Churned users are the most honest audience you will ever have.
They have no reason to flatter you and every reason to tell you the truth, because they are already leaving. There is no relationship left to protect. For a few seconds, on the way out the door, they will tell you things a happy customer never would.
Most churn surveys waste that honesty.
The typical one is a single dropdown: "Why are you canceling?" with five options the team brainstormed in a meeting. Too expensive. Missing features. Not using it. Switching to a competitor. Other. The user picks one, clicks cancel, and the team files away a result that mostly confirms what they already believed.
That is the trap. A churn survey is not a satisfaction score. It is a short forensic interview about a decision that has already been made. Your job is to reconstruct that decision before the memory fades, not to grade your product against a list you wrote in advance.
Here is how I think about doing it well, and then exactly how we run one in Formaly.
Ask at the moment of cancellation, not a week later
Timing is the whole game. If you email a churn survey three days after someone cancels, you are asking them to reconstruct a feeling they have already moved on from. The answer you get back is a vague, sanded-down version of the real one.
Ask inside the cancel flow, while the decision is fresh and the frustration is still warm. The same question answered at the moment of cancellation versus a week later produces completely different data. One is a memory. The other is the actual reason.
It is a simple rule: the closer you are to the moment, the truer the answer.
The reason they click is rarely the reason they left
"Too expensive" is the most common churn answer, and it is almost always incomplete.
Price is rarely the real problem. "Too expensive" usually means "I did not get enough value to justify the price." Those are different problems with different fixes. One is a pricing page. The other is your onboarding, your activation, or your product.
A dropdown can never tell you which one it is. That is why the single most important question in any churn survey is the open-ended follow-up:
What is the main reason you decided to cancel today?
The structured choice tells you the category. The open text tells you the story. You need both, and the story is where the money is.
Structure: one reason, one story, one save attempt
A good churn survey is short. Churned users are doing you a favor, and drop-off climbs fast past a handful of questions, so three to five questions is the right size. I structure it like this:
- Primary reason (multiple choice). A clean, mutually exclusive list so you can trend categories over time.
- The open-ended why. "What is the main reason for that?" This is the question that earns its place.
- A specifics question. You do not need branching to get detail. One short optional question, such as "If a missing feature pushed you to leave, which one?", lets the people it applies to answer and lets everyone else skip past it.
- The save question. "What would have changed your mind?" This is the most underrated question in the whole survey, because it tells you what is recoverable.
Do not lead the witness
The fastest way to ruin a churn survey is to write questions that fish for the answer you want.
"What did we do wrong?" assumes you did something wrong and primes blame. "Was the price the issue?" hands the user an easy excuse and buries the real reason. Keep the wording neutral and let the answer come from them. "How would you describe what led to your decision?" gets you somewhere honest. A loaded question gets you a loaded answer.
Leave a door open
Some churn is recoverable, and the churn survey is your last cheap chance to recover it.
End with a low-pressure offer: a way to talk to a human, a pause option instead of a full cancel, or a simple "mind if we follow up?" Not everyone will take it, but the ones who do are often the most valuable conversations you will have all quarter. A churned customer who agrees to a fifteen-minute call will tell you more than a hundred satisfied ones.
How we do this in Formaly
That is the thinking. Here is the practical part: building and running this exact survey in Formaly, start to finish. It takes about ten minutes.
Step 1: Generate the survey from a prompt
Instead of dragging fields around, describe what you want. I start a new survey and type the goal in plain language, including the audience and the decision I am trying to understand.
Create a short churn survey for SaaS users who are canceling their
subscription. I want one multiple-choice question for the primary
reason, an open-ended follow-up asking why, an optional question about
which feature was missing, and a final question asking what would have
changed their mind. Keep it under five questions and keep the tone
respectful, not defensive.
Formaly turns that into a structured survey with the right question types and ordering, so you start from a real draft instead of a blank page.
Step 2: Tune the questions
The draft is a starting point, not the final word. This is where your judgment goes in. I tighten the multiple-choice options so they are mutually exclusive, make sure the open-ended "why" question is doing the heavy lifting, and keep the optional "which feature?" question short so it never feels like work.
This is also where you set the tone of the copy so the survey reads as curious, not wounded.
Step 3: Put it where the churn actually happens
A churn survey is only as good as its placement. You have two strong options in Formaly:
- In the cancel flow (recommended). Embed the survey on your cancellation page, or link to it from the "Cancel subscription" button, so people answer while the decision is fresh. This is the highest-signal placement there is.
- By tracked email invite. If you cannot touch the cancel flow yet, send a personalized invite immediately after the cancellation event so you can see exactly who responded.
Pick the one that matches where your users actually cancel.
Step 4: How you get the insights out
This is the part that makes the whole thing worth it. Once responses start coming in, you do not have to read a spreadsheet line by line.
Open the survey's analytics and Formaly does the first pass for you:
- AI summary groups the open-ended answers into themes, so you can see at a glance whether "too expensive" really means "weak onboarding."
- Sentiment shows you which reasons carry the most frustration, which is where churn is most preventable.
- The completion funnel shows whether people are dropping off the survey itself, which tells you if a question is too heavy.
- Cross-tabs let you ask the strategic question: do enterprise customers churn for different reasons than self-serve ones?
You can read the headline in thirty seconds, then drill straight into the raw answers behind any theme. More on that in Advanced Analytics in Formaly.
Step 5: Close the loop
The last step is the one most teams skip. Take the top recoverable reason from the summary and do something with it this week. Reach out to the people who said yes to a follow-up. Fix the one feature gap that shows up the most. Then run the survey continuously, not as a one-off, so you can watch the reasons shift as you ship.
The point
A churn survey is not about collecting reasons. It is about understanding a decision well enough to prevent the next one.
Ask at the right moment, get past the dropdown to the real story, keep it short, and actually read what comes back. Do that, and your most honest audience will quietly hand you your product roadmap on their way out.