpage, let’s call it page A is about “chicken recipes” and then you click a link from that page to another webpage, page B. It is safe to assume that page B is somehow related to page A, and by default “chicken recipes”, and why wouldn’t you use that information in your search results?
This is valuable and relevant information that Bing that should be included in their algorithm but more than likely place little, very little weight on in that algorithm.
But now let’s look back at the examples Google uses. They use very rare queries that would never have been searched before like: hiybbprqag and delhipublicschool40 chdjob, which both mean absolutely nothing, right?
Bing has no data, no links, no pages crawled in fact absolutely nothing on these queries (no other ranking signals). Until one of these Google engineers sends them a small piece of information, which warrants Bing listing the page that was clicked onto by the Google engineer as a search result.
This is absolutely the same as the “chicken recipes” example. Page A the Google search result is about hiybbprqag or delhipublicschool40 chdjob or whatever nonsensical query you put in. The user goes from page A to page B and Bing tracks this. The result Bing believes that Page A is similar to page B. It is the exact same as the “chicken recipes” example.
The reason these particular examples look like Bing have copied Google is because they had no data on these queries other than that provided by the Google engineers. So for regular searches that Bing have data on, this click tracking probably has little or no relevance to the search result you get back.
Now let’s look back at the same model we used for chicken recipes and you wouldn’t have any click from a Google search result. In fact there is no proof that this tracking was based solely on Google results and most likely isn’t. In fact Bing haven’t even intentionally tracked a Google search result, that is merely one of the websites that Bing ends up tracking. It could have been any other website on the World Wide Web.
This tracking wasn’t targeted at Google and wasn’t copying their search results. We know this because Bing are tracking the clicks between Page A and Page B and making a correlation between the two. It has nothing to do with tracking Google.
Google’s media storm, multiple blog posts and Matt Cutts speaking at the Future of Search conference have been totally unwarranted and could certainly be called a smear campaign. But Bing’s response has been weak and an easy defense like this post would have quelled much of the attention placed on this story.
Until there is a study into whether a Bing takes into account clicks from pages related to certain query to another page unrelated to that query, we can’t say accurately whether Bing is copying Google’s results.
In fact there is no evidence or reason to believe that Bing is copying Google’s search results. The only conclusion that can be made out of this is that Bing uses the click data of user who clicks from Page A to Page B and makes a correlation between the two.
My conclusion: Bing DOESN’T copy Google’s search results.
Credits: This Quora thread and this post from TechCrunch are good reads that both touch on same logic published in this post.
About the author: Mark Collier is the author of Link Building Mastery an ethical link building guide, that gives you the true strategies for link building that actually work. LBM also features 15 expert interviews including Yaro Starak, Ann Smarty, Marko Saric, etc. You can join the ethical link building revolution now.
