Medicare Advantage (MA) plans understand the importance of retention and make it a key part of their growth strategy. The challenge then for these plans is how to stay ahead in the retention game. This is a competitive market where there’s a continuous stream of new entrants and most health plans already have advanced retention analytics and intervention processes in place. To be successful, health plans need to keep evolving retention analytics capabilities by adopting industry-leading practices faster than the competition.
Companies can achieve industry-leading practices by focusing on two key components of evolving retention analytics:
- Expanding channels for retention intervention: This includes product updates, member engagement and agent engagement.
- Building mature analytics capabilities: This is achieved through upgrades in data, smarter analytics and streamlining intervention execution.
Every step either on the horizontal axis or vertical axis is a step toward evolving retention analytics.
There are three primary ways to influence churn: Engage high-retention risk members at the right time and with the right message, partner with brokers to improve retention and make updates to MA products to reduce changeover.
Retention rates are higher in populations that are more engaged with health plans. The progression of better member engagement occurs by moving from cohort-based analytics and interventions to ever more personalized outreach addressing potential reasons for turnover. An example of personalized engagement would be outreach to a member impacted by a specific formulary update to share alternate drug options. Another could be targeted annual notice-of-change calls to members that are at a higher risk of disenrollment as a result of the updates.
A majority of MA sales happen through brokers and they play a big part in member retention and renewal. With advanced analytics we can predict which agent has the higher probability of losing a member during the annual enrollment period or open enrollment period. With that information, broker relationship managers can increase engagement with those brokers and incentivize them to take corrective action. The evolution in agent engagement occurs by moving from focusing entirely on high producers to producers that can help reduce churn.
Winning at product design
Members most often switch plans because of dissatisfaction with coverage and out-of-pocket costs, both of which are tightly linked to the MA plans offered by competitors in a given market. Using analytics to find out which plan features are gaining or losing traction in a geography can be a big help in making plan updates to reduce unenrollment. Additionally, when using the MA product and enrollment information by county, which is shared by the Centers for Medicare & Medicaid Services (CMS), it’s possible to identify features of plans that are gaining traction and may pose a threat for member loss in that geography.
The second and more commonly understood dimension of evolving retention is analytics maturity, which involves using data, analytics and tools to implement insights.
Data are the backbone of advanced analytics. Upgrades in available data can occur by expansion of data, by improving the quality of data or by improving the velocity of data. Expansion refers to adding new data sets for retention analytics, including social determinants of health data, or data on preferred channels of engagement for MA members and agents selling MA products.
Data quality—as it relates to lives covered, information about fill rates and elimination of bias—should always be viewed as an opportunity for continuous improvement. Velocity, in this case, means how quickly a data set is made available for retention analytics. It could, for example, refer to how quickly network changes are added to the retention analytics engine. Velocity times can span months, weeks or even a real-time feed.
Evolution in advanced analytics means improving or building advanced machine learning models for retention analytics programs as Medicare member preferences and behaviors keep evolving. Improving models is a function of staying up to date on the newer machine learning techniques to improve predictive power, working with a smaller set of data and using an ensemble of models to improve accuracy and bring consistency and transparency into model results. Those advancements in analytical models will assist health plans in improving the accuracy of predicting high-retention risk members. It also will help forecast competitive product changes in a geography and refine MA member targeting by considering lifetime value.
Improvement in the execution aspect of insights is as important, if not more important, than generating insights. Execution has human, operations and technology components. It includes everything from executing retention interventions to improving the rigor with which data curation contributes to retention analytics programs and member engagement platforms.
Following the two-dimensional framework involving expanding channels for intervention and analytics capability will help health plans think through opportunities for improving their retention program. Health plans need to keep beating their previous best in retention analytics to stay ahead of the competition. Organizations that follow the analytics capabilities and expand channels for a retention intervention framework will set themselves up to be competitive well into the future.