Generally, it's more difficult and expensive to acquire new customers and win back attriting customers than to retain existing customers. Thus, it’s important for companies to understand the reasons customers churn and estimate the risk associated with individual customers. But often marketers find it challenging to identify which customers will drop out.
Areas of improvement
Goal
To effectively and efficiently retain the spending of an attriting customer
FocusKPI provides the output of a predictive churn model as a measure of the immediate or future risk of customer cancellation. Our predictive behavior analytics use your data to identify the customers that are at a high risk of attrition so you can target these customers with retention measures before they drop out.
Data Sampling
Model Universe: around 695,000 online customer sites in P5 FY14 with tenure 12+ periods
Model Sample
Business Sites: 87,890, 13% of total universe
Revenue Attrition Customers: 15,996, 18.2% of Model Sample
Non-revenue attrition customers in next 3 periods: 71, 894, 81.8% of Model Sample
Variable Selection
Over 4,000 variables have been tested in the modeling development. 51 variables have been identified to be significant drivers
Key Drivers Identification
Score and identify 10 key drivers from a total of 51 drivers
Customers with Risk Ranking
Rank customers with risk (who are most likely to stop repurchasing papers)
In prioritizing spending on retention efforts, results can be greatly improved via an emphasis on an early intervention based on customer behavior.
Recommendations
Generally, it's more difficult and expensive to acquire new customers and win back attriting customers than to retain existing customers. Thus, it’s important for companies to understand the reasons customers churn and estimate the risk associated with individual customers. But often marketers find it challenging to identify which customers will drop out.
Areas of improvement
Goal
To effectively and efficiently retain the spending of an attriting customer
FocusKPI provides the output of a predictive churn model as a measure of the immediate or future risk of customer cancellation. Our predictive behavior analytics use your data to identify the customers that are at a high risk of attrition so you can target these customers with retention measures before they drop out.
Data Sampling
Model Universe: around 695,000 online customer sites in P5 FY14 with tenure 12+ periodsModel Sample Business Sites: 87,890, 13% of total universe Revenue Attrition Customers: 15,996, 18.2% of Model Sample Non-revenue attrition customers in next 3 periods: 71, 894, 81.8% of Model Sample
Variable Selection
Over 4,000 variables have been tested in the modeling development. 51 variables have been identified to be significant drivers
Key Drivers Identification
Score and identify 10 key drivers from a total of 51 drivers
Customers with Risk Ranking
Rank customers with risk (who are most likely to stop repurchasing papers)
In prioritizing spending on retention efforts, results can be greatly improved via an emphasis on an early intervention based on customer behavior.
Recommendations