The company is a file-sharing subscription service with over 10Mpaying subscribers and 500M users spread across its consumer and enterprise offerings. The company suffered from high rates of accidental churn—subscriber drop-off caused by a failure to collect payments. Despite a healthy user acquisition funnel, the company struggled with user retention on its platform with rates hovering below 85%monthly.
The company needed a solution to address the leaky bucket of silently churning users. For every 100 previously paid users, only 35 had explicitly asked to cancel their subscriptions. The company was bleeding users who had not actively requested to stop using their product—causing pain both for the company, and for its customers who were unaware that their access to the service was at risk.
The company proactively prevented payment lapse by identifying at-risk users and updating payment information prior to scheduled billing dates. An ML-based approach to scheduling payment collection attempts was implemented—including:
Targeted interventions led to a massive decline in accidental churn at the company—from 65% of overall churn to just 25% for its enterprise product. The company realized massive improvements in the rate of payment updating for lapsed users—from 10.9% to 33.5% on Day 0 and from 15.7% to 43.4% on Day 24. As a result of these cumulative changes, the company realized a massive benefit in revenue saved that accounted for $22M in top-line annualized recurring revenue—$14M on the enterprise side and $8M on the consumer side of the business.