Your availability settings decide what strangers can do to your week. Get them right and a booking link quietly fills the gaps you actually want filled. Get them wrong and you'll find a call at 8am on a Monday, three meetings back to back with no gap, and something scheduled for twenty minutes from now while you're mid-task.
This guide covers each setting, what it actually controls, and how to combine them.
Weekly hours
The foundation is a weekly pattern: which days you take meetings, and between which times. Most people set something like Monday to Friday, 9:00 to 17:00, and stop there.
That's usually a mistake, because it publishes your whole working day as bookable.
Split hours are the point
You can set more than one block per day. Two blocks — say 09:00–12:00 and 14:00–17:00 — means the two hours in the middle are never offered. That's your lunch, your focus block, or just the buffer that stops your day being a corridor of meetings.
This matters more than it sounds. Publishing 9–5 as one block tells the world every minute of your day is available for anyone to claim. Publishing two blocks says the same thing about six hours and keeps two for you.
Buffers
A buffer is padding either side of a meeting that can't be booked over.
Buffer before gives you a few minutes to read the notes and get to the right room or the right tab. Buffer after gives you time to write things up before the next call starts. Ten to fifteen minutes each is typical.
Without buffers, a system that's working correctly will happily schedule you 10:00–10:30, 10:30–11:00 and 11:00–11:30. Technically no conflict. Practically, a bad morning.
Minimum notice
How far in advance someone must book. This is the setting that prevents the worst experience of all: a meeting appearing on your calendar for fifteen minutes from now.
Two to four hours suits most people. Twenty-four hours suits anyone who needs to prepare. Going much beyond that starts costing you meetings — the person who wants to talk today and can't will often just not talk to you at all.
Scheduling window
How far ahead people can book. The default is usually a couple of months.
Long windows sound generous and aren't. A meeting booked eleven weeks out has a high chance of being cancelled, rescheduled, or forgotten, and it blocks a slot you can't use in the meantime. Thirty to sixty days is a reasonable ceiling for most work.
Date overrides
Your weekly pattern is the rule; overrides are the exceptions. Use them for public holidays, annual leave, a conference, or a day where you're free at unusual times.
An override either blocks a date entirely or replaces that day's hours with different ones. The important habit is putting leave in as soon as you book it, not the week before — the whole value of a booking link is that it works without you thinking about it, and that only holds if it knows about your time off.
Daily limits
A cap on how many meetings a day can hold. Once the cap is hit, the day closes.
This is the setting that stops one bad Tuesday. Without it, an enthusiastic week can put seven calls on a single day while the rest of the week sits empty. Three to five is a common cap for meeting-heavy roles.
Focus-time protection
The subtlest one. It refuses to offer a slot if taking it would leave you without an unbroken block of free time that day.
The problem it solves is fragmentation. Four meetings scattered across a day can destroy it more thoroughly than six meetings in a row, because you never get a stretch long enough to do real work. Focus protection makes the system pack meetings together rather than scatter them.
It measures unbroken time per availability block, so a lunch break between two blocks doesn't count as your day being broken up. And it will never make you completely unbookable — if the rule would wipe out every slot on a day, it stands down for that day rather than leaving an invitee staring at an empty calendar.
Connect your calendar
Everything above describes when you're willing to meet. Your connected calendar describes when you're actually free. You need both.
With a calendar connected, existing events block the matching slots automatically — including events created outside the booking tool. You can connect more than one account, which matters if you keep work and personal calendars separate: a dentist appointment in your personal calendar should block a work slot even though nobody at work can see what it is.
Time zones
Set your own time zone once. Invitees see times in theirs, detected from their browser, and can change it manually if the detection is wrong.
This is handled for you, but it's worth knowing what it means: 2pm to you is not 2pm to them, and the confirmation email each of you receives shows the time in your own zone. Nobody needs to do arithmetic.
A worked example
For someone who takes external calls but needs to get real work done:
That publishes twelve hours a week rather than forty, keeps Mondays and Fridays clear, and still comfortably supports a dozen meetings a month. The instinct to open up more availability is strong and usually wrong — a full calendar of low-value meetings isn't better than a half-full one of good ones.
The short version
Split your day rather than publishing one long block. Add buffers. Set a minimum notice so nothing lands on you unannounced, and a daily cap so no single day gets destroyed. Put your leave in as overrides the moment you book it. Then connect every calendar you actually use, because willingness and availability aren't the same thing.