In my opinion, for Medicare marketing, modeling gives you the most bang for your buck. Here’s why:
At the end of the day, by leveraging modeling, you are able to make your marketing dollars work harder for you. Modeling allows marketers to allocate marketing dollars to the AEP prospects most likely to respond and eliminate contacting those less likely to respond. For example: in the chart below, one of our Medicare clients during 2014 AEP last year did not use modeling, and their marketing cost per new member was over $508.
By applying an AEP acquisition response model, and targeting only the top 3 deciles, they had the opportunity to cut their acquisition cost in half - down to $268.
Based on their budget and their lead/new member goals, they have options of how deep within the model they want to go. If they decided to mail the top 5 or 7 deciles, their overall cost-per-new-member increases, but would still be less than the year before. The key here is to be efficient with every marketing dollar you have and get it to work as hard as possible for you.
Modeling is proven over time to consistently outperform other targeting methodologies - typically improving response rates by 30% or more. It allows you to communicate less to achieve the same results, or to communicate the same amount for improved results. Lastly, while I have used AEP acquisition response as an example, modeling can be used to predict other behaviors for marketing such as: who is likely to leave your plan (“churn” modeling) or, of your former members, who are you likely to win back?
So how can you determine whether one, two or all of these data analytic methodologies is right for your 2016 AEP marketing campaign? Some best practices include the following:
Begin with the end in mind: Analytics means little if it is not grounded in your defined and very specific business objectives. At the end of a campaign or measurement cycle, what will be the benchmark for success?
Use a variety of data sources: When appending data, make sure your data provider uses a variety of sources to cross-validate information. If you ever hear complaints about mailing to deceased seniors, that means your list is old - so make sure that the data is updated frequently.
Adding customized variables allows you to go beyond basic demographics, which can help improve results whether you are using profiling, segmentation, modeling, or all three.
Localize: Rather than a one-size-fits-all approach, evaluate data analytic strategies at a local level (e.g., county or neighborhood) to best understand the differences that exist between locations - because no two areas are alike. For example, for one of our clients that services two counties, we built a model for each.
Small steps: Using data analytics is an iterative process that never ends - as a process always looking for solutions that improve performance.
Most importantly, unlike other marketing, direct mail is measurable and controllable. So don’t go on hunches about what will work. You need to test, test, and then test some more.
So whether your Medicare prospects look more like Gene Simmons or the Breakfast Club, you can’t go it alone. Modeling, profiling and segmentation all have a place in developing a deeper understanding of your members, and a combination of all of them will likely find its way into your marketing. But plain and simple, let data do the work for you, and you’ll have your strong AEP results to thank you.
Rick Berman is SourceLink's Senior Director of Business Development, and as the title suggests, knows more about "seniors" and Medicare marketing than almost any marketer in the country. You can reach Rick at firstname.lastname@example.org or find him on LinkedIn here.