Buffer time is padding placed before or after an appointment so the next booking cannot land immediately against it. Ten minutes of buffer after a thirty minute call means the following slot cannot start until forty minutes have passed.
It is one of the two or three settings that most changes how a booking page behaves, and it is the one most often set by guesswork. Too little and every overrun cascades through the rest of the day. Too much and you have quietly removed a third of your capacity without deciding to.
What follows is how buffers behave in practice, when to put them before rather than after, the arithmetic of what they cost, and a worked example you can hold your own settings against.
Why Back To Back Scheduling Fails
Back to back looks efficient. It is efficient in exactly one case: the appointment always finishes on time, requires no preparation, produces no follow-up work, and happens in the same place as the next one. Almost nothing meets all four.
The four things that break it:
- Overrun. Appointments that run long are not an exception, they are a distribution. A thirty minute booking that averages thirty two minutes is fine on its own and disastrous stacked eight times.
- Notes and admin. Anything that has to be written up, invoiced, filed or handed on takes time. If there is no gap, that work either happens during the next appointment or piles up until the end of the day.
- Travel and turnaround. Room changes, site visits, sanitising a chair, resetting equipment. Real minutes that no calendar shows.
- Human limits. Nobody delivers the eighth consecutive appointment as well as the first. A schedule with no breaks is a schedule that degrades.
The cost of getting this wrong compounds forwards. A ten minute overrun in the first appointment of a fully packed day is a ten minute delay for every client after it, and the last one waits an hour. Buffers exist to absorb variance so it does not propagate.
Before Buffers Versus After Buffers
Most systems let you set both, and they do different jobs.
An after buffer protects the schedule from the appointment that just happened. It absorbs overrun, gives you time for notes, and covers turnaround. This is the one most businesses need, and if you only set one, set this.
A before buffer protects the appointment from everything else. It guarantees preparation time: reading the file, setting up a room, reviewing what the client sent in their intake form, getting to the location. It is worth using when arriving cold has a real cost.
Some rough guidance on which to reach for:
| Situation | Before | After | Why |
|---|---|---|---|
| Sales or discovery calls | Short or none | Yes | Notes and CRM entry happen afterwards |
| Therapy and clinical sessions | Yes | Yes | Preparation before, notes and reset after |
| Salon and treatment chairs | Sometimes | Yes | Turnaround and cleaning dominate |
| Site visits and mobile work | Yes | Yes | Travel sits on both sides of the appointment |
| Tutoring and coaching | Short | Short | Light setup, some overrun tolerance |
One point of confusion worth clearing up: buffers between two of your own appointments do not usually stack. A ten minute after buffer followed by a ten minute before buffer typically produces ten minutes of gap in most implementations, not twenty, because the systems treat the buffer as a minimum separation rather than as two blocks laid end to end. Test yours before assuming either behaviour, since it changes your slot count considerably.
How Buffers Silently Reduce Bookable Slots
This is where the surprises live. Buffers do not remove capacity in proportion to their length. They remove capacity in steps, because slots have to fit whole inside your availability window.
Consider a straightforward setup: a 9:00 to 17:00 window, one hour unavailable for lunch, and thirty minute appointments. Seven working hours, so fourteen slots with no buffer.
Add a five minute buffer and each appointment now consumes thirty five minutes. Seven hours is 420 minutes, and 420 divided by 35 is exactly twelve. You lost two slots for five minutes of padding.
The step effect gets sharper as the buffer grows relative to the appointment. Here is the same window at four buffer settings:
| Buffer | Effective Slot Length | Slots In A 7 Hour Day | Change | Minutes Of Padding Per Day |
|---|---|---|---|---|
| 0 minutes | 30 minutes | 14 | baseline | 0 |
| 5 minutes | 35 minutes | 12 | 2 fewer | 60 |
| 10 minutes | 40 minutes | 10 | 4 fewer | 100 |
| 15 minutes | 45 minutes | 9 | 5 fewer | 135 |
Two things to take from that table. First, the jump from zero to five minutes costs two slots, and the jump from five to ten costs another two. The pain is front-loaded, not gradual. Second, at fifteen minutes you are giving up more than a third of the day's slots, which may be entirely correct for mobile work and clearly wrong for phone consultations.
Run this arithmetic on your own numbers before changing the setting. The formula is simply your available minutes divided by appointment length plus buffer, rounded down. Rounded down is the part people miss: a remainder of twenty five minutes at the end of the day is not a slot, it is nothing.
Slot Granularity Makes It Worse Or Better
There is a second setting quietly involved: the interval at which start times are offered. If your page only offers starts on the hour and half hour, a thirty five minute effective slot cannot start at 9:35. It has to wait until 10:00, and you lose twenty five minutes rather than five.
This is why some people set a buffer and see their availability collapse far beyond the table above. The buffer was not the problem, the interaction with a coarse start interval was. If your product offers a start interval or increment setting, align it with your effective slot length, or set it finer, typically five or ten minutes, and let the system pack the day properly.
The related rules for how the window itself is defined are covered in setting your availability rules, and they matter here because a buffer applied to a badly shaped window produces strange results that look like a bug.
Choosing A Buffer Without Destroying Capacity
A method that works better than picking a round number:
- Measure your actual overrun. For two weeks, note the finish time against the scheduled finish time. You do not need a system for this, a column in a spreadsheet is enough.
- Take something near the typical bad case, not the average. A buffer sized to the average is exceeded roughly half the time, which defeats the purpose. Sizing it so most days absorb cleanly is the goal.
- Add fixed costs separately. Travel and turnaround are not variance, they are known minutes. Add them on top rather than hoping the overrun buffer covers them.
- Check the slot arithmetic before saving. Work out what the new setting does to slots per day. If it costs four slots and you are already turning people away, look for another answer.
- Consider a longer appointment instead. If overruns are constant, the appointment is mis-sized. A forty minute booking with a five minute buffer is more honest than a thirty minute booking with fifteen.
That last point is the one worth sitting with. Buffers are for absorbing variance. If you are using them to hide a systematic underestimate of how long the work takes, you are lying to your own calendar, and the client is the one who experiences it as a rushed appointment.
When A Fixed Buffer Is The Wrong Tool
Some patterns need something other than a uniform pad:
- Different services, different needs. A consultation and a full treatment should carry different buffers. Set them per service rather than globally.
- Batching. Some work is genuinely more efficient back to back, and the real need is a longer break every few appointments. That is a break in your availability window, not a buffer.
- Travel between locations. A fixed buffer cannot know that the next client is forty minutes away. This needs either location-aware rules or deliberately separated blocks per site.
- A hard stop you must protect. If a school run or a shift end is non-negotiable, protect it in the availability window. Do not rely on a buffer to keep the last appointment away from it.
Buffers also interact with how soon someone can book. A tight buffer plus a very short minimum notice produces the classic bad morning: a booking appears twenty minutes out, right behind an appointment that is running long. The relationship between the two is covered in minimum notice and lead time.
Reviewing The Setting
Buffers are not set once. Revisit them when appointment types change, when you add staff, and when you notice the day consistently ending late. Two signals are worth watching: how often you finish more than fifteen minutes past the last scheduled end, and how often the page shows no availability inside a week. The first says the buffer is too small, the second says it may be too large, and seeing both at once means the appointment length itself is wrong.
If capacity is genuinely the constraint rather than the padding, the wider question is how many appointment hours you can offer at all, which is capacity planning territory. And if bookings are landing on top of each other despite a buffer being set, that is usually a sync problem rather than a buffer one, dealt with in double booking prevention.
How You Would Automate This
appntmnts.io applies before and after buffers per service rather than only globally, so a sixty minute treatment and a fifteen minute consultation can carry different padding on the same booking page, and the start interval can be set finer so a buffer does not cost you a whole slot to rounding. Because availability comes from two-way calendar sync, the buffer is applied against what is genuinely on the calendar rather than against a static timetable.
The features overview covers how buffers, availability windows and service durations fit together, and the free plan is enough to test a buffer against a real week before committing to it.