Retention Foundation

Churn diagnosis: find what is breaking before it compounds

Walk through a structured analysis of your churn patterns, identify the leading indicators of departure, and build an early-warning system for at-risk customers.

5 steps

55 min total

Foundation

Track progress as you complete steps

Step 1 of 5

Pull and structure your churn data

15 min

The diagnosis begins with the data. You need at least 90 days of churn records to identify patterns. If you have less than that, use all available data and note the limitation. The output of this step is a structured churn log that lets you analyse by cohort, channel, plan, and time.

Monthly Churn Rate = Customers lost in month / Customers at start of month x 100

Action items

Export all customer cancellations or non-renewals from the last 90 days

For each churned customer, record: signup date, cancellation date, revenue at churn, acquisition channel, plan or tier, and any cancellation reason captured

Calculate your overall churn rate for the period

Monthly churn rate = customers lost in month / customers at start of month

Identify which 20% of your customer base accounts for 80% of churned revenue

Unlock to continue

You have seen how it starts. Unlock the full playbook to finish it.

Choose the option that works for you.

$4.99

This playbook only

One-time. Yours forever. No subscription required.

Best value

$49.99/mo

Growth subscription

All 8 playbooks + Dashboard + deep-dive reports.

Step 2 of 5

Identify churn patterns across four dimensions

15 min

Raw churn numbers are less useful than patterns. You are looking for whether churn is concentrated in a specific segment, a specific time window, or a specific acquisition channel. Pattern concentration tells you where to intervene.

Tenure pattern

When in the customer lifecycle is churn highest? Month 1? Month 3? After 12 months? Early churn usually signals onboarding failure. Later churn usually signals value delivery failure.

Segment pattern

Is churn concentrated in a specific plan tier, company size, or industry? High churn in a specific segment often means the product is solving a different problem than that segment actually has.

Channel pattern

Do customers from one acquisition channel churn more than others? Channels that produce cheap customers often produce low-fit customers with higher churn.

Reason pattern

If you collect cancellation reasons, what are the top three? These are the symptoms. The diagnosis requires going deeper.

Action items

Sort your churn data by tenure and identify when the largest concentration occurs

Cross-reference churn rate by plan or segment to find the highest-churn cohort

Cross-reference churn rate by acquisition channel

List the top three stated reasons for cancellation (if data exists)

Step 3 of 5

Interview churned customers

15 min

Data tells you where churn is happening. Interviews tell you why. The stated reason at cancellation is rarely the real reason. A short conversation with three to five recently churned customers will produce more diagnostic insight than any amount of data analysis.

Action items

Email 8 to 10 recently churned customers asking for 15 minutes of honest feedback

Offer a small incentive if needed. Frame it as helping you improve, not as a retention attempt.

Ask: what made you sign up originally? What outcome were you hoping for?

Ask: at what point did you feel the product was not delivering that outcome?

Ask: what would have needed to be different for you to stay?

Record verbatim phrases, not paraphrases. The exact words matter.

What you are listening for

The gap between what customers expected when they signed up and what they actually experienced. This gap is almost always visible in the first 30 days. It is either an onboarding failure (they never reached the value moment) or a messaging failure (the product was never right for this customer, but the marketing suggested otherwise).

Step 4 of 5

Build your early-warning signals

10 min

Most churn is predictable. Customers do not suddenly decide to leave: they gradually disengage. The goal here is to identify the two or three behavioural signals that appear in your data before customers churn, so you can intervene before the decision is made.

Action items

Look back at churned customers: what product activity (or inactivity) did they show 30 days before cancelling?

Define your two leading indicators of churn risk based on this data

Set up an alert or report to flag customers showing these signals automatically

This can be as simple as a weekly spreadsheet review or as sophisticated as an in-app trigger.

Step 5 of 5

Create your 30-day intervention plan

10 min

With your diagnosis complete, define the specific interventions you will run over the next 30 days. Prioritise based on where your pattern analysis showed the highest concentration of churn.

Action items

Define one onboarding improvement if early churn (0 to 30 days) is your primary pattern

Define one proactive outreach action for customers who trigger your early-warning signals

Define one win-back campaign for customers who churned in the last 60 days

Win-back rates for recent churners can be surprisingly high when you address the specific issue they left over.

Set a churn rate target for 60 days from today and schedule a review

One rule

Work on one churn intervention at a time. If you improve onboarding and launch a win-back campaign simultaneously, you will not know which one moved the number.

Playbook complete.

Run the Growth Audit again to see how your score has moved since you started this playbook.

Audit my score Next: Referral Programme