CASE STUDY
Finding 86% of Potential Churners Brings Down Revenue Leakage and Customer Acquisition Costs
Leading mobile service provider uses AI to successfully target at-risk churners, trimming budget costs and reducing revenue leakage.
Churn prevention is a significant concern across the telecom industry. Finding insights into why customers leave their service provider – and what makes them stay – is essential for successfully competing in the market.
Our client, a multinational telecommunications company with a strong presence in Africa and the Middle East, needed more profound insight into their customers’ churn propensity. Precisely, they needed answers to questions like:
Which customers are at-risk to churn? What is their behavioral segment? Of these, who can we retain?
What can we do to retain and increase high-value customers?
What specific actions can we take to reduce churn for each customer segment?
The answer required a comprehensive study of customers’ usage and creating behavioral micro-segmentation, which is leveraged in multiple predictive modeling, and a self-learning & heuristic-based recommendation engine that could produce the most effective retention tactics and offers.
Targeted Offers Save Churners (and Revenue)
We used a customized NAVIK MarketingAI to find at-risk customers and map them to the most effective offers. This resulted in the following gains:
- Identified 86% of potential churners in the top 3 deciles.
- Prevented churns led to 5% revenue savings (estimated).
- Effective utilization of marketing budget by targeting customers having a high probability to churn also saves customers acquisition cost (by retaining customers)
In addition, the models also highlighted areas of growth within the current customer base utilizing cross-sell and upsell opportunities. For example, it found 60% of potential upgraders in the top 3 deciles, which would contribute an estimated 7% increase in potential revenue. Additionally, merging the output of Churn analysis with CLTV and Profitability helped us identify High Risk-High Value customers that should be preferred for sending offers.
AI, Machine Learning Provide Deep Insight into At-Risk Customers
To make this happen, we used a customer value management analytics engine to isolate churn factors, identify at-risk customers, and create targeted offers. Behavioral micro-segmentation formed a vital part of the strategy: We used it to group customers into segment based on their lifetime value, life stage, usage, etc.
The next step was to develop cross-sell, upsell, and retention offers that mapped with the services each customer would most need (e.g., additional high-speed data, more minutes, etc.). To do this, we employed multiple predictive models. This was combined with the results from the customer profitability and CLTV models and the output was fed into the optimization algorithm. The final step was generating campaign lists with targeted offers for each micro-segment.
A Powerful Ally Against Churn
Thanks to the NAVIK MarketingAI platform, our client can now use personalized outreach and offers to connect effectively with each part of their customer base. They have a clear picture of the churn risk for various subscribers and can predict customer behavior around key business objectives. This allows them to save money on their marketing campaigns, but more importantly, it also helps them hang on to their customers.
