Designing a Subscription People Don't Want to Cancel

Published 2018-11-12 · Updated 2026-06-23 · By Alan Wizemann

Topics: Product Strategy & Development, Subscription, User Experience, Digital Marketing

There are really only two ways to keep a subscriber, and most companies, when the pressure is on and a churn number is staring back at them, instinctively pick the wrong one. The first way is to make leaving hard. Bury the cancel button somewhere it takes three screens to find, make them call during business hours (and only during business hours), hide the skip option, lock the contents of the box so they cannot meddle. This works, in the narrow and short-lived sense that it slows down cancellations, and it is a catastrophe in every sense that actually matters over time, because it quietly converts a relationship into a hostage situation. The member who finally fights their way out does not just leave. They tell people, they never come back, and every month they stayed angry they were a churn waiting for a trigger to set it off. The second way is to make staying easy – so easy, and so clearly in the member's own control, that there is no accumulated resentment left to cancel over in the first place.

That second way is the design philosophy we built the membership experience around at Dollar Shave Club, and it rests on a bet that sounds backwards until you sit with it: give people more control over the thing, including the explicit power to take less of it, and they stay longer. The instinct in the room is always that if you let people skip a box, change their frequency, edit what is inside, and walk away whenever they feel like it, then of course more of them will. In practice the opposite happens, and once you understand why, it stops feeling like a paradox at all. The thing that makes people cancel a subscription usually is not the product itself. It is the feeling of being trapped by it: a box arriving they did not need, at a time they did not choose, full of things they did not pick, with no obvious way to fix any of it except to quit the whole arrangement entirely. When quitting is the only control you have given them, quitting is the control they will eventually reach for.

So we gave them the smaller controls instead, the ones that let a member fix the specific problem in front of them without burning the whole relationship to the ground. Tell us how often you actually want a shipment, rather than how often our default would like to send you one. See what is coming in your next box before it ships, and change it if it is wrong. Add the things you are about to run out of in a single tap, skip a month when the cabinet is still full, and build a profile so the box reflects who you are over time instead of staying frozen at whatever you happened to pick on day one. Every one of those is a release valve, and each one lets a member relieve the small pressure that would otherwise build, quietly and unnoticed, into a cancellation. There is one feature in particular that captures the whole philosophy for me: a savings gauge that simply shows members, plainly and without theatrics, what they are getting out of the membership. No trick, no dark pattern, just honesty about the value, because here is the bet sitting underneath all of it – if the membership is genuinely worth it, then the single most powerful retention tool you have is making that value obvious and putting the member visibly in charge. You only need to trap people when you are quietly afraid the value will not hold them on its own. The cage, when you really look at it, is a confession.

The deeper point is about what a subscription actually is, underneath the billing mechanics. It is not a clever way of extracting recurring revenue on autopilot. It is a standing promise that you will keep being worth it, renewed silently every single cycle whether you acknowledge it or not. The member re-decides to stay every time a box ships, and the only question is whether you make that re-decision a comfortable one or a contested one. Companies that treat the subscription as a trap are, in effect, trying to suppress that re-decision and hope nobody notices they are doing it. Companies that treat it as a relationship are trying to win it, every cycle, by being genuinely easy to live with. I will name the obvious tension, because someone in the room always does and they are right to: does making it this easy to skip and cancel cost you revenue in the short term? Yes, in the immediate and literal sense, some boxes that would otherwise have shipped to an annoyed member simply will not ship. But that member was going to leave anyway, and probably leave angry, and what you get in exchange is a member who sticks around for years because the thing never once made them feel cornered, and who speaks well of you instead of warning their friends off you. Run the lifetime math honestly and it is not close. The skipped box is a rounding error set against the retained relationship.

The test I would hold any subscription design up to is a simple one, and it is uncomfortable in a useful way. Imagine the member trying to do the reasonable thing – take a little less, change the timing, fix what is in the box – and then ask, honestly, whether your product helps them or fights them in that moment. If it fights them, you have designed a trap, and traps leak no matter how cleverly you build them. If it helps them, you have designed a relationship, and relationships compound quietly in your favor for years. So make staying easy and stop trying to engineer leaving out of existence. The people who would have cancelled anyway will cancel a little sooner, which costs you almost nothing, and the people who would have stayed resentfully will stay happily instead, which is worth nearly everything. That is a trade I would make every single time, without much agonizing over the quarter it shows up in.