Seasonality is not a new phenomenon in search marketing – every marketer worth their salt knows that queries for terms like “ski” go up in winter and “suntan” peaks in the summertime. However sometimes seasonal trends are less obvious to the naked eye which is why Google’s Insights For Search (among other tools) can be helpful, allowing users to trace back search volume for particular keywords and to compare the volumes over time for multiple keywords.
Now Google aims to do the predicting for you with their latest update to Insights, and they have also published a nifty companion paper available here. While much of the paper is filled with the kind of statistical modeling formulas that drove my decision to get a degree in History the top level conclusions are quite interesting. For instance, according to Google:
- Over half of the most popular Google search queries are predictable in a 12 month ahead forecast, with a mean absolute prediction error of about 12%.
- Nearly half of the most popular queries are not predictable (with respect to the model we have used).
- Some categories have particularly high fraction of predictable queries; for instance, Health (74%), Food & Drink (67%) and Travel (65%).
- Some categories have particularly low fraction of predictable queries; for instance, Entertainment (35%) and Social Networks & Online Communities (27%).
What this allows for in the Insights tool is the ability to see that global query volume for terms like “pancake recipes” show clear seasonality that allows for predictability:
The red line at the bottom of the chart is for the query “pancake coupons” and while there isn’t enough data to allow Google to forecast moving forward, the slight lift over the last year maps to the recession related findings in the report, per Google again:
We show several examples that demonstrate possible influences of the recent recession on search behavior, like an observed increase of query share for the category Coupons & Rebate compared to the forecast. We also show a negative deviation between the query share for category Restaurants compared to the forecast, where as the category Cooking and Recipes shows a similar positive deviation.
I would also point out what they’ve noticed in the car industry as a function of tough economic times:
We show that in the recent 12 months there is a positive deviation relative to the forecast baseline (i.e., an increased query share) in the searches of Auto Parts and Vehicle Maintenance while there is a negative deviation (i.e., a decrease in query share) in the searches of Vehicle Shopping and Auto Financing.
While Google takes pains to point out that they can’t foresee the unforeseeable, these trend extrapolations are interesting guides that marketers can find helpful.
It’s also instructive to look at the categories that Google found hard to predict, such as entertainment. I suspect a big part of the difficulty there is reflected in the lack of long-term branded keyword data. In other words, someone in the auto sector might consistently search using the brand name “Chevrolet” no matter what other marketing efforts are in market – whether they be incentives, new model launches, or other efforts. Similarly, marketers are likely to always want to capture the traffic around “Chevrolet” as a search term.
On the other hand folks searching for “Transformers 2” are very unlikely to know or care about the studio releasing the film, and there is little (if not zero) point in using, say, “Warner Brothers Records” as a search term when you want info on the new Flaming Lips record. As a consequence branded campaigns are a very small part of marketing in the entertainment world and search term volume and interest comes and goes with whatever specific initiative is on tap.