Quick post this – I just came across this last night, so I thought I’d put down some thoughts.
Google’s newest lab rat, is Google Correlate and it complements Google Trends in the sense that it allows you to upload your own data series and look for corresponding data trends.
Google’s mission statement for this is: “Google Correlate finds search patterns which correspond with real-world trends.”
What’s really interesting is that it works like Google Trends in reverse. With Google Trends, you type in a query and get back a series of its frequency over time. But what Google Correlate allows is to enter a data series (the target) and get back queries whose frequency follows a similar pattern.
What does this mean in real world terms?
My understanding of this, is that it allows you to upload a series of data, let’s say for example how may people visited your website (and you know that it’s during say a slower period of the year), and they came to your site looking for “widgets”.
You could take the search term “widgets” and the data series (of visits), upload the info to Google Correlate – and it would spit out other related data streams that follow the same series. This means that (in search terms) you could possibly target online categories that follow a similar online cycle – and further optimise (or spread) your search budget.
It effectively cuts out all the complicated data analysis required to try and find similar search patterns, but based on your own data. This is where it differs from Google Trends in that all the data is coming from Google in Google Trends – but with Google Correlate you can upload your own data to be queried against what’s already stored by Google. (It’s pretty chunky in that it goes back to 2003, too).
The maths behind the algorithm etc is pretty complicated – the number wizards out there can check it out here.
There’s a whitepaper on it here – and the Google FAQ’s are here.
I haven’t played around with it enough to see of it’s much more than a number-crunching exercise in terms of Google showcasing their ability to generate relevant results, based on your data – but I’m guessing that if it proves to work that it’ll (no surprise) get online marketers spending more online dollars as they try and exploit other data (search) trends that mimic their own.
In other news, the www.igamingsupershow.com was busy, and it was good to meet so many old faces that had made it to Dublin. I’ll wrap up some thoughts on that, and #bluemonday – when I get time to draw breath.