A churn interview template for SaaS founders

A familiar churn-research pattern goes like this.

A bootstrapped founder pulls the last quarter’s cancellations out of Stripe. Twenty-three customers left. Eleven picked “too expensive.” Six picked “switching to a different tool.” Three picked “not using it enough.” Three picked “other” and left the box empty.

The founder spends the next two weekends building a cheaper plan. Six months later, churn is exactly where it was, and the new plan is mostly absorbing customers who would have paid the old price.

The cancellation form was not lying, exactly. It was just the wrong instrument for the question. “Too expensive” might mean the product was genuinely overpriced. It might mean the customer never reached the promised outcome. It might mean their company cut budget. It might mean a competitor mapped better to their workflow. It might mean they stopped using the product eight weeks before the invoice became annoying.

A churn interview is how you find the real story.

This template is for tiny B2B SaaS teams: post-launch, fifty to a thousand users, no dedicated research team, and not enough time for a heavyweight retention programme. You want the next product decision to be based on what actually happened, not on a dropdown reason in a cancel flow.

The method is simple: ask about the past, reconstruct the switch, separate the type of churn, and listen without trying to win the customer back during the call.

Why cancellation forms hide the real reason

A cancellation form catches a customer at the worst possible moment for careful reflection. They have already decided to leave. They want to finish the task. They may be annoyed, embarrassed, rushed, or simply done.

That context shapes the answer.

Broad reason codes are useful for routing and rough reporting. They are poor evidence for product strategy. Churnkey’s documentation on structured follow-up questions gives a clean example: a customer who chooses “too expensive” might mean company budget cuts, weak ROI, price too high for the value, or a cheaper alternative. Those are four different problems with four different fixes.

Churnkey’s cancellation-flow guidance also argues for keeping the flow short. That may be sensible for a save flow, but it is also a warning about research quality: the shorter the flow, the less explanatory depth you collect.

Paddle’s guide to why customers churn makes the same point from the retention side. Customers leave for wrong-fit acquisition, missing outcomes, weak support, competitor pull, bugs, weak perceived value, and price. The cancel form may show the final label. It rarely gives you the causal chain.

The form should start the investigation, not end it. It tells you who left and when. It does not tell you why.

The rule: anchor everything in the past

The first rule of churn interviews is the same rule Rob Fitzpatrick teaches in The Mom Test: ask about their life, not your idea; ask about specifics in the past, not opinions about the future; talk less and listen more.

Churn is a good fit for this rule because the story is recent. The last useful session happened on a real day. The renewal conversation happened in a real Slack thread. The workaround started before the customer cancelled. You are not asking them to predict behaviour. You are asking them to walk you through behaviour they already had.

Teresa Torres makes this practical in her guide to story-based customer interviews. People are poor at summarising their own behaviour in the abstract. They are much better when you ask them to tell a specific story.

So do not open with “Why did you cancel?”

Open with:

• “Can you walk me through the last time you used the product?”
• “What were you trying to get done?”
• “What happened next?”
• “When did you first notice it was not fitting the way it used to?”
• “What were you doing instead in the weeks before you cancelled?”

These questions feel slower, but they get you closer to the truth. A customer may not know the root cause when asked directly. They can usually describe the last few weeks in enough detail that the cause becomes visible.

Use the switch interview for leavers

Once you understand the recent past, reconstruct the switching decision.

Bob Moesta’s Jobs-to-be-Done switch interview is useful here because churn is a switch in reverse. In his Intercom interview on unpacking customer motivations, Moesta points teams toward recent buyers and recent leavers because both groups have fresh decision stories. One chose you. One chose something else.

The four forces of progress give you a simple map:

• Push: what made the current situation unacceptable?
• Pull: what made the alternative attractive?
• Anxiety: what made switching feel risky?
• Habit: what kept the customer in the old way longer than expected?

For churn, “the old way” is often your product. The push is the friction, weak outcome, confusing workflow, missing integration, or internal change that made staying harder. The pull is the competitor, spreadsheet, agency, internal tool, or “do nothing” path that seemed easier. The anxiety is what they worried they would lose by leaving. The habit is what kept them paying for another month after the value had already faded.

Moesta’s Demand-Side Sales framing calls the trigger a struggling moment: the point where someone starts believing the current situation is no longer good enough. In churn research, your job is to find that moment.

Ask:

• “What was pushing you away from the product?”
• “What did the alternative promise?”
• “Was there anything about leaving that made you nervous?”
• “What kept you paying as long as you did?”

Do not treat the four forces as a questionnaire. Treat them as a checklist for what you need to understand before the call ends.

Talk to almost-churned customers too

Do not only talk to people who cancelled. Customers whose usage dropped by half, downgraded, hit a billing objection, changed champions, or went quiet after onboarding often have the churn story without the final break.

Teresa Torres argues on Churn.fm that continuous discovery is well suited to churn because churn is not a one-off research question. It is an evergreen product-health signal. For a tiny team, the practical version is one recurring interview slot each week, filled from three lists:

• Customers who cancelled in the last 14 days.
• Customers who downgraded or nearly cancelled.
• Customers whose usage dropped sharply after previously healthy use.

The third list is usually the best early-warning system. Cancellation is a lagging indicator. Behaviour change is a live story.

Classify the churn before you fix it

Not every cancellation deserves the same response.

Lincoln Murphy’s churn classification framework is useful because it asks two blunt questions: was this avoidable, and did you see it coming? An expected, unavoidable departure is different from an unexpected, avoidable one. The first may be acceptable. The second is a product or customer-success failure you need to understand.

You also need to separate voluntary and involuntary churn. Paddle’s guide to voluntary and involuntary churn defines the split plainly: voluntary churn is an active decision to leave; involuntary churn happens when payment or operational failures end the subscription without that intent. Recurly notes that some industry reports put involuntary churn at 20-40% of customer churn, though the exact share depends heavily on customer type, payment method, and billing setup.

That distinction changes the research plan.

Voluntary churn needs a conversation. Involuntary churn needs billing instrumentation, dunning, card updates, retry timing, and clearer payment communication. A customer who finished a one-time job may need a win-back path or a different pricing model, not a roadmap fix.

Before you schedule calls, tag each cancellation:

• Voluntary: the customer actively cancelled or chose not to renew.
• Involuntary: payment failed, card expired, or the account lapsed without a clear decision.
• Completed job: the customer achieved the thing they came for and no longer needs the product.
• Wrong fit: the customer should probably never have bought.
• Unknown: you need more information.

Only the voluntary and unknown buckets are obvious candidates for a 30-minute churn interview.

The 30-minute churn interview script

Here is the working template.

1. Open: remove the save attempt (2 minutes)

“Thanks for talking with me. I’m not here to win you back on this call. I want to understand what happened from your side, especially where the product stopped fitting. There are no right answers, and blunt feedback is useful.”

Most leavers expect a save attempt. Naming that you are not doing one lowers the pressure. ChurnZero’s exit-interview Q&A makes this explicit: the interview should be for understanding the decision, not for trying to sell the customer back.

2. Reconstruct the last useful session (5 minutes)

“Can you walk me through the last time the product was genuinely useful?”

Probe for:

• What they were trying to do.
• Who else was involved.
• What outcome they got.
• What made it useful.
• What happened after that.

This establishes the value you used to provide before you investigate where it broke.

3. Reconstruct the last frustrating session (6 minutes)

“Now walk me through a recent time when it did not work the way you needed.”

Probe for:

• What they expected.
• Where they got stuck.
• What they did instead.
• Whether they told anyone internally.
• Whether support, docs, onboarding, or the product itself could have helped.

Do not explain. Do not correct. Just listen.

4. Find the trigger (5 minutes)

“Tell me about the moment you decided not to continue.”

Probe for:

• The day or week it happened.
• The conversation around the decision.
• The person who had the final say.
• The budget, timing, or workflow event that triggered action.
• Whether they had already emotionally left before they cancelled.

Many customers cancel long after the real decision was made.

5. Map the four forces (8 minutes)

Ask the four questions one at a time:

• “What was pushing you away?”
• “What were you going to use instead?”
• “What worried you about leaving?”
• “What kept you paying as long as you did?”

You want the forces that moved them and the forces that delayed them.

6. Ask the return condition carefully (3 minutes)

“If we wanted to earn your business again in the future, what would have to be different?”

This is the only forward-looking question in the template. Keep it concrete. If they say “better onboarding”, ask what needed to happen in their first week. If they say “lower price”, ask what value would have made the old price feel reasonable.

7. Close with positioning (1 minute)

“If a peer asked you what we do, what would you tell them? And what would you warn them about?”

That final answer often gives you sharper positioning data than the rest of the call.

Should the founder run the call?

If you are the only person available, yes. It is better to run the call than to avoid the conversation. But recognise the cost: a churned customer is speaking to the person whose product they left. That changes what is easy to say.

Indi Young’s Listening Deeply is a useful corrective. The listener’s job is not to defend, explain, categorise, or sell. The listener helps the person describe their own thinking in their own terms.

That means:

• Do not correct misunderstandings during the call.
• Do not promise that a feature is coming.
• Do not ask whether they would return if you changed one thing.
• Do not debate whether their use case was a good fit.
• Do not turn the interview into a reactivation attempt.

If a save attempt makes sense, send it separately later. The interview is for learning.

This is also a place where Maren can help. A churned customer may be more direct with a neutral AI researcher than with the founder whose product they just cancelled. Maren can run the same structured conversation across many leavers, probe for specific stories, and bring back themes.

What to do with the answers

After each call, write a five-part summary while the details are fresh:

• Last useful moment.
• Last frustrating moment.
• Trigger.
• Replacement.
• Return condition.

Then add two tags: churn type and primary theme.

After eight to twelve interviews, look for repeated stories rather than repeated labels. “Too expensive” is not a theme. “Never reached value before the first renewal” is a theme. “Champion left and nobody else knew how to use it” is a theme. Each one points to a different fix.

Use the customer’s words when you share the finding. Verbatim language is more persuasive than a tidy internal summary, and it often becomes better onboarding copy, pricing-page copy, lifecycle email, or sales enablement.

Finally, reconcile what people said with what they did. If churners say onboarding was confusing, check activation events, support tickets, docs searches, and first-session drop-off. If they say price was the issue, check usage depth, feature adoption, plan limits, and whether the account ever reached the outcome they paid for.

The interview gives you the story. The behavioural data tells you where to look next.

The template in one page

Use this when you need the short version.

Churn interviews are one of the cheapest research tools a SaaS founder has. The customer already lived the story. You do not need a clever question. You need to ask them to walk you through what happened, in the past tense, without making them responsible for your feelings.

The cancellation form told you they left. The conversation tells you what to do next.