[Retention & lifecycle]() / [Use cases]() / [Cancellation saves]() [How it works]() [What you get]() [Scope]() [Start a review]() [Retention & lifecycle]()[Subscribers]() # Learn why subscribers cancel, and which saves last. An honest read on why subscribers cancel, and tests that show which saves still leave a paying subscriber 90 days later. [Start with a retention review]() [How it works]() Cancellation reasons Last quarter Reason Share Save to test Still have plenty 41% Skip or delay Away or moving 18% Pause Price 15% An offer Not using it 14% Swap the product Something else 12% Ask Why Care Kit subscribers cancel, grouped by reason, with the save worth testing for each. ## Sound familiar? - The cancel survey says “too expensive” for most of them, and you don’t believe it. - A bigger discount gets accepted more often, and nobody checks who’s still paying. - The reasons sit in surveys, support tickets and reviews that never meet. A save that’s accepted and canceled a month later cost you the discount and kept nothing. ## What changes. ShowTodayWith the engine Reasons A dropdown, taken at its word Survey, tickets and orders read together Offers The biggest discount for everyone Tested, reason by reason Judging a save The offer was accepted Still paying at 90 days Survey changes A new option quietly reshuffles the data Reason codes versioned, so results stay comparable ## How it works. 6 steps, 4 stages. - AI**2** - Rules and automation**2** - Your team**2** - Approval 1. 1 Collect 1. Rules and automation ### Capture the reason and the context The cancel survey, support contacts and recent orders, joined on one subscriber. - Recharge - Gorgias 2. 2 Group 1. AI ### Group the reasons Free-text answers become reasons your team can name, with examples kept beside each. 2. Your team Approval ### Agree the reason codes Your team approves the categories, so the data stays comparable when the survey changes. 3. 3 Test 1. Rules and automation ### Randomize the offer Each reason gets a randomly assigned save: a skip, a swap, a discount or nothing. - Recharge - Stay 4. 4 Decide 1. AI ### Read what lasted Counts the subscribers still paying at 90 days, and what each save really cost. 2. Your team Approval ### Choose what stays in the flow Keep, change or drop each save, with a date to check again. Each read sets the next test ## What you get. Real work, every week. Save tests · Care Kit Extra payers per 100 offers - Offer a skipTo “still have plenty” **+11**Keep - 10% offTo price reasons **+4**Keep - 20% offTo price reasons **+5**Drop Paying at 90 days, against a no-offer group Judged at 90 days. A bigger discount earns its place only if more subscribers are still paying. Test plan Care Kit cancel flow Which offer keeps a subscriber paying? Groups A skip, 10% off, 20% off, no offer Measured Still paying at 90 days, after discounts Reasons Coded the same way throughout Decision Keep an offer only if it pays for itself Every save is a test with a no-offer group, agreed before the flow changes. ## How you’ll know it’s working. 1. Subscribers still paying at 90 days The number a save should be judged on, which acceptance alone can’t show. 2. Cost per saved subscriber Discounts and gifts, divided by the subscribers still paying who wouldn’t otherwise have been. 3. Cancels with a usable reason The reasons only help if most cancels leave one your team can act on. Baselines agreed before we build. ## Works with what you have. Commerce - Recharge - Stay - Shopify Email & SMS - Klaviyo Support - Gorgias We configure first Recharge and Stay both have cancel flows with reasons and offers. We use them to ask the questions, then fix how the results are read. We build when Reasons sit in several systems, or you want randomized offers read at 90 days against a no-offer group. ## What a first build includes. Built in your accounts, around your team. Scope, fee and baseline agreed before we start. 1. 1 Design 2. 2 Build 3. 3 Run it, or hand it over [Start with a review]() [Scope this build]() ### We build - One set of reason codes across survey, tickets and orders - A cancel flow with randomized offers - A no-offer group in every test - A 90-day readout of what each save kept ### We start from - Cancellation and survey data - Support contacts from subscribers - Subscription and order history ### From your team Owner Head of retention or lifecycle Involves - Lifecycle - Customer service - Finance ## Questions teams ask. Isn’t this the same as subscriber control? They’re neighbors. Subscriber control makes skipping, delaying and swapping easy. This job learns why people cancel anyway, and which saves keep them paying. We measure the two separately, because a skip that only delays a cancel isn’t a save. Why not just offer a bigger discount? ProsperStack, a cancel-flow vendor, reported randomized tests in 2026 where a 20% offer roughly doubled acceptance against a 10% one. It added only about one more paying subscriber per 100 offers by day 90. The vendor didn’t disclose sample sizes, so treat it as a signal. Acceptance is easy to count, and still paying at 90 days is what pays. Can we trust the reasons people pick? Treat them as a starting point. Wording changes what people choose, so we keep the raw answers, version the reason codes and check them against what the same subscribers did next. Who owns it once it’s running? You do. It’s built in your accounts, documented, with training for the people who run it. Ownership usually sits with your head of retention or lifecycle. We can keep running it with you, or step back. How do we start? Usually with a retention review: two to four weeks, a fixed fee credited to the build, and a plan that’s yours. If the job is already clear, we can scope the build directly. ## Part of your retention engine. One of 12 use cases, in 6 parts. [See the whole retention engine]() 1. [Second orders **3**]() 2. [Subscribers **3**]() 3. [Contact that fits **1**]() 4. [Owners & loyalty **2**]() 5. [Lifecycle production **1**]() 6. [Proof & decisions **2**]() In subscribers - [**Subscriber control**Let subscribers skip, delay and swap without canceling]() - [**Payment recovery**Recover failed subscription payments]() - **Cancellation saves**You’re here Works with - [**Retention proof**Prove which messages and discounts earn their cost]() See it at work - [**Also in**Customer service & support]() ## Keep the subscribers worth keeping. Start with a retention review. [Start with a retention review]() [**Talk it through first**15 minutes]() We reply within one business day. --- Source: [https://www.thekeating.co/use-cases/cancellation-saves/]() [Contact Keating & Co](mailto:nelson@thekeating.co) · [Agent guidance](https://www.thekeating.co/llms.txt)