How Typeform's Cancellation Flow Could Say Goodbye Better
+20 XPCase study
How Typeform's Cancellation Flow Could Say Goodbye Better
A team that no longer runs surveys cancels its Typeform plan and walks through every step of the offboarding flow. You will see which steps push people away, which ones help, and how a warmer ending can keep a customer likely to come back.
+20 XP when you reach the endWhat you will learn
- Show what people will lose in a few points, not a wall of features
- Replace mandatory, guilt driven surveys with short, useful questions
- Use the reason someone leaves to write a personal goodbye
Scroll to follow the story. The screen changes with you.
Based on the learnings from Growth.Design: Typeform: How to offboard users the right way, presented in our own format.
Step 1 of 11
- Cancel my plan
01
The setup
The team has used 827 of 10,000 responses, so it is time to cancel.
After wrapping up their last survey, the team sees nobody is using Typeform anymore. Finding billing takes a detour: the profile menu opens on 'Preferences' rather than the account. Once there, the Plan and billing page is clear, with a visible 'Cancel my plan' link and the date of the next invoice.
TakeawayMake it easy to find the cancel option. Hiding it only makes people angrier when they do.
Nineteen red crosses02
Problem 1
A list of nineteen lost features tries to make leaving feel costly.
The first step lists everything the team will lose, each line marked with a red cross. It is meant to trigger loss aversion, but it misses: they already know they will lose features, the list reminds them they paid for things they never used, and it is so long that most people skip it.
TakeawayPressure that ignores why someone is leaving rarely changes their mind.
Loss aversionLosing something feels roughly twice as painful as gaining the same thing feels good. Losing 20 dollars hurts more than finding 20 dollars pleases. It only works when the loss matters to the person.
Original Too long to readFocused 
What really matters03
Repair
Show only the core differences, and people will actually read them.
A focused version keeps three lines: you lose access to your previous reports, you are limited to 10 responses a month, and you lose the pro features. It is quick to scan, and it points to the thing that matters most to this team, their reports.
What changed
- Nineteen lines become three
- The most valuable loss, past reports, comes first
- Minor features collapse into one link
TakeawayName the one or two losses that matter, and link to the rest.
- Data-Access plan
04
What works
A cheaper plan just for keeping reports matches what the team wants.
Beside the list, a green box asks whether they would be interested in a Data-Access plan for 10 dollars a month to keep and export their responses. That is exactly their worry, so they say yes.
TakeawayThe best retention offer solves the specific reason the person is leaving.
Lifecycle opportunitiesPeople leave for a reason. Once you understand their hopes, pains and barriers, you can find new offers that fill a gap in their journey. This applies to every stage, not only offboarding.
Required to leave05
Problem 2
The downgrade is not complete until you finish a survey.
The only way forward is 'Continue to survey', and the intro screen promises it will be quick. For someone who has already decided, a required survey feels like a guard at the door.
TakeawayMake exit surveys optional, short and clearly useful, or the answers will be noise.
The doorman surveyMandatory steps that are not needed block a natural flow and create reactance. Aim for fewer, better answers. Forced surveys collect hundreds of replies like 'didn't need it', which are hard to learn from.
- Waving goodbye
06
Problem 3
A sad illustration tries to make the user feel bad for leaving.
Above the text, a figure waves a handkerchief goodbye. It is a small touch, but it leans on emotion to make the person feel guilty about a reasonable business decision.
TakeawayA simple, friendly goodbye works better than trying to make people feel sorry.
The professional mournerUsing emotion to make someone feel bad for leaving is an exaggeration everyone sees through. It is meant to seem more human but reads as manipulative.
Pick one, any one07
Problem 4
The first required question has no answer that fits.
'Will you replace Typeform with an alternative tool?' offers a scale from 'Definitely' to 'Definitely not'. The team does not need a tool now, but might one day, so they pick an answer at random.
TakeawayIf people cannot answer honestly, the data you collect is not worth much.
The actual reason08
What works
The second question finally captures why they are leaving.
A list of specific reasons includes 'The project I was working on has finished', which is exactly right. This is the valuable answer, and it could have been the first and only question.
TakeawayAsk the one question that explains the churn, and put it first.
New use cases09
Problem 5
Good ideas for new uses appear only after the user has decided to go.
Typeform suggests registration forms, feedback and market research as other ways to use the product. The suggestions are relevant, but by now the team just wants to finish, so they click 'No thanks'.
TakeawaySuggest new uses when usage starts to drop, not during cancellation.
Nurture adoptionA key retention strategy is making sure current customers benefit from all your features. Prevent churn instead of reacting to it: introduce new use cases when you detect a decline in usage.
Cold 
A flat endingWarm 
Personal goodbye10
Repair
Use the reason they gave to write a warmer last screen.
After an optional comment box and a confirmation step that could have come first, the flow ends with a flat 'Your cancelation is confirmed.' A better ending uses the answer the team gave: 'Hope to see you for your next project!', plus a one click way to hear about the Data-Access plan.
What changed
- The headline reflects the reason they left
- A friendly illustration replaces the plain icon
- A light question keeps a door open for return
TakeawayThe last screen shapes how people remember the whole product. End it well.
Peak end rulePeople judge an experience mostly by its peak and its end. Even customers who leave can recommend a product for years if the last interaction felt meaningful.
Start Easy to startBetter end Door left open11
The result
A clear cancel link, then friction at every step after it.
The mood started fine at the billing page, dropped at the long loss list and the mandatory survey, rose briefly at the relevant suggestions, and ended low on a cold goodbye. Growth.Design scored the offboarding experience a D.
What changed
- Short, focused loss summary
- Optional survey with the key question first
- Relevant offers based on the reason
- A personal, warm last screen
TakeawayOffboarding is part of the product. Every step either protects your reputation or spends it.
Every takeaway, in one place
- The setupMake it easy to find the cancel option. Hiding it only makes people angrier when they do.
- Problem 1Pressure that ignores why someone is leaving rarely changes their mind.
- RepairName the one or two losses that matter, and link to the rest.
- What worksThe best retention offer solves the specific reason the person is leaving.
- Problem 2Make exit surveys optional, short and clearly useful, or the answers will be noise.
- Problem 3A simple, friendly goodbye works better than trying to make people feel sorry.
- Problem 4If people cannot answer honestly, the data you collect is not worth much.
- What worksAsk the one question that explains the churn, and put it first.
- Problem 5Suggest new uses when usage starts to drop, not during cancellation.
- RepairThe last screen shapes how people remember the whole product. End it well.
- The resultOffboarding is part of the product. Every step either protects your reputation or spends it.
Source: Growth.Design: Typeform: How to offboard users the right way
Make leaving easy, and make coming back likely.
Cancellation flows are often built last and reviewed never. Zyxware can audit your offboarding and downgrade journeys, cut the steps that frustrate people, and design exits that collect honest reasons and keep your reputation intact.