If you worked for a publishing company, and you send out New Releases Newsletters to your subscribers, how would you determine what genre newsletter you should send to which customers? Should you ask them which genre they like the most when they subscribe (preferences)? Or should you determine what newsletter to send by looking at what books the subscriber usually browses or buys from your company (clusters)?
We decided to work with a New York based publisher to put this question to the test.
The original segmentation this client used was based on genre preferences that customers declared when they subscribed to the newsletters. The company then created dynamic clusters that were based on the actual purchasing and browsing activity of the subscribers.
Clustering is 7x more effective than a preference center
It turns out that deciding which genre newsletter to send to whom based on clusters had a 2-time better open rate, a 4-time better click-through-rate and a 7 time higher click through rate, than those sent based on the consumers’ self-stated preferences.
The conclusion: it is equally, if not more, important to observe what people do as compared to what they say.