Customer & Marketing Data Science case study
Multi-channel customer churn modelling and lifetime value forecasting
An anonymised summary of the challenge, approach and outcome. Client names and measured figures are published only with written consent.
Anonymised client work
We do not publish client names or identifying details without written consent, and we do not publish numbers we cannot substantiate. The summary below describes the shape of the work without exposing the client.
Client
Customer & Marketing Data Science sector organisation, identity withheld pending consent
Challenge
A high-volume digital subscription business struggled to detect early customer disengagement signals across fragmented web, mobile, and customer support touchpoints.
Solution
Acmez built feature store pipelines and gradient-boosted survival analysis models scoring churn probability and projected LTV, triggering personalized retention workflows across CRM channels.
Outcome
Retention teams received actionable risk tiers 45 days prior to contract renewal, improving intervention timing and customer cohort retention rates.
Capability involved
Next step
Facing a similar challenge?
We can describe comparable work in a private conversation even where it is not published publicly.