Jonathan Saipe

Google Correlate Tool Helps Search Marketers

2 June 2011, Jonathan Saipe


This week Google announced a great new tool called Google Correlate which gives users correlation search data by time and region (although currently it’s only showing data from the US).

It’s a bit like Google Trends or Google Insights for Search in reverse where users can enter a query and receive search trends by time and location. With Google Correlate, users enter data and get back a list of search queries which follow a similar pattern. Hence the name Google Correlate. Clever eh!

If you’d prefer, you can even upload your own data or draw your own line chart and Google Correlate will try to publish the same correlation search data as your own. But that’s for later.

Keyword research for search marketers

So I’m a keen cyclist and I wanted to know what correlation data existed around the search term “cycling”. Google found the following correlations after entering “cycling”.

Google Correlate results for cycling

The results may not be particularly surprising in this instance as search terms like “road biking” and “road bicycle” are obvious synonyms of cycling. But nonetheless, this is useful when carrying out keyword research for your search marketing activity.

And it’s one step up from the Google Keyword Tool we all know and love.

Google Correlate will also output either a line chart or a pretty cool scatter plot showing correlations by time (which you can zoom in on by selecting a time period):

Google Correlate line chart

Google Correlate scatter plot

If you want a better understanding of correlation data by region (currently US only), Google Correlate will accommodate:

Google Correlation regional data

Interestingly, in this instance, Google Correlate chose to correlate cycling with cycling clothing regionally, possibly due to climate differences across the US.

Draw your own line chart and find correlation data…

Turning Google Correlation in reverse, you can draw your own line chart and Google Correlate will find correlation data that matches. This took some attempts on my part and eventually I came up with the following results from my drawn series.

Google Correlate drawn series

Not a bad tool amidst a dizzying week of announcements including the launch of  the Twitter “follow” and Google +1 buttons. But that’s for another blog post…