How We Raised Click through Rate by 50% with a Simple Change

By on March 14th, 2015 in Machine Learning
TrueAccord Blog

The technical and analytical vision behind TrueAccord is to add data-driven decisions to the communication model in debt colelction. Digital communication enables better data collection, and better understanding of customer behavior patterns. We can collect and observe open, click and browsing patterns that sometimes do more to explain how to engage with a customer than explicit communication. By connecting customer behavior to their mental state, and responding to that state with corrent language, we were able to substantially increase engagement on our collection communication.

Data allows us to segment customer types to identify the equivalent of “buying intent” – intent to settle a debt – without harassing the customer. We saw different segments of customers, but also different stages in the “buying” process. Customers go through a mental process leading up to paying a debt. They need to get informed about the obligation, accept TrueAccord’s involvement, negotiate, and finally feel comfortable with the payment. We’ve developed a model to track these changes in the customer’s disposition and commitment to paying a debt.

Once we understood that the model applies to debt collection, we found differences in how customers respond to language and design elements differently based on who they are, but also their stage of the process. Initially, our intuition was that with every communication, the call to action should be “Pay Now”. We were trying to get customers to pay ASAP. Since there are different stages to the customer’s commitment, this CTA didn’t make sense for each stage, and our click through rate was underperforming for specific subsegments. Once we realized that, we changed the CTA in some stages to “View my statement”.

The result? An immediate 50% rise in click through rate on emails taht were sent to customers in lowcommitment stages. They were more likely to get informed, respond back, and eventually pay – because we didn’t force the issue. We were able to do that because we have the data to track customer behavior, but also because we’re not forced to aggressively collect as much as possible, as early as possible. Our clients understand that data-driven, flexible user experience increases recovery in the long term.