Keep your patients longer,and know what one is worth

What you will have at the end.

A two-year patient value per category, a retention curve showing the visit where people leave, and one reactivation campaign running with its results measured.

A clinic that keeps patients an extra visit each does not need to spend more on advertising to grow. That sentence is true, boring, and almost never acted on, because acquisition is visible and retention is not.

What a patient is worth

1. Calculate two-year value, per category

   Two years, not lifetime. Lifetime value in a clinic is a number nobody can verify and everybody inflates. Two years is long enough to matter and short enough that you have the data.

   The calculation is three columns and one multiplication.

   —Average revenue per visit, by service category.

   —Average number of visits in the first twenty-four months, for patients whose first visit was two to three years ago.

   —Multiply. That is your two-year value for the category.

   In practice

   Use a cohort whose two years have actually elapsed. Taking patients who started six months ago and extrapolating produces a number that is wrong in a direction that flatters you.

2. Turn it into the number you actually use

   The two-year value on its own does not decide anything. What decides things is the ceiling it implies for acquisition.

   —Take the two-year value, subtract your direct cost of delivering it, and you have contribution.

   —A common working ceiling is a fifth to a third of contribution as your maximum cost per new patient.

   —Sit at the higher end when you have empty capacity. Sit at the lower end when the schedule is tight.

Finding the drop-off

1. Build the retention curve by visit number

   This is the most useful chart in a clinic and almost nobody has it. For a cohort of new patients, count how many made a second visit, a third, a fourth, and so on. Plot it.

   There is always a cliff. It is never where owners expect, and once you see it, it is obvious what to fix.

   —Take all patients whose first visit was two to three years ago.

   —For each, count total visits since.

   —Count how many reached visit two, visit three, visit four, and so on.

   —Express each as a percentage of the ones before it. That is your curve.

2. Look at the visit before the cliff, not at the cliff

   If most people leave after visit two, the problem is in visit two, or in what happened between two and three. The cliff tells you where to look, not what is wrong.

   —Was the next appointment booked before they left the building, or were they told to call.

   —Did anyone contact them when they did not rebook.

   —Did the cost change at that visit, without warning.

   —Did they see a different practitioner.

   —Was that the visit where the treatment plan got long.

   In practice

   The single highest-return fix in most clinics is booking the next appointment before the patient leaves. Telling someone to call back later is a reliable way to lose about a third of them.

3. Split the curve by first service

   Retention differs enormously by how somebody entered. A patient who came in for an emergency behaves nothing like a patient who came for a checkup, and averaging them hides both.

   —Build one curve per entry category.

   —The entry category with the best retention is the one worth acquiring, even if it costs more per patient.

   —An entry category with terrible retention is not a growth channel, it is a treadmill.

Reactivation

1. Define lapsed honestly, by category

   Lapsed does not mean the same thing everywhere. For a recall-based practice it might be fifteen months. For an acute injury practice, a patient who finished treatment and left is not lapsed, they are discharged, and messaging them as though they abandoned you is offensive.

   —Pick the interval from your own data: the typical gap between visits, plus a margin.

   —Exclude anyone who was discharged, referred out, or asked not to be contacted.

   —Exclude anyone whose file suggests a reason not to contact them. Someone has to look at this list.

2. Check that you are allowed to contact them

   This is where a reactivation campaign becomes a complaint. Consent to receive care is not consent to receive marketing, and in most jurisdictions the second one has to be specific.

   —A reminder that a recall is due is usually care communication and is treated differently from a promotion.

   —An offer is marketing and needs marketing consent, with a way to opt out in every message.

   —Never send a patient list to an advertising platform as an audience without consent that specifically covers it. The list itself reveals health information by implication.

3. Write one message, not a sequence

   Two messages maximum, spaced two weeks apart, and then stop. Patients are not a funnel and a clinic that chases them feels like something other than a clinic.

   —Short. Four sentences.

   —Specific. Name when they were last in, and what is typically due.

   —One action. A link to book, or a number to call.

   —An easy opt out, honored immediately.

   —No discount, unless you have thought hard about what a discount says about your pricing to the patients who paid full price.

4. Measure it against a holdout

   Hold back ten percent of the lapsed list and send them nothing. Compare rebooking rates after eight weeks.

   Without the holdout you cannot tell the difference between a campaign that worked and a group of people who were going to come back anyway. A lot of reactivation reporting is measuring the second and calling it the first.

When it goes sideways