« Searches for Hybrids Up 43% from March to July 2008 | Main | SEW Experts: Leveraging Social Media to Educate »
August 27, 2008
Google's Search Experiments are Sometimes Subtle
Have you ever been around a friend who points out a flaw about themselves and you say, "I would never have noticed that if you hadn't pointed it out?"
Google is doing the same thing sans flaws.
On the Official Google Blog, Ben Gomes explains that sometimes Google's search experiments are very subtle. They'll test things like white space in the search results or how big the plus box should be next to a stock quote link.
The differences among the options being tested are very difficult to detect without being told what's going on. But the results of tests show that even a small number of pixels can make a big difference. For example, it could mean the difference between a searcher spending a lot of time on one search result - and not necessarily the first one - or a searcher being able to scan the results more easily. And then Google has to decide which is better.
All in all, Google's search experiments are not always about the alogorithm. What do you think of these experiments? Leave your thoughts in the comments!
Related Reading:
Google News Testing Updates
Google Testing Time Elements In Search Results
Google Testing Expandable "More Google" Links On Home Page & More
Posted by Nathania Johnson at August 27, 2008 11:14 AM
Stumble It!
Digg this!
Add to del.icio.us
Comments
I always wonder how Google measures success in these experiments. If they were focused just on themselves, I'm sure they'd look at an effective CPM calculation. But part of how they've done so well is by being focused on the customer and on advertisers, but Google wouldn't have great data on those segments.
Posted by: Eric at August 27, 2008 3:38 PM
I guess being the most trafficked site on the web enables them to do this. With so much traffic, even little adjustments are going to generate significant improvements.
Posted by: Robin at August 28, 2008 11:44 AM
