How it works
Patient cohort analysis works by fixing the group first, then watching that group over time. A cohort is a set of patients who share one fact. They all booked their first injectable appointment in March. They all came from the same paid campaign. They all bought a membership in the same quarter. Once the group is defined, its members never change. New patients go into new cohorts.
From there the method is simple:
- Choose the shared trait: intake month, treatment line, acquisition channel, or offer.
- Choose one metric: rebooking rate, revenue per patient, visits in the first year, or churn.
- Set checkpoints, such as 30, 90, and 365 days after the first visit.
- Measure the same group at each checkpoint and leave the group closed.
- Line the cohorts up side by side and read the differences.
The value comes from the comparison. A single clinic-wide average blends a loyal three-year filler patient with someone who redeemed a discounted laser package last week and never came back. Cohorts pull those stories apart. You can see that January's cohort rebooked at twice the rate of February's, or that patients from referrals outspend patients from deal-site offers by a wide margin. That is something you can act on, not just a number you report.
Why it matters for aesthetic clinics
Aesthetic clinics live on repeat visits. Most of the money in a med spa arrives after the first appointment, through maintenance treatments, packages, and memberships. A blended average cannot tell you whether that repeat behavior is holding up, because last year's loyal patients quietly prop up this year's weak intake. The total looks fine right up until it does not.
Cohort analysis exposes what the average hides. It answers the questions clinic owners ask constantly. Is the new consultation flow producing patients who come back, or just patients who show up once? Is the discount offer buying revenue today and costing retention for the next two years? Which treatment line actually creates a long-term patient, and which one fills a chair and ends there?
It also sharpens spend. Acquisition cost only means something next to what a cohort turns out to be worth over time. Two channels can look identical on cost per lead and be completely different businesses once you follow each group for twelve months. Cohorts are how you tell them apart before you scale the wrong one.
Patient cohort analysis vs patient segmentation
Both group patients, but they answer different questions. Clinics often build segments and call it cohort analysis.
| Patient cohort analysis | Patient segmentation | |
|---|---|---|
| Group defined by | A shared starting point, usually a date or first treatment | A shared attribute, such as age, treatment interest, or spend tier |
| Membership | Fixed. The group never changes | Fluid. Patients move in and out as attributes change |
| Question answered | Is behavior changing over time, and for which intake? | Who are my patients right now, and how should I talk to them? |
| Typical use | Retention, payback, and quality of new patients | Campaign targeting and offers |
The Ownerized take
Most clinics we audit run on one blended number and make expensive decisions with it. We treat cohorts as the default view instead, because a clinic-wide average is just an average of good decisions and bad ones. As AI answer engines and search start sending a different kind of patient, the only honest way to know whether that patient is better is to follow their cohort forward, not to watch the monthly total tick up. That measurement layer is part of what we build under the AI Growth System.
Common mistakes
- Redefining the cohort after the fact so a campaign looks better than it was.
- Cutting cohorts so small that one high-spend patient swings the whole result.
- Comparing cohorts of unequal maturity, such as a 3-month cohort against a 12-month one.
- Measuring revenue only and ignoring visit count, which moves earlier and warns sooner.
- Stopping at 90 days, when injectable and laser cycles play out over a year or more.
- Letting patients drift between cohorts when they change treatment lines, which breaks the whole method.
- Building the report and never bringing it to the decision, so spend keeps following cost per lead.
Frequently asked questions
What is a good cohort size for a small clinic?
Group by quarter rather than by month if monthly cohorts hold only a handful of patients. You want enough people that a single outlier cannot move the result. A noisy cohort is worse than no cohort, because it invites confident decisions built on random variation.
How far out should I track a patient cohort?
Track at least twelve months. Aesthetic treatment cycles run long, and a cohort that looks weak at 90 days often turns strong once maintenance visits begin. Use 30 and 90 day checkpoints as early warning signals, but judge acquisition quality and payback on the full year.
Do I need special software to run cohort analysis?
No. Most practice management systems and healthcare CRMs already export first-visit date, treatment, and spend, which is everything a basic cohort needs. A spreadsheet handles the first version fine. Buy tooling once cohorts are changing your decisions, not as a way to start making them.
How is cohort analysis different from patient lifetime value?
Patient lifetime value is a number. Cohort analysis is the method that makes the number trustworthy. A blended lifetime value averages every patient you ever treated, including ones you would not acquire again. Cohorts show how that value is trending for the patients you are winning now.
Which cohort should a med spa start with?
Start with intake month split by acquisition channel, measuring rebooking rate at 90 days. It is the fastest way to see whether your marketing is buying patients or one-off transactions. Once that view is stable, add treatment line cohorts to find which services create long-term patients.