MediaMarkt - Combining site-search with paid search
-
Year
2019
-
Company
MediaMarkt
-
Services
Google ads, Marketing tech
Problem
With an inventory of 30.000 products, it can be very difficult to determine at scale what keywords you should advertise. While keyword research is a good source, it is a manual process and difficult to scale across the long tail of 30k products.
SOLUTION
Three datasets were connected to each other on a daily basis in Google Bigquery.
1/ Site search. Tens of thousands of searches were done every day on MediaMarkt's website.
2/ Organic data was pulled from Google search console
3/ Paid query data was pulled via the Google Ads API
Next, these 3 datasets were joined. This made it very easy to identify searches that were happening on the website or in organic that were not covered by paid search. In addition, given that all these datasets had search volume tied to it as well it was very easy to prioritize which keyword additions to do first. The data connections were made on a daily basis, so every day new keywords that weren't targeted in paid search would come flooding in.
Impact
Entire new keyword categories were added in paid search. For example, it was discovered that specific TV inch sizes weren't added to the paid keyword portfolio that were being searched for on the site or organic.
As a result, total keyword coverage grew by 20% and through that 12% of additional revenue + profit was generated without sacrificing profit on ad spend.
