Beating Back Member Churn with Big Data

According to the latest Current Population Report from the U.S. Census Bureau, over one in five Americans changed their health insurance during 2016. Among those with “direct-purchase” policies this increased to almost 60 percent! For healthcare payors trying to sustain profitable growth, that’s an unsettling figure. And for good reason! Recent Gartner research reinforces payors continued emphasis on retention vs. acquisition, finding that 80% of a health insurance company’s future revenue comes from only 20% of existing customers. As a general rule, increasing customer retention rates by 5% increases profits by anywhere from 25% to 95%.

In the post ACA era, the rapid expansion of the individual insurance market has increased the risk of member churn as members who bear personal costs of their policy have proven to be understandably more price-conscious and therefore more susceptible to churning. In order to find ways to retain the members in the individual market, accurately predicting when a member is about to leave a plan is becoming an important part of the insurance business model.

What’s Data Got to Do with It?

Changing healthcare challenges like member churn serve as perfect examples of how advances in data and analytics can work to provide easier-than-expected solutions to difficult new problems. With the critical data already available for most payors, it possible to create new models and strategies that will accurately predict when a member is about to move to a competitor, understand what is driving that behavior and take preemptive action to save that member.

A Case in Point

For one major health insurance company, digging in to their member churn challenge involved working with Amitech to develop a new model that would help them regain perspective on accurate predictors of customers’ churn behavior and begin proactively targeting those at-risk.

Via proportional hazard models built on existing data, Amitech was able to identify multiple, actionable indicators that significantly impacted modern churn risk.

Key findings included:

• When a third claim category is added, odds of churning were reduced by 19%
• Older customers were less likely to churn. Each additional year of age reduced the odds of churning by 7%
• The more overall claims a member had, the less likely they were to churn
• Female subscribers were 24.4% were more likely to churn than male subscribers

Turning Insight into Income

As it relates to member churn, knowing the who, when and why is half the battle. But if you want to win the war, you need to know how to turn those insights into a plan of action with a direct line to business value.

Thankfully for the client in question, we’re not in the business of providing half-answers to hard problems. Based on our findings, we made several additional business recommendations that would enable the info at hand to have a true impact on the bottom-line.

Key recommendations included:

• Spending more marketing dollars on female customers since they generally had a higher probability of churning after the first year of being a customer
• Spending less on heavy users as it indicated a higher relationship bonding with the firm, or that the customers had pre-existing conditions that prevented them from switching
• Developing initiatives that would encourage customers to use more than one line of coverage, such as adding Prescription or Dental to Medical

Real-Time Value

For this client, data and predictive analytics was the key to recapturing their ability to reduce risk, lower member churn and increase profitability. Data has the same potential to add real-time value to a limitless variety of other healthcare challenges, as well. Please visit our website to read the full case study and explore the experiences of additional clients to learn how Amitech could help you solve some of your most pressing problems through data and analytics.