Google’s new Insights for Search allows us to see what the world is looking for.
A few months ago, I wrote a post for my German blog that described Google as a global database of people’s will to knowledge and of their wishes (this of course echoes John Batelle’s famous phrase). In that particular post I then put forth a wish myself: that we receive some kind of API to access this global database of searches.
With Google Insight for Search (”Google Trends on steroids” as Andrew Chen calls it) the search company made a huge step in the right direction. Finally, they’re giving us back some of the information we type into the famous form field day on and day off. Our information.
We can now access information about what terms are searched for when and where. This cannot be underestimated. It’s not only possible to trace whether the global search population prefers red wine or white wine. It’s also a great tool for monitoring brands, especially with the possibility to show the headlines for data points that are somehow standing out of the general trends. It also makes comparing search terms within different categories a breeze - so is this interface to the inner workings of the noosphere a dream come true?
Some things seem unfinished yet. For example the algorithm that is used to match headlines to the graphs. When I’m displaying search terms for Austria, I get headlines from Germany that seem to have nothing to do with Austria. And when I’m analyzing search terms for Germany, I get a lot of headlines from the Netherlands which even is another language. I don’t know how this is done, but it can certainly be improved.
On the other hand, what I like about this service: it does not omit results for Google’s own projects as has been the case with Google Trends for Websites. It does not deny that Knol is not a serious rival for Wikipedia yet and that the distance between Google and Yahoo is not that great (besides: that’s a whole lot of people using Google to search for Google).
A few uses for Google Insight for Search that immediately crossed my mind:
- Monitoring Brands and Corporations
- Analyzing where people using your product live
- Search Engine Optimization
- Meme tracking / Buzz monitoring
- Calculating relationships / Semantic Web
- Tracking a person’s reputation
- Looking for seasonal trends
What else could this be used for? Please add your suggestions in the comments below.