Walk a competitor's cancellation flow, step by step

Where does a competitor spend its most defensive design effort? Not the homepage. The homepage is written for strangers. The cancellation flow is written for people who already paid, already know the product, and are halfway out the door — and every screen in it is a bet about why they are leaving.

That makes it one of the most honest surfaces a SaaS company ships. It is also one almost nobody on a competitive team ever looks at, because you have to be a customer to see it.

This playbook is for walking it properly, once or twice a year, on your top two or three competitors. It takes a paid month on the cheapest plan and about an afternoon.

Step 1 — Get a real account, honestly

You need to be past the trial. Many products show a different, lighter exit path to trial users than to paying customers, and the paying one is where the retention budget goes.

• Buy the cheapest self-serve paid plan, monthly. Budget it as research spend.
• Sign up under your own name and work email. Do not invent a persona or a fake company — you want the flow a normal customer sees, and misrepresenting yourself is not worth the ethical or legal exposure.
• Use the product lightly for a week or two: create a project, invite nobody, connect one integration. A dormant account and an active one often get different offers.
• Screenshot the billing page before you touch anything. Note the plan name, price, and renewal date exactly as shown.

If you already ran the 30-minute onboarding audit, reuse that account — you are simply finishing the lifecycle.

Step 2 — Find the exit before you use it

How hard is the cancel button to find? That is your first data point.

• Start from the dashboard and count clicks to the first screen that mentions cancelling.
• Note where it lives: billing settings, account settings, a help-center article, or "contact support." Each is a deliberate choice.
• Search their help center for "cancel", "downgrade", and "refund". Save the articles — the wording of those pages is public and changes over time.
• Check whether downgrade and cancel are separate paths. Many products push you toward downgrade first; which plan they steer you to tells you where their margin floor is.

A cancel path buried behind a support ticket usually means monthly churn is a board-level worry. A one-click cancel usually means they are confident in reactivation — or selling to a segment that punishes friction.

Step 3 — Read the exit survey like a product brief

The reasons list on the cancellation survey is the competitor telling you, in their own words, the top five or six reasons customers leave them.

• Record every option verbatim, in order. Order matters: the first option is often the one they most want to hear about, or the one they have a canned answer for.
• Look for competitor names in the list. "Switching to [Your Company]" as a pre-filled option is about as direct a signal as you will ever get.
• Note what is missing. If "too expensive" is absent from a product that just raised prices, that omission is a choice.
• Pick a reason that is true for you — usually "just evaluating" or "no longer needed" — and continue. Do not fabricate a complaint to fish for a bigger discount.

Step 4 — Log every retention offer

This is the step that pays for the month. Whatever they put between you and the confirm button is their real answer to "what saves a customer."

• Discount offers: size, duration, and whether it is a coupon or a plan change. "50% off for three months" and "switch to annual and save 20%" are very different retention philosophies.
• Pause options: a pause-instead-of-cancel screen tells you seasonal or project-based churn is a known problem for them.
• Feature nudges: "Did you know you can…" screens name the feature they believe is most under-discovered.
• Human intervention: a "talk to us first" prompt or a calendar link means someone on their side is paid to save accounts at your price point.

Screenshot every screen. Write down the exact copy — the phrasing ages faster than the offer itself.

Step 5 — Test the data exit

What can a leaving customer take with them? This is where you learn how much they rely on lock-in.

• Look for an export option before and after cancelling. Note the formats offered: CSV, JSON, API-only, or "contact support."
• Check what happens to data after cancellation — immediate deletion, a grace period, or indefinite read-only access. The help-center article usually states it.
• Note whether an export requires an active subscription. Tying export to payment is a lock-in lever, and it is a sales argument you can use honestly if it is true.

If their export is weak, that becomes a line in your migration story. If it is excellent, your switching pitch cannot lean on "we make it easy to leave them."

Step 6 — Watch what follows, then file it

The flow does not end at the confirm screen.

• Keep the inbox open for 30 days. Log every win-back email, its timing, and its offer — the pattern is covered in more depth in the sales-sequence mystery-shop playbook.
• Write a one-page summary: path length, survey options, offers shown, export terms, win-back cadence. Date it.
• Hand the survey options and retention offers to sales and product. Hand the export terms to whoever writes your migration pages.
• Put a reminder on the calendar to repeat it in six months. The value is in the diff, not the single snapshot.

That last point is the part people skip. A cancellation flow walked once is an anecdote; walked twice, it tells you whether their retention problem is getting worse. The logged-in screens need a human with a card. The public half does not: their help-center articles on cancelling, refunds, and data retention, plus the plan terms on the pricing page, are exactly the kind of surface Seeto watches continuously, flagging each wording change as a discrete event so you know when it is time to walk the flow again. It won't tell you what the retention offer is — only that the policy around it just moved.

Before you start

Keep it clean. Pay for the plan, use your real identity, give honest survey answers, and don't pressure their support team for concessions you would not really accept. The flow you are studying is designed for real customers; the most useful version of this exercise is the one where you behave like one.

The exit is the only surface a competitor writes for people it has already convinced once. Read it like it matters to them — because it does.