Campaign Predictive Analytics
Predictively Analyze Your Responders to Score the Best Non-Responders
When deploying a direct mail or email campaign, a relatively small percentage will typically respond or convert. Often one or a small number of touches is just not enough to produce a result. This does not mean that nobody else on the list will respond. But, unless targeting and timing was 100% perfect, it means that some of the non-responders are more likely to respond to a follow on campaign than others. One way to predict who the potential responders are is to use logistic regression to determine the most important demographic factors, and create a multi-variate regression model that places non-responders in one of 10 buckets or “deciles”. Typically, the highest ROI is achieved by mailing to the top two to four deciles.
Different Approaches: Self-Contained Analytic Solutions vs. Custom Analytics
Custom Projects for Campaign Predictive Analytics
Custom analytics — Using R, Alteryx or other statistical software tools — which your data scientists and analysts can perform (or which Data Marketing Strategies can help working together with one of our Analytics Partners) will ultimately perform better than self-contained / self-service solutions. If you do undertake this project, we highly recommend Enhancing Your Customer Data with additional third-party data to improve model performance.
The downside is that custom models will take time to develop. Perhaps the most time-consuming aspect involves “data engineering” which essentially means getting your data into the proper formats required for various types of statistical processes to work properly. Tools such as Alteryx are very good at this…even so this is typically a substantial project. And unless you have a local instance of a U.S. Consumer database, you will not be able to apply scores to individual prospect records, if reaching best prospects is your goal.
Self-Contained/Self-Service Campaign Predictive Analytics
We work with multiple vendors that offer self-service Campaign Predictive Analytics solutions. This allows us to compare model results to make sure that nothing important was missed by one or the other vendor.
The Campaign Predictive Report indicates which data elements are most predictive in the model and provides a Lift Chart which predicts how much better responders in the first few deciles will perform relative to the list overall.
Some solutions also provide the Principal Component Analysis results including model weightings by demographic element.
As well as the breakdown of ranges within each element and their indvidual index values.
In addition to the modest model development fee, there is a small fee per to provide the decile score to the records in the prospect list file. These fees are low compared to the increased ROI from prioritizing which records to mail, thereby saving thousands of wasted mailers or emails.