Using a segment-specific retention model helps a health-tech firm reduce annual churn rate by over 11%

About

The client is a leading health-tech firm which offers subscription-based health and wellness plans to customers in the US and other countries in the world.

Challenges

  • The client  faced high customer churn post 30 days of usage resulting in revenue losses. 
  • The marketing team was keen to understand and identify key influencers of churn behavior.
  • The present rule-based approach to retention was inaccurate and not scalable.

Solution

Armed with multiple decades of experience in customer retention, the Tenzai team devised a three-stage approach to address customer churn. 

Data from multiple sources like CRM, product usage, subscription behavior and campaign system were considered for the analysis. Key drivers of churn were identified based on in-depth exploratory analysis. Based on the insights from analysis, marketers were able design strategic interventions/programs to address churn. 

Next, Tenzai developed a customer segmentation model to understand customer behavior among diverse customer groups. Customers were segmented into clusters based on their product usage, value, loyalty, and subscription frequency.  It helped in understanding the variances between multiple customer segments.

Tenzai employed a unique approach, to identify potential churners and then prioritize the right customers for retention. A two-stage stacked ensemble model was built, the first model was used to predict the propensity of customers who are most likely to churn. 

Then to prioritize the right customers for retention, customer lifetime value (CLV) for each customer was calculated using a regression model. Customers were then prioritized for retention campaigns based on their churn propensity and CLV score.

Why not use data to gain insights, guide efficiency efforts, and make predictions? Contact us.

Results

The segment-specific churn prediction model resulted in higher accuracy compared to a single model for the overall customer base. 

The segment-specific approach also helped them to devise targeted offers and messaging for each customer group. 

The value-based retention strategy helped the marketers to prioritize the top 10% of customers who were contributing to more than 45% of the revenues for retention campaigns. 

The new retention solution helped the client to improve the effectiveness of retention campaigns, reduce customer churn, and also devise strategic initiatives to reduce churn.

Key benefits include

  • The segment-specific churn prediction model had a high accuracy of greater than 90%.
  • Prioritizing customers for retention based on value and propensity score helped increase ROI on marketing spends by 35%.
  • Post-deployment, the client was able to reduce the annual churn rate from 36% to 25% resulting in revenues savings of millions of dollars.