The cheapest patient to book is the one you already treated

Clinics spend many times more attracting a new patient than bringing back an existing one, even though the returning patient already knows you and needs no convincing. The gap is not budget. It is the absence of a system that remembers who should come back and when.

6 min read

In most clinics the patient database is the largest unused asset in the building. Thousands of people visited once or twice and then disappeared — not because anything went wrong, but because nobody reminded them. Meanwhile the entire marketing budget goes toward finding new faces.

Recall is the system that decides who should return, when, and with what message, and then executes without depending on a staff member's memory. Building it takes about six weeks, and the effect shows up in the appointment book long before it shows up in any marketing report.

Why retention beats acquisition in clinic economics

A returning patient differs from a new one in four ways that touch profitability directly, and every one of them works in your favour:

  • Acquisition cost close to zero: one message instead of an advertising campaign.
  • Higher attendance: someone who has been before, knows the route, the parking, and the clinician is far less likely to drop out at the last minute.
  • Shorter consultations: the file and the history already exist, so nothing starts from scratch.
  • Higher treatment acceptance: trust is already established, and that is the hardest variable in any first visit.

The consequence is that one percentage point of improvement in return rate is usually worth more than one point of additional advertising, at a fraction of the cost. That does not remove the need to be visible in local search — it decides which one you build first.

Four recall cohorts worth automating

Do not start with a list called 'all patients'. Effective recall is built on a clear clinical reason for each group to return, and that reason is what turns a message from an intrusion into a service.

CohortWho it containsWhen to reach out
Scheduled reviewPatients whose condition needs periodic review or a seasonal checkTwo weeks before the review falls due
Unfinished treatment planPatients who started a plan and stopped before completing itFour to six weeks after the last session
Repeated non-attendancePatients who cancelled or missed a visit and never rebookedWithin a week of the missed appointment
LapsedPatients with no visit in twelve months or moreA seasonal campaign, twice a year

The first two cohorts return the most, because the reason to come back is clinical and specific. The lapsed group is the largest and the least responsive, so do not begin there however tempting its size.

Writing a recall message patients do not resent

What separates a useful message from an irritating one is content, not design. A good recall reminds the patient of something that belongs to them specifically, and asks for exactly one action.

  • Name the reason: 'It has been a year since your last check' is clearer and more honest than 'we miss you at our clinic'.
  • Offer a single action: one booking link or one reply. Multiple options postpone the decision.
  • Use the patient's own language: Arabic for those who registered in Arabic, rather than one language for everybody.
  • Respect timing: sensible working hours only, and nothing during holidays or public occasions.
  • Make opting out effortless: one word should be enough, and its presence raises trust and lowers complaints.
  • Never send more than twice: a message and one follow-up. A third reads as pursuit and loses the patient permanently.

Consent is a requirement, not a courtesy

Marketing contact with patients falls under data protection rules in Saudi Arabia and the UAE, which require a documented lawful basis and a clear way to stop. Capture consent for the contact channel at registration and keep the record — recall without consent is a regulatory exposure before it is an annoyance.

A six-week build for your recall cycle

The plan is designed to end with a system that runs itself, not a campaign that gets executed once and forgotten during the first busy week.

Week one: clean the database

Merge duplicates, normalise mobile number formats, and remove empty records. A message sent to a wrong number counts as a failure in your measurement later.

Week two: define the cohorts and their rules

For each group write the entry condition, the send timing, and the number of attempts. A written rule is what can be automated and audited.

Week three: write and approve the messages

An Arabic and an English version per cohort, reviewed by the responsible clinician so no message promises something the clinic does not provide.

Week four: launch one cohort only

Start with scheduled reviews on a limited segment. A full launch hides mistakes and makes them harder to correct.

Week five: read the results and adjust

Look at the booking rate, not the open rate, and change the timing or the wording based on who actually booked.

Week six: expand and automate

Add the second cohort and let the rules run on their own, so no recurring manual work is left in the process.

The numbers that prove recall is working

Most out-of-the-box reports show figures that look encouraging and mean nothing. Three metrics are worth reviewing monthly, and all three connect directly to revenue:

MetricHow to calculate itWhy it matters
Return ratePatients who booked after a recall ÷ patients contactedThe only measure that reflects a real effect on the schedule
Annual retention ratePatients seen this year who were also seen last year ÷ last year's patientsExposes the slow leak that visit-count reports never show
Lifetime patient valueAverage revenue per patient across all of their visitsTurns retention from a marketing idea into a line in the budget

Add these three to the operational metrics you already follow instead of tracking them in a separate file, and review them monthly rather than weekly — retention is a slow indicator by nature, and reading it too often produces noise rather than information.

Recall starts with a consent record you can prove

A marketing consent record on every patient, per channel — email, SMS, and WhatsApp, with grant and revocation dates and the privacy-notice version — plus automatic no-show follow-up and package-expiry notices, inside one plan.

See the pricing

Frequently asked questions

How does a recall differ from an appointment reminder?
A reminder concerns a visit that is already booked and exists to secure attendance. A recall concerns a patient with no booking at all and exists to create one. The first is a short operational task; the second is an ongoing system built on a clinical reason for each group to return, which is why it is measured by booking rate rather than by attendance.
How often should we contact a patient once treatment ends?
Tie the timing to the clinical situation rather than a marketing calendar: a stalled treatment plan warrants contact within four to six weeks, and a periodic check warrants contact two weeks before it falls due. A patient who has finished treatment and needs no follow-up needs no more than seasonal contact, twice a year at most.
Which patients are worth recalling first when time is limited?
Those who began a treatment plan and never completed it, because the reason to return already exists and both sides know it, so the decision is half made. Periodic-review patients come next. People who lapsed years ago are the largest group and the least responsive, and should never be the starting point however tempting the list size looks.
Do recall messages breach patient consent rules?
Not when they rest on documented consent for the contact channel, with a one-step way to stop and immediate respect for that request. Problems arise when messages go to imported lists or without any record proving consent, which is precisely what the data protection regimes in Saudi Arabia and the UAE treat strictly.
What return rate should a clinic expect from recall?
It varies widely by cohort, and the only useful comparison is against your own result from the previous month rather than a published figure. Build a baseline from the first cycle, then improve it by adjusting timing and wording. The unfinished-treatment group typically outperforms every other cohort by a visible margin.

Related articles