"Who is likely to churn?"
Imagine that you can know which customer to is planning to leave. It would allow you to more smartly deliver marketing spend & thus save signficiant costs; the cost to acquire a customer is actually 5 times bigger
than the cost to retain a customer.
Invest in retaining your existing customers. It is therefore crucial to focus your attention to customers who are most likely to stop buying from you (i;e. churn). But how do you determine those customers? This is what we mean deliver with churn prediction.
What's Churn Prediction?
Churn Prediction is the process of predicting & identifying customers who are likely to churn (leave, stop buying) within X months. It’s proven to be a tangible use-case which often brings direct ROI to the business: e.g. Churn scores can be automatically added in e.g. Sales Force Automtion tools or CRM-systems and leveraged by sales or call center agents. By focusing on those customers, your retention budget can be spend in the best way possible.
From our experience, Churn is often highly driven by Recency, Frequency and Monetary value (RFM).
What do we understand with these measure:
- Recency: how recently did the customer purchase?
- Frequency: how frequent does the customer purchase from us?
- Monetary value: how much does the customer spend?
However, in order to make more accurate predictions, we should augment those dimensions with other drivers like seasonality, promotions and general ordering patterns of customers.
If we combine all these elements, we can use a mix of statistical techniques to deliver a (transparant) predictive model. We can then tag each customer with a likelihood to churn. By setting the right threshold, we can immediately detect the customers we should focus on in retaining and thus optimizing our retention spend.
For more information on how we can help you out, please contact us.