Telehealth Retention: Why Month Three Is Where You Lose People

Most telehealth companies can tell you their cost per acquisition to the cent. Far fewer can tell you what share of patients are still there at month six.

That second number decides whether the business works. If your payback period is four months and your median patient leaves at three, no amount of funnel optimisation fixes it.

Where people actually drop

The curve has a specific shape in subscription care, and it isn't smooth.

The first refill. The biggest single cliff. The patient got their initial supply, and now something has to happen again — a check-in, a renewal, a payment. Every one of those is a chance for the process to fail or for them to reconsider.

Around month three. The point where the initial motivation has faded and results either have or haven't materialised. Also, frequently, when the introductory pricing ends.

Any friction event. A failed payment. A refill that arrives late. A message that went unanswered for three days. These don't cause immediate cancellation; they cause quiet disengagement that shows up as a cancellation two weeks later.

Notice that two of the three are operational, not clinical. Patients often leave because the machinery around the care was annoying, not because the care didn't work.

Failed payments are silent churn

A meaningful share of subscription cancellations aren't cancellations at all. The card expired, the charge failed, and the system treated it as an ending.

This is the cheapest retention win available, and it's covered in more detail in our subscription billing guide. The short version: retry on a sensible schedule, use your processor's card updater, notify through more than one channel, and give a grace period that doesn't interrupt care.

Fixing involuntary churn typically recovers more patients than any campaign you could run.

Refill timing is a retention feature

If a patient runs out before the next shipment arrives, two bad things happen at once: their treatment is interrupted, and they experience your service as unreliable.

The fix is unglamorous:

• Trigger the refill process on days-of-supply remaining, not on a calendar date. A 90-day supply dispensed three days late shouldn't push everything downstream.
• Build in buffer. Start the renewal conversation with two weeks of supply left, not two days.
• Make the required check-in easy. If the refill needs a provider touchpoint, that touchpoint should take the patient two minutes, not require scheduling.
• Tell them where it is. Shipping visibility removes an entire category of support ticket and anxiety.

Response time is the trust signal

In a service where the patient never meets anyone in person, how fast you answer a message is the relationship.

Set a response commitment you can actually keep and then keep it. Twenty-four hours, honoured consistently, beats "usually a few hours" that sometimes takes three days. Patients calibrate to what you promise; they churn on what you break.

Route clinical questions to clinical staff and everything else to support, so a billing question doesn't sit in a provider queue for two days.

Continuity of provider matters more than speed

Given the choice between seeing anyone tomorrow and their usual provider on Thursday, patients in ongoing care generally take Thursday.

Continuity improves adherence and it improves retention, and most scheduling systems optimise against it by defaulting to first-available. Weight the patient's previous provider heavily and offer the choice explicitly rather than assuming speed wins.

What to measure

Cancellation rate alone tells you very little. You need:

Cohort retention curves. Group patients by signup month and track what share remain at 1, 3, 6, 12 months. This is the only view that shows whether your product is actually improving, because a headline churn number moves with growth rate.

Voluntary vs involuntary churn, separated. If a third of your churn is failed payments, that's a billing project, not a product project. Most teams don't split these and end up solving the wrong problem.

Refill adherence. What share of patients complete their next refill on time? This leads cancellation by weeks and gives you a chance to intervene.

Time to first value. How long from signup to the thing the patient actually came for. Shortening this improves everything downstream.

Exit reasons, collected properly. Not a required dropdown on the cancellation page — an optional free-text field, read by a human weekly. The dropdown gives you clean data about nothing; the free text tells you what's actually wrong.

The intervention that works and the one that doesn't

Works: reaching out before the drop. A patient who's missed a check-in or is approaching a refill without engaging is showing you they're going. A short, human, non-salesy message at that moment recovers a real share of them.

Doesn't work: the retention offer at the cancellation screen. By then they've decided, and a discount mostly delays the same outcome by a month while training people that cancelling is how you get a better price.

Make cancellation easy anyway

It feels counterintuitive, and it's still right.

Hard cancellation flows generate chargebacks, complaints, and public reviews — and chargebacks cost you more than the churn would have. US regulators have also been increasingly active on subscription cancellation being as straightforward as sign-up, and health services attract more scrutiny, not less.

Offer a pause as a genuine alternative. A patient who pauses for two months and returns is worth far more than one you made angry enough to dispute the charge.

The honest framing

Retention in telehealth is mostly operations wearing a product costume. The clinical care is usually fine. What loses patients is a late shipment, a failed card nobody told them about, a message that went unanswered, and a renewal that required more effort than it was worth.

Fix those four and you'll move the curve more than any acquisition work would.

---

We build the billing, refill, and messaging systems that keep patients from falling through the gaps. See the telehealth build.