Churn – loss of customers – is a major risk area for D2C and subscription businesses. Churn models help you identify high churn risk customers so you can act to retain them
Customers produce revenue and if they leave, revenue will fall. Many businesses operate on a profit margin of 10-20%. If you lose 20% of your customers, your profits could potentially fall (depending on how active the lost customers are), and the business could incur increased losses.
Churn can be minimised by identifying which customers are likely to churn and by taking pre-emptive action to minimise losses.
But to do this it is important to identify which customers are most likely to leave. We identify these customers using Churn models. Churn models look at the customers who have churned recently and measure their characteristics.
Customer Churn Drivers
Possible churn indicators include:
- Declining frequency of purchase
- Time since last purchase
- Declining revenue
- Declining engagement
- Visits to “close account” pages
- Payment type
- Contract length
- Need for previous win-back activity
- Number of products or services held
- Postcode District or Sector
- Not responding to information update requests
Churn Modelling
We build machine learning models to quantify and predict the risk of churn for each customer using the types of variable listed above. When we have a measure of the importance of each of the variables listed, we can ‘map’ those to each customer record to produce a churn risk score. Customers can be segmented using these scores and you can engage in Social, eMail or Direct Mail activity to hold these customers.
Contact for more info HERE