Capacity is the number of appointments you can deliver in a period without cutting corners. Most people carry an approximate version of it in their head, and the approximation is usually optimistic, because it is calculated from open hours rather than from bookable hours.
The useful version is a number you can write down and check against reality: appointments per week, per person, at a service mix you recognise. Once you have it, three decisions get easier. You know whether to hire. You know whether your prices are right. And you know whether a waitlist is a temporary measure or an admission that the diary is permanently oversubscribed.
This article gives the calculation step by step with a worked example, defines utilisation in a way you can measure from your own booking data, and covers the four honest responses to demand exceeding capacity.
Theoretical Versus Realistic Capacity
Theoretical capacity is open hours divided by appointment length. If you are open forty hours a week and each appointment is an hour, that is forty appointments. Nobody delivers forty.
Realistic capacity is what survives after everything that consumes bookable time without producing an appointment:
- Buffers between appointments for notes, resetting the room, and the ordinary overrun. See buffer times between appointments.
- Breaks, including a lunch that is genuinely taken.
- Admin blocks at the start and end of the day, plus recurring internal meetings.
- Slot granularity losses, the dead minutes created when appointment lengths do not tile neatly into your open hours.
- Leave and public holidays, which reduce the number of weeks you work rather than the hours in a week.
- Absence, both yours and the client failing to arrive.
The last two are different in kind from the others. Buffers, breaks and admin reduce the capacity of a working week. Leave and holidays reduce how many working weeks there are. Keep them separate in the arithmetic or you will double count.
The Calculation, Step By Step
- Start with open minutes per week: the hours on your booking page, times five days, times sixty.
- Subtract breaks: lunch and any fixed daily break, times the number of days.
- Subtract admin and internal meetings: daily admin times five, plus any weekly meeting.
- You now have bookable minutes per week.
- Work out the block length: service duration plus buffer.
- Divide bookable minutes by block length, rounding down per day rather than per week, since a leftover twenty minutes on Tuesday cannot be combined with a leftover twenty on Thursday. This is weekly capacity at full occupancy.
- Work out your working week factor: working weeks per year divided by fifty two, where working weeks account for leave and public holidays.
- Multiply to get average weekly capacity across the year.
- Apply your own measured no-show rate to get delivered appointments, unless you backfill cancellations from a waitlist.
Step nine is the one to be careful with. Do not use a figure you read somewhere. Your no-show rate is a number you can calculate from your own bookings in about five minutes, and it varies enormously by trade, price point and reminder practice. Count the appointments in the last quarter, count the ones where the client did not arrive and did not cancel in time, divide. That number is yours and it is the only one worth planning with.
A Worked Example
One practitioner, one service, a schedule you can adapt.
- Open Monday to Friday, 09:00 to 17:00.
- Lunch 12:00 to 13:00.
- Admin 30 minutes at the start of the day, none at the end.
- Team meeting, 60 minutes on Monday.
- Service 45 minutes, buffer 15 minutes, so a 60 minute block.
- Leave 25 days a year, plus 8 public holidays.
| Step | Working | Result |
|---|---|---|
| Open minutes per week | 8 hours x 5 days x 60 | 2,400 |
| Less lunch | 60 x 5 | 2,100 |
| Less daily admin | 30 x 5 | 1,950 |
| Less weekly meeting | 60 x 1 | 1,890 |
| Blocks per normal day | 390 usable minutes on Tue to Fri, divided by 60 | 6 per day |
| Blocks on Monday | 330 usable minutes, divided by 60 | 5 |
| Weekly capacity at full occupancy | (6 x 4) + 5 | 29 |
| Working weeks | (260 working days less 33 off) divided by 5 | 45.4 weeks |
| Working week factor | 45.4 divided by 52 | 0.873 |
| Average weekly capacity across the year | 29 x 0.873 | 25.3 |
So the answer is roughly 25 appointments a week averaged across the year, against a theoretical forty. The gap is not waste, it is the buffers, breaks, admin and leave that make the work sustainable. What matters is that 25 is the number you plan revenue and hiring against, not 40.
Two adjustments people usually want to make next. If you run a mixed service list, calculate the block length as a weighted average using your real service mix from the last quarter rather than assuming an even split. And if you deliberately hold slots back for urgent or repeat clients, subtract those before you quote yourself a capacity figure, because a held slot is not available capacity.
Utilisation: The Number Worth Tracking
Utilisation is delivered appointment time divided by bookable time, expressed as a percentage. It is the single most useful operational number a booking desk has, because it moves before revenue does and it explains why revenue moved afterwards.
Measure it monthly per person, from your booking data:
- Bookable minutes: the capacity calculation above, for the actual days worked that month.
- Delivered minutes: appointments that happened, counted at service duration, excluding buffers.
- Utilisation: delivered divided by bookable.
Count the service duration and not the block, otherwise a buffer change moves your utilisation without anything real changing. And count delivered rather than booked, or a month of no-shows will look like a good month.
Track two companion numbers alongside it and the picture gets much sharper: the proportion of offered slots that were booked at all, and the proportion of booked appointments that were delivered. The first tells you about demand. The second tells you about your reminder and policy discipline, which is covered in waitlist management and the no-show material.
Reading The Number
Utilisation is a diagnostic, not a target, and the right level depends on how variable your work is. Some general reading:
| Pattern | Likely Meaning | Sensible Response |
|---|---|---|
| Low and flat | Demand shortfall, or a booking page that is hard to use | Check the page as a client sees it before changing anything else |
| Low with slots going unbooked at specific times | Your hours do not match when clients want you | Shift hours rather than adding them |
| Rising steadily | Demand growing into capacity | Plan the response before you hit the ceiling, not after |
| Very high, sustained | No slack for overruns, urgent work or illness | Treat as oversubscribed, not as a success |
| High but revenue flat | Busy on low value work | Look at the service mix and at pricing |
Running near the top of your capacity permanently is fragile rather than efficient. One overrun cascades through the day, one illness cannot be absorbed, and there is no room for the urgent client who would otherwise become a good long term one. Most single practitioner businesses find they need visible slack in the diary, and the amount is something you learn from your own overrun pattern rather than from a rule of thumb.
When Demand Exceeds Capacity
There are four honest responses, and most businesses need a combination.
Run A Waitlist
The cheapest first move. A waitlist converts cancellations into filled slots and turns "no availability" into a queue rather than a lost client. It raises delivered capacity without raising theoretical capacity, which is exactly what you want when the shortage is intermittent. See waitlist management for how to run one that people do not resent.
Extend The Booking Horizon
If you are full for three weeks and your page only shows two, clients see nothing and leave. Extending the horizon lets people book further out, which is fine for planned work and poor for urgent work. The cost is that a longer horizon locks in more of your future diary and makes changes harder, so extend deliberately rather than opening a year.
Add Capacity
More staff, longer hours or shorter appointments. Longer hours are the fastest and the least sustainable. Shorter appointments raise throughput and lower quality, and usually mean shorter buffers too, which is where days start overrunning. More staff is the real answer but it needs enough sustained demand to carry the cost, which is where a few months of utilisation data earns its keep.
Raise Prices
The response people avoid and often the correct one. If you are consistently full weeks ahead and turning work away, the price is below the market clearing level. Raising it reduces demand to what you can serve and increases revenue on the same capacity, which is the only one of the four options that improves the numbers without adding hours. Pricing your appointments covers how to approach that.
When Capacity Exceeds Demand
The reverse problem is more common and gets less attention. Before adding marketing spend, check the mechanical causes, because they are cheap to fix: minimum notice that is longer than clients plan ahead, a booking horizon too short to catch people who plan further out, availability that sits in hours your clients are at work, buffers wide enough to halve your slot count, or a daily cap someone set during a busy month and never removed. Setting your availability rules covers each of those.
Reviewing It Monthly
Capacity planning is not a one off exercise. Once a month, spend ten minutes on four numbers: utilisation per person, the proportion of offered slots booked, the proportion of booked appointments delivered, and the average lead time between booking and appointment. That last one is the earliest warning you have. When lead time is stretching month on month, you are approaching your ceiling, and you have time to choose a response rather than having one forced on you.
Getting The Data In appntmnts.io
appntmnts.io records bookings, cancellations and no-shows against each service and each team member, which gives you the delivered minutes and the counts the calculation above needs without exporting anything by hand. Availability rules, buffers and caps sit in the same place, so a capacity change is a settings change rather than a rebuild of the schedule, and the REST API is available if you want to pull the figures into your own reporting.
The features page covers the scheduling and reporting side, the developer documentation covers the API, and pricing shows what the free plan includes. If your capacity problem is intermittent rather than structural, start with waitlist management before you change anything else.