Significant increase in response rate
Historical conversions or demographic information
Plausibility analysis and bootstrapping
Customer analysis for VR Bank Mitte eG
VR-Bank Mitte eG is striving to optimize its sales processes in the area of customer approach and support. Specifically, the goal is to assess and evaluate customers’ latent interest in a particular product in advance of a sales campaign.
Machine learning algorithms are used to calculate the affinity of a customer for the product to be marketed. A high affinity promises an increase in the response rate and a more efficient use of resources in marketing and sales.
To achieve a reliable score, eoda brings together 20 different data sources – historical conversions or demographic information. Numerous features are prepared or generated for later modeling. To determine the score, an ensemble of 1,000 classification trees with previous bootstrapping is formed. With a visual plausibility analysis, the predictions of the algorithm could be visually validated.
After only a short campaign period, a significant increase in the response rate could be observed. Thanks to eoda’s customer affinity analysis, VR-Bank is able to talk to the right customers about the right topics even more frequently. Targeted and effective sales activities increase revenue potential, reduce costs and increase customer satisfaction.