Mercedes-Benz used a machine learning algorithm to drive repurchase, ultimately achieving both the best-ever ROI and strongest repurchase rates all within an increasingly challenging market.
The Australian car market is burdened by long purchase cycles, disconnected customer experiences and blurred product segments. With increasing competition from other manufacturers, existing Mercedes-Benz customers have more choice than ever before when it comes to purchasing a new car; there are 67 brands fighting for 1.1 million annual Australian sales whereas in the US just 53 brands compete for well over 15 million.
Due to a period of recent growth, Mercedes-Benz Australia had a large number of new-to-brand customers – those who had only ever purchased one Mercedes-Benz. Ensuring these drivers repurchase and become loyal was vital for the long-term health of the brand, but we had an incomplete understanding of who they were.
The Repurchase Reminder strategy was developed to identify and target existing customers who were most likely to purchase a new Mercedes-Benz. There was already a model in place but it was simple and there was room for optimisation. The new predictive model used the previous version as a basis, and overlaid multiple data points such as third party geodemographic data and Mercedes-Benz Financial Services data to improve the performance of the model and create a single more holistic view of the customer. The top 10% of customers who the model said were most likely to repurchase received a series of personalised email and direct mail pieces.