Journey 8 · Churn prevention: the first sign of a slip

The decisioning matrix

Journey 8 · Defend

The churn-prevention journey

Step in at the first sign of a slip with two touches then rest: a value-led save on the customer's own rhythm, a second knock only if the first goes unseen, then designed silence to day 60 and a hand-off to win-back, replacing a day-50 calendar canvas measured at minus 10.2% against its own control.

This journey’s strategy

There is no subscription to cancel: a customer who is leaving simply goes quiet, and once lapsed most are gone for good. The churn-risk model spots the first sign of a slip, so we step in with one personalised reason to come back while they are still cheap to keep.

One window we own: the 60 days from the first sign of a slip to the hand-off to win-back, where today nothing happens until a generic email at day 50.
The model decides who enters and when, never a calendar timer; a customer it has not flagged is sent nothing.
Two touches, then rest: one personalised reason to come back on their usual route, a second only if the first went unseen, then silence until the hand-off.

The CRM brief: every comms change, today beside proposed

▶ Watch the customer experienceOpen full screen

The film of this journey as the customer receives it: every segment, outcome and channel state, with the change and the evidence at each moment. The static walkthrough below carries every change in full beside its evidence.

The static walkthrough

The same journey with every moment on screen at once: each change, its evidence and its full logic, moment by moment below.

One live defect sits under this journey: the in-app banner’s impressions and clicks do not register, so the only surface reaching about 830 no-channel customers a month is unreadable. The detail, with the delivery fixes: the Engineering brief.

ValidatedAnalytics validation requiredData / build outstanding

Choose the viewEvery change redraws the walkthrough below: one traveller, one channel state, one ending, one phase.

The D pins: the data the changes need

The pins on the walkthrough above mark the data each change needs, coloured by the sort of ask: COMPUTED TRAIT · ANALYTICS APP AND JOURNEY OWNED NEW EVENT · ENGINEERING BUILDS EXISTS · PIPE TO BRAZE READY · NOTHING TO BRIEF. Tap any pin, anywhere in the pack, for what it is. The full contracts, the computed traits included, live on the Engineering brief.

Pin	The ask	Where it stands
D1	COMPUTED TRAIT · ANALYTICS	Does not exist yet
D3	READY · NOTHING TO BRIEF	Live in braze
D4	READY · NOTHING TO BRIEF	Live in braze
D6	APP AND JOURNEY OWNED	Does not exist yet
D7	READY · NOTHING TO BRIEF	Live in braze

Who walks through this journey

The volumes behind the walk: how many customers take it, split by who they are and what we can reach them on. These numbers prioritise what gets built first.

Who	Push and email	Push only	Email only	No marketing channel	All channels
Established actives entering a save episode	1,089	116	594	42	1,841 since 20 Aug
of those, new customers: one win, then quiet (the first-win register)	TBC	TBC	TBC	TBC	not yet measurable
of those, returning power users (the recognition register)	TBC	TBC	TBC	TBC	not yet measurable
Still quiet at day 60, handed to win-back	TBC	TBC	TBC	TBC	not yet measurable

Replaces the consent-state proxy and corrects it sharply: 97.7% of entrants are reachable, against the proxy's 62.7%. A pool since measurement began on 20 August, not a stable monthly figure. The first-slip sub-splits need 60 days of history, so 19 October is the earliest they can be pulled. Source: Analytics (Ivi), 2 Sep 2026. The three rows below the head need 60 days of history from the 20 August measurement start, so 19 October is the earliest they can be pulled.