The Challenge
The client found it challenging to identify key controllable and uncontrollable factors that drives customer churn at the dealer level. This in turn made it impossible to develop effective strategies to reduce churn.
Our Solution
A combination of driver analyses and customer lifetime value analysis was determined as the best approach. Initially focusing on a single dealer, our team built a data mart, segmented customers based on spend patterns, and identified drivers to generate a final forecast model. An action item matrix was then developed to help client prioritize retention efforts and also in determining ROI for the retention campaign.
In any competitive market, customer churn is a concern. It becomes particularly important to be able to predict and mitigate churn in businesses like heavy equipment manufacturers, where long lead times make it difficult to respond dynamically to changes in demand and customer base. Both short- and long-term strategies are needed to increase customer engagement and bring down the rate of customer churn.
Business Challenge The client wanted to design effective retention strategies to reduce their customer churn rate. This could only be done by gaining a better understanding of customer pain points and identifying controllable and uncontrollable drivers of churn. Disparate data sets and complex business environment made this tough.
Evalueserve Solution Initially, a model was created for one specific dealer based in the U.S. This was tested and refined, and then could be applied to the full set of 50 dealers. A data mart was created for the dealer. Customers were segmented as unlikely to churn, volatile or churned, based on their spend patterns, and their behavior was regressed against several parameters to identify the driver variables. “Results from the models developed were used to develop an action item matrix which helped the client prioritize retention efforts and also in determining ROI for the retention campaign.” These parameters included controllable factors (e.g., service quality and frequency and mode of interactions) and uncontrollable factors (e.g., macro-economic indices, competitive intensity). Each driver variable was checked for significance in correlation, lead-lag and seasonality. Multiple statistical validation metrics were applied to select the most appropriate model. The final model predicted with ~85% accuracy customers likely to churn in short and long term. The results were used to then develop a priority matrix which helped client in designing cost efficient retention campaigns. We also developed an interactive dashboard (Figure 1) that enables continuous monitoring of customer churn rates and success of retention campaigns. Benefits Achieved It was estimated that the pilot dealer could have used the framework to generate additional sales opportunities of up to $25 million through churn prevention. The client was very satisfied with the results and insights generated. The analysis is now being extended to other dealers within the client’s sales and distribution network.
“Results from the models developed were used to develop an action item matrix which helped the client prioritize retention efforts and also in determining ROI for the retention campaign.”Business Impact
It was estimated that the pilot dealer could have used the framework to generate additional sales opportunities of up to $25 million through churn prevention. The client was very satisfied with the results and insights generated. The analysis is now being extended to other dealers within the client’s sales and distribution network.
