Solutions / Retention & churn

“Customers leave and we find out too late.”

Know who’s about to leave. And keep them.

Models that spot customers at risk weeks in advance, explain why, and trigger the right offer or call before they go.

Sound familiar?

If any of this is true, we can help.

  • You learn about churn when the cancellation arrives.
  • Retention offers go to everyone, including people who would stay anyway.
  • Your churn score says who, never why, so nobody knows what to do with it.

What we put to work

Three things, working together.

01

Predict who and why

Causal models that give each customer a risk and the specific reasons behind it.

02

Choose the action

A call, an offer, a fix to a complaint, or nothing at all. The cheapest action that keeps the customer.

03

Act and measure

Agents and your team act through the CRM; every campaign is measured against a control group.

Proof

Where we’ve done it.

Insurance · Retention

A causal retention engine tells a European insurer who will leave, why, and which offer keeps them.

Client anonymised. Full case file and reference available under NDA.

Read the case study
+23%customer retention
−18%churn
−15%retention operating cost
Quarterlymodel retraining
70%

default rate in the segments a model flagged early across a €10bn book at a tier-1 European bank.

Your size, your version

For a 40-person company or a 40,000-person group.

SMEs & mid-market

Ready-made, live in weeks

A churn model on your customer data with a monthly list of who to call and what to offer.

Setup fee + monthly plan · How it works
Enterprises & PE portfolios

Built into your stack

Causal retention engines integrated with your CRM, pricing and contact centre, retrained every quarter.

Value Map → first system → run & scale · How it works
Connects to what you already use
CRMPolicy / billing systemsData warehouseContact centreEmail & messagingPricing engine

Questions

What people ask us.

How much data do we need?+

Usually what you already have: contracts, payments, claims or orders, and interactions. We check it in the first two weeks.

Why causal and not a normal churn model?+

A normal model tells you who correlates with leaving. A causal one tells you what would change their mind, so the offer works.

Will the team understand the output?+

Every score comes with its reasons in plain words, in the CRM screen they already use.

Other problems we solve

Let’s fix this one.

Start with a free Value Map: three hours with your team, and a fixed-price plan for your first system five days later.