Google’s Gift of Correlation


Looking for a deeper dive on your search results? Google’s new Correlate system is a data-mining tool that can be used for virtually any data search. A bonanza for marketers, educators, economists, publishers and many other professions, the genesis of the new service is integral to the model.

During the flu epidemic of 2008, Google noted that the “activity of certain search terms are good indicators of actual flu activity” and subsequently launched Google Flu Trends, a warning system based on the statistics. 

The ten-ton penny drop: a quantifiable correlation between the search world and the real world, in an ‘aha!’ moment illustrated in a comic that recalls Google’s Chrome comic book explainer by Scott McCloud.[more]

Although Google Trends and Google Insights for Search let users enter a search term and see a trend, feedback from the research community called for the ability to enter a real-world trend and see what search terms match, i.e., Google Trends in reverse.

So Google built Correlate, whose methodology a blog post describes as: “upload your own data series and see a list of search terms whose popularity best corresponds with that real world trend. In the example below, we uploaded official flu activity data from the U.S. CDC over the last several years and found that people search for terms like [cold or flu] in a similar pattern to actual flu rates. Finding out these correlated terms is how we built Google Flu Trends.

You can also enter a search term and find other terms whose activity corresponds well over time with the one you’re interested in. While cell biology isn’t so popular in the summer time, continues Google, “What’s interesting is that the ups and downs of web search activity for cell biology terms is unique enough that searching on Correlate for ribosome brings up searches for other biology terms, such as mitochondria.”

In a broader world than biology, “Brand managers keen to see if their promotional TV ad campaigns result in echoed online activity searching for their brands (and possibly compared to their competition) can do so with a few clicks. Publishing houses can see whether readers are searching for their magazine, or their online entity,” comments Fast Company.

Consider it another free tool from Google, which will benefit as their tool is widely used and advances an even more granular understanding of data correlation; but the benefit for all of us, garnered from relational and predictive analysis in a highly correlative world, is a gift of unsearchable magnitude.