Most brands build a cancel flow once, wire up whatever save offer feels reasonable, and never look at it again. The offer that ends up in the most important spot is usually there by accident — someone picked it in a meeting two years ago and moved on.

We just went deep on one of our supplement clients' cancel flows, tracked what actually happened to the subscribers who went through it, and found that the single most valuable screen in the whole funnel was defaulting to the weakest possible save. Fixing that one thing is, dollar for dollar, one of the highest-leverage changes we've made all year.

Here's what we learned.

When a subscriber clicks cancel, the first thing they hit is a splash screen — a well-produced video, then a single call to action: skip your next order. If they decline the skip, they move into a second layer of offers tailored to their stated cancellation reason:

It's a thoughtful flow. But when we pulled the data on who went through it and looked at which accepted offer led to the longest LTV afterward, the picture got very clear, very fast.

The first finding is the one that reframes everything: the splash screen is where all the juice is. That first screen — before anyone reaches the reason-based menu — is where the actual saves happen. It's the highest-traffic, highest-intent moment in the entire flow, and whatever offer sits there does the overwhelming majority of the work.

So the question isn't "do we have good offers in the flow?" You do. The question is: is the right offer on the first screen? Because that's the one that matters most, by a wide margin.

Right now, it isn't.

The splash screen currently offers a skip. And skips look like saves — the subscriber doesn't cancel today, so the flow records a win. But when we followed those people forward, only about 20% of the folks who skipped ever made it to their next shipment. Roughly four out of five skips churned anyway. The skip didn't save them; it just moved the cancellation out a cycle and let the flow take credit in the meantime.

That's the trap with skip and pause offers in general: they feel accommodating, but they interrupt the habit. A subscription only survives because the product keeps showing up and keeps getting used until it's automatic. The moment you let someone stop receiving it, the routine goes dormant, the reason they subscribed fades, and by the time the skipped order comes back around, canceling is the path of least resistance.

Compare that to the discount. When subscribers were offered a discount instead, about 75% made it to their next order. Same decision point, same "I'm about to cancel" intent — a completely different outcome.

The mechanism is the mirror image of the skip. A discount keeps the product arriving and the habit intact; the only thing that changes is the price for a cycle or two. That temporary margin hit buys you the one thing that actually retains a subscriber: continuity. The routine reasserts itself, and the customer comes back to full price worth far more than the discount ever cost.

One honest caveat, because it matters: the discount number is drawn from a small cohort — right now the discount is only being surfaced to a limited slice of users, so the sample behind that 75% is smaller than the skip sample. It's an early signal, not yet a fully scaled result. But the size of the gap — 75% versus 20% — is large enough, and the mechanism behind it clear enough, that it's well worth acting on and validating at volume.

The action is simple and immediate: change the splash-screen offer from skip to discount.

That's it. Keep the rest of the flow — the reason-based menu, the swap, the reschedule, even the skip for the people who genuinely have too much product. But the first, highest-leverage screen should lead with the save that actually retains, not the one that quietly defers the loss. You're not adding a new offer or rebuilding the funnel. You're moving your best-performing save into your most important slot.

The reason this was such a big unlock isn't really the discount. It's what the exercise exposed: this flow was optimized for the wrong moment. It counted the cancel that got avoided today instead of the subscriber who was still paying next month — so it happily treated skips as saves when four out of five of them were just deferred cancellations.

Measure the cancel flow downstream. Look at who's still on the subscription a cycle or two later, not who clicked away from the cancel button. When you do, the best offer usually stops being a matter of opinion — and, at least here, it clearly wasn't the one sitting on the first screen.