For the complete documentation index, see llms.txt. This page is also available as Markdown.

How to Reduce Churn and Cancellations

Find out why customers cancel, save some of them on the spot, and fix the reasons the rest are leaving

A cancellation is the most informative moment in the customer lifecycle and the one most companies waste. The customer has already decided to leave, which means they have nothing to lose by telling you the truth — and for a few minutes, they're still reachable.

A good cancellation survey does two jobs at once: it saves a portion of the customers who take it, and it tells you what to fix so the next cohort doesn't reach this page at all.


Building the survey

Slide 1 — The reason

Add a Single Choice slide: "What's the main reason you're cancelling?"

Write options that map to different fixes, not to different feelings:

  • It's too expensive

  • I'm not using it enough

  • I'm missing a feature I need

  • I had a bad experience with the product

  • I had a bad experience with support

  • I'm switching to another product

  • My needs changed

Enable a dynamic "Other" option. Keep the list to seven or fewer — a long list on a cancellation page gets skipped.

Slide 2 — The detail

Add a Long Answer slide: "What could we have done differently?"

Make it optional. A required text box at the moment someone is trying to leave feels like a hostage negotiation, and you'll get "asdf" instead of an answer.

Slide 3 — The save offer

This is where Slide Logic earns its keep. Branch on the reason from Slide 1:

They said
Do this

Too expensive

Show a Promo Code slide, or use the Generate Reward action on Shopify to issue a discount on the spot

Not using it enough

Use a Redirect To action pointing at onboarding, a tutorial, or a booking link

Missing a feature

Ask which feature, then offer to notify them when it ships — capture the email and push it to Klaviyo

Bad support experience

Route to a human immediately: create a Zendesk ticket or notify Slack with a Slide Logic action

Switching to a competitor

Ask which one with a Short Answer slide. Competitive intelligence you cannot buy.

Don't offer a discount to everyone. A customer leaving because a feature is missing doesn't want 20% off — and offering it trains customers to threaten cancellation for a discount. Match the offer to the reason.


Where to run it

  • On the cancellation flow itself — Trigger the survey when a customer clicks cancel with a Custom Survey Trigger. This is the only placement that lets you save the customer.

  • Email after cancellation — For customers already gone. You lose the save opportunity but you still get the diagnosis, and the answers are often blunter. See Email Zigpoll Surveys.

  • Before they churn — The best cancellation survey is the one you never have to send. Target at-risk customers (declining usage, no orders in 60 days) with a check-in survey and catch the problem while it's still fixable.


Reading the results

Track the reason mix over time, not the raw counts. Total cancellations move with the size of your customer base, but the mix tells you whether you're fixing the right things.

  • "Too expensive" dominates → It's rarely price. It's usually value: the customer isn't seeing enough benefit to justify what they already agreed to pay. Look at whether the cancellers activated properly.

  • "Not using it enough" dominates → An onboarding and activation problem. The product never became a habit.

  • "Missing a feature" dominates → Read the specifics. If the same feature appears repeatedly, you've found a roadmap item — see How to Prioritize Your Product Roadmap.

  • "Bad experience" dominates → Fix this first regardless of volume. These customers tell other people.

Use Z-GPT Insights to summarize the open-ended responses into themes automatically, and sync everything to Google Sheets so you can chart the mix month over month.


Measuring the save rate

Track how many people who start the cancellation survey don't complete the cancellation. That's your save rate, and it turns the survey from a research cost into a revenue line.

If the save rate is near zero, your offers are mismatched to the reasons. If it's very high, you may be discounting customers who would have stayed anyway — tighten the conditions in your Slide Logic.


How to Measure Product-Market FitHow to Prioritize Your Product RoadmapCustomer Effort Score (CES)

Last updated