Almost every article about concierge medicine — including several of ours — treats the membership fee as a number. You pick it, patients agree to it, and it arrives. The operational reality is less tidy. The fee has to be collected, on a schedule, from a payment method that expires, from people whose circumstances change, in a way that does not require your front desk to make awkward phone calls.

Practices that get this wrong rarely notice immediately. The revenue loss shows up as a slow, unglamorous drip: a card that expired in March and was never updated, a renewal that nobody followed up on, a patient who meant to rejoin and drifted. None of it is dramatic. All of it compounds.

The two kinds of churn, and why only one gets attention

Voluntary churn is a patient deciding not to renew. It is visible, it stings, and practices think about it constantly. It is also, in a well-run concierge practice, usually modest — people who value the access tend to keep valuing it.

Involuntary churn is a membership that lapses for reasons nobody chose: an expired card, a replaced card after a fraud alert, a declined charge on a full credit line, a bank that flagged a large annual charge as unusual. The patient still wants the membership. They simply are not paying for it, and frequently do not know.

The distinction that matters operationally

Voluntary churn is a value problem — the patient decided the membership was not worth it, and the fix is in the care, the access, or the price. Involuntary churn is a plumbing problem, and the fix is entirely in your process. Practices that report a "retention problem" are often looking at a plumbing problem and trying to solve it with better medicine.

Card-based subscription businesses have studied this for two decades, and the consistent finding is that a meaningful share of cancellations in any recurring-payment model are involuntary rather than chosen. A concierge panel of 400 members is a recurring-payment business, whether or not it thinks of itself that way. The difference is that a consumer subscription loses $12 a month to a failed card. You lose several thousand dollars a year, from a patient who is still walking into your office.

Annual, monthly, or quarterly: the cadence decision

How often you charge changes your cash flow, your conversion rate, and your exposure to payment failure. There is no universally correct answer, but the trade-offs are consistent.

Annual

Best cash flow. One charge, one collection event, a full year of revenue up front. Lowest total processing cost.

Hardest sell. A four-figure single charge is a real decision for many patients, and it is the version most likely to be declined by a card issuer or to bump a credit limit.

Monthly

Easiest sell. A monthly figure reads as affordable in a way the same amount annually does not, and it removes the single largest objection at signup.

Twelve times the failure surface. Twelve charges a year means twelve chances for a card to decline, and twelve small collection events for someone to manage.

Quarterly

The compromise. Four charges a year keeps the individual amount digestible while limiting failure events and administrative load.

Less common. Patients are conditioned to monthly and annual pricing; quarterly sometimes needs explaining, which is friction at exactly the wrong moment.

Two practical notes. First, offering annual with a modest discount alongside monthly captures both preferences and pulls some cash forward — but price the discount deliberately rather than by instinct, because it comes straight off your margin. Our guide to setting membership fees covers the underlying pricing work.

Second, whatever cadence you choose changes your first-year model. A practice that assumed annual billing and then sells mostly monthly memberships will collect substantially less in its opening quarter than the projection said. That is a timing gap, not a revenue gap, but it is real and it has to be funded. The costs that get left out of first-year projections covers the rest of that category.

What actually happens when a card fails

A declined charge is not a single event. It is the start of a sequence, and the quality of that sequence determines whether you keep the member.

The common failure reasons, roughly in order of frequency:

  • Expired card. The most common and the most preventable. Cards expire on a schedule you can see in advance.
  • Card replaced. Fraud, loss, or a bank reissue. The old number simply stops working, usually without warning to you.
  • Insufficient funds or credit limit. Frequently temporary. A retry days later often succeeds where an immediate retry does not.
  • Issuer decline on an unusual charge. A large annual membership charge can look like fraud to a risk model, particularly the first time.

The important operational insight is that these have different remedies. An expired card needs the patient to act. An insufficient-funds decline usually needs nothing but patience and a retry. Treating every failure as the same event — an immediate email saying the payment failed — is both more alarming and less effective than it needs to be.

The silent lapse

The worst version of this is a failed charge that generates no follow-up at all. The patient keeps coming in, keeps texting, keeps receiving the full membership experience, and has not paid in four months. When somebody finally notices, the practice has an uncomfortable choice between a large back-bill and writing it off. Both are bad, and both are the result of nobody owning the failure queue.

A dunning sequence that does not feel like collections

"Dunning" is the industry term for the retry-and-remind process after a payment fails. Most consumer versions are aggressive and impersonal, which is exactly the wrong register for a practice whose entire value proposition is a personal relationship.

A sequence that works in a clinical setting looks roughly like this:

  • Before it fails: the expiry sweep. Thirty days ahead, identify cards expiring before the next charge and ask for an update. A single message that says "your card on file expires next month" prevents a large share of failures and never mentions money owed.
  • Day 0: retry quietly. Do not message anyone on the first decline. Many resolve on their own.
  • Day 3: retry, then notify. If it fails twice, tell the patient — warmly, briefly, with a direct way to update the card. Frame it as a card problem, which it almost always is, not as a debt.
  • Day 7 and day 14: retry, remind. Two further attempts, spaced. Retry timing matters more than message volume; funds and limits change with pay cycles.
  • Day 21: a human. Someone from the practice calls. Not to demand payment — to ask whether anything has changed. Occasionally the answer is that the patient lost a job and is embarrassed, which is a conversation worth having rather than a lapse worth processing.

Note what is not in that list: suspending access on day one. A membership that quietly stops delivering the day a card declines converts a plumbing problem into a relationship problem.

The renewal cliff

Annual memberships create a moment that monthly ones do not: an anniversary, where a patient who has not thought about the fee in eleven months is asked to think about it. If the renewal is a silent auto-charge, some patients feel ambushed. If it is a re-decision, some will decide no.

What tends to work:

  • Announce the renewal, do not spring it. Thirty days of notice before the charge, stating the amount and date. Nobody who values the membership cancels because you told them it was coming. Some who were surprised by the charge cancel because you did not.
  • Attach a year in review. A short note on what they used — visits, messages, coordination, after-hours contacts — reframes the fee against what it bought. This is the single cheapest retention intervention available to a concierge practice, and it works because the value is genuinely there and simply invisible.
  • Handle fee increases separately and early. If the fee is going up, that message should not arrive in the same breath as the renewal charge, and it should arrive with more than 30 days' notice. Renewing at a new price is a new decision, and it deserves the same care the original conversation got. Our piece on discussing membership fees applies here as much as it does at signup.

Who owns this

The most common structural mistake is that nobody does. Billing is "the front desk's job" in the same way that the supply closet is everyone's job, which is to say it is attended to when there is time.

Membership billing wants a named owner with a recurring slot in the week. The task list is small and boring, which is precisely why it gets skipped:

The weekly membership billing pass

  • Review the failed-payment queue. Every failure has a next action and a date.
  • Run the expiry sweep for cards expiring within 30 days.
  • Confirm renewals due in the next 30 days have had their notice sent.
  • Reconcile active members against members actually paying. These two numbers drift, and the gap is your involuntary churn.
  • Flag any member more than 30 days unpaid for a personal call, not another email.
  • Log cancellations with a reason, so voluntary and involuntary churn stay separable.

That last item deserves emphasis. If cancellations are recorded without a reason, every churn number you look at afterward is uninterpretable — you cannot tell whether your price is wrong or your card-update process is. Ten minutes of categorization a week makes the difference between a metric and a guess.

The number to watch

Practices track member count. Member count is a lagging indicator and it flatters you, because it includes the people whose payments quietly stopped.

Track collected members instead: how many memberships actually produced revenue in the period. The gap between that figure and your roster is the money that is leaking, and it is almost always larger than expected the first time anyone measures it.

Pair it with payment success rate — the share of attempted charges that clear on the first try. In a healthy practice with a working expiry sweep, that number is high and stable. A drop is an early signal, usually weeks before it shows up as missing revenue, and it points at process rather than at your patients.

Neither number requires sophisticated tooling. Both require somebody to look. If you are still building the operational side of a transition, the sequencing for all of this sits in the twelve-month transition timeline — billing belongs in the middle quarter, tested before go-live rather than discovered after it.