KPIs for Clinic Owners: The Seven Numbers That Actually Move a Private Practice

Most clinic software produces reports the way a printer produces paper: in quantity, on demand, and mostly unread. Owners describe exporting to a spreadsheet to answer the one question they actually had, and reviews of the UK incumbents list thin reporting among the reasons for leaving. The problem is rarely too little data; it is too many numbers and no agreement on which ones matter. Practice management software should produce the seven below without an export, and an owner should be able to read them in ten minutes on a Monday.

Each number has a definition, a place to read it and a lever. Three describe the diary, two describe the front door, two describe the business.

The diary

1. Diary utilisation

Definition: booked hours as a share of available hours, per clinician and per room. Available hours are the hours the clinician is rostered, not the hours the building is open, and the definition should be written down once so the number means the same thing every month.

Where to read it: the diary report, per clinician, per room, per location, with the trend over twelve weeks.

The lever: booking settings. Adjacent-slot offering closes islands between appointments; a waitlist refills cancellations; lead-in times stop unprepared bookings; buffers set for a reason and reviewed. A clinician at a persistently low share with a short lead time has a demand problem; the same share with a long lead time has a settings problem.

The UK Private Practice Barometer gives the sector a shared vocabulary for this number. Use it for vocabulary, and use your own trend for decisions.

2. DNA rate

Definition: did-not-attend appointments, missed without notice, as a share of booked appointments, per clinician and per appointment type. Late cancellations are a separate number, because the lever is different: a late cancellation is a patient who told you; a DNA is a patient who did not.

Where to read it: the attendance report, per appointment type, with the reminder delivery status beside it.

The lever: reminders that patients act on, with a one-tap confirm; a confirmation request on the appointment types that run high; a deposit or card on file for first visits where DNAs cluster; and a waitlist so that a late cancellation becomes a filled slot rather than a gap. The no-show calculator turns the rate into hours lost, which is the version that persuades a partner meeting.

3. Rebooking rate

Definition: the share of patients who leave an appointment with their next one booked, for appointment types where a next one is clinically expected.

Where to read it: the diary report, per clinician, weekly.

The lever: a rebook prompt at checkout and a one-tap rebook in the reminder, so the patient who cancels can move the appointment rather than drop it. This is the number that predicts next month's utilisation. A clinician with a low rebooking rate this week is a clinician with gaps in six weeks.

The front door

4. New-patient conversion

Definition: new enquiries that became a booked first appointment, as a share of all new enquiries, by channel: website booking, WhatsApp, phone, email, referral.

Where to read it: the inbox and booking reports together. Enquiries that arrive on a channel the software does not track are invisible, which is the strongest argument for running every channel through one inbox.

The lever: the booking page (no account wall, slots visible in three taps) and the reply. An enquiry answered within the hour books; one answered the next day has often booked elsewhere.

5. Enquiry response time

Definition: time from a patient's message to the clinic's first reply, per channel, per location, in clinic hours.

Where to read it: the inbox report, with open conversations at close of day.

The lever: a named person on the inbox each morning, templates for the ten most common questions, and routing overflow to another site. Response time is the only number here that reception can see move within a day, which makes it a good one to share with the team.

The business

6. Visits per episode of care

Definition: average number of attended appointments from first visit to discharge, per condition or appointment type, per clinician.

Where to read it: the clinical and diary reports together; it needs episodes to be opened and closed in the record.

The lever: this one is a conversation, not a setting. Wide variation between clinicians for the same condition is worth a clinical discussion about discharge criteria; a steady fall across the clinic is worth asking whether patients are drifting off before discharge, which the rebooking rate will usually confirm.

7. Revenue per clinician-hour, and open invoices

Definition: invoiced revenue divided by rostered hours, per clinician and per location, beside invoices raised and not yet paid.

Where to read it: the finance report, which is only trustworthy if invoices are raised from the appointment at the time.

The lever: utilisation and DNA rate move the numerator; the open-invoice list is a workflow fix, raise at checkout and send the payment link with the follow-up message.

A Monday routine

Open one report with the seven numbers per clinician and per location, with twelve-week trends. Spend ten minutes:

1. Find the one number that moved most against trend.
2. Find the clinician or appointment type behind it, because the clinic average hides the cause.
3. Change one setting or have one conversation.
4. Note it, and look at the same number next Monday.

Owners who try to move all seven at once move none. One number, one lever, one week.

What your software has to do

The seven numbers need five things from the platform: appointments with a status (attended, DNA, late cancellation) recorded at the time; reminders with a delivery status; every patient channel in one inbox with timestamps; invoices raised from appointments; and episodes of care opened and closed in the record. A platform that holds those produces the seven without an export. A platform that does not produces the spreadsheet the reviews describe.

Numbers to ignore

Almost every reporting screen offers numbers that feel like progress and change nothing. Website visits, social followers and newsletter opens describe marketing reach, not the clinic; they belong in a marketing review, not the Monday report. Total patients on file grows forever and says nothing about this month. Portal accounts created measures how firmly reception insisted at check-in. Average anything across the whole clinic hides the clinician or appointment type that is the actual cause. If a number cannot be tied to a lever that one person can pull this week, leave it off the page.

Sharing the numbers with the team

The seven numbers are per clinician, and a per-clinician report can become a league table that nobody wanted. Two rules keep it useful. First, share each clinician's own numbers with them, privately, with the trend and the lever, so a rising DNA rate arrives as "your follow-ups are being booked too far out, shall we shorten the window" rather than as a ranking. Second, share the clinic-level numbers with everyone, including reception, because response time and rebooking rate are the front desk's numbers as much as anyone's, and a team that can see them move will move them.

Treat visits per episode with particular care. It is a clinical conversation about discharge criteria, not a productivity target, and the moment it is used as one the number stops being honest.

How Capra reports them

Capra reports all seven per clinician, per appointment type, per room and per location, from the diary, the inbox and the invoices it already holds, with twelve-week trends and no export. The appointment reminder page covers the delivery status that makes the DNA number trustworthy, the practice management page covers the diary settings that are the levers, and a demo will show the Monday report built for your clinic.