It’s a healthcare marketer’s dream: the ability to understand the needs of prospects and patients so thoroughly that you can predict what they’ll likely do next. Armed with such valuable insights, campaigns could be created with a much greater level of personalized effectiveness.
Enter propensity modeling—a predictive modeling tool that’s popular with marketers across industries.
Consulting firm Bain & Company describes it like this: “A propensity model calculates the likelihood of a prospective or current customer’s next steps. Understanding likely next steps helps companies deliver better experiences, increase loyalty, reduce churn and build value for the organization.”
Mercury Healthcare—a “performance-driven technology company” that specializes in using data to help healthcare organizations drive patient/consumer engagement and optimize outcomes—refers to propensity modeling as a “subset of predictive modeling” and defines it as “a family of multivariate statistical analyses used to optimize the prediction or likelihood of a specific event to occur.”
“In healthcare, propensity modeling involves using health analytics to identify the best prospects for targeted marketing efforts,” the company says. “Roughly translated, it’s a statistical approach that considers many variables at the same time, deriving insights from patterns and relationships in the data and using that information to create optimized and strategic marketing campaigns.”
But to get the most from propensity modeling, Mercury Healthcare says there are certain dos and don’ts that healthcare marketers should keep in mind.
Propensity Modeling Dos
Propensity modeling can be a valuable tool to guide marketing strategy—as long as certain best practices are followed along the way. According to Mercury Healthcare, these include:
- Defining your target or goal—by asking the right questions to help hone your efforts; identifying the “primary purpose” of your campaign; and integrating the “scope of the campaign.”
- Using the best data—which means it’s as recent and “clean” as possible. Best practices in this context include finding out when data sources were updated last; obtaining data summaries “to ascertain depth, breadth, and fill rate of data elements;” checking into “data cleansing and matching processes;” and, when possible, using “data that include standardized recodes.”
- Using multiple data sources and the most appropriate analytics—such as creating messaging and prospect lists by using “standardized socio-demographic and socio-economic segmentation schemes” and integrating additional filters to enhance specificity. Regardless of the approach adopted, it’s essential to “always use the best analytics to meet your goal.”
- Ensuring data are vetted and validated by clinical/academic experts—when selecting a propensity modeling vendor: “In healthcare, it is critical that data, modeling, analytics, and targeted marketing are based on current empirical and best practices.”
- Deploying validated analytics and follow-up testing: “Propensity modeling becomes a lifecycle when you use follow-up testing to better define your targets and goals.”
Propensity Modeling Don’ts
On the flip side, Mercury Healthcare says there are five common mistakes healthcare systems should avoid when it comes to predictive—aka propensity—modeling. The following offers snippets from the list. For complete descriptions and recommended remedies, please see the full post.
- Building predictive models with incomplete or inaccurate data: “User error, bias, accidental deletion, and improper coding all contribute to a problematic propensity model, and ultimately to problematic campaigns. …”
- Sticking to one predictive model: “There’s no such thing as a universal algorithm that constructs the perfect predictive model for all healthcare marketing endeavors. Different algorithms may produce entirely different results from the same datasets. …”
- Overfitting predictive models: “When creating a predictive model, it can be tempting to adjust your model so it conforms perfectly to your current dataset – to the point that it’s really only describing a random error, or noise, as opposed to the legitimate relationship between variables. …”
- Believing that predictive modeling replaces rules-based analytics: “While predictive modeling reduces the legwork in deciphering healthcare datasets, it doesn’t replace the need for rules-based analytics. All possible conditions and outliers must be built into the model upfront to flag problems or outliers. …”
- Failing to adjust or alter predictive models over time: “As more data is made available…it’s important to refine your predictive models to accommodate changes. …”
Propensity modeling is a valuable tool that can help healthcare marketers optimize their efforts. By integrating recommended dos and don’ts like these, you’ll be well on your way to getting the most out of your next campaign.
Contact us today to find out how we can help level up your healthcare marketing strategy.