What Meetings Actually Cost

A recurring meeting is a standing purchase order that nobody ever re-approves. It gets authorised once, usually informally, and then renews every week until someone actively cancels it — which almost never happens, because cancelling is a visible act and continuing is not.

Putting a number on it is a way of making the invisible purchase visible. Below is how to do the arithmetic properly, what it does and does not capture, and how to use it without becoming the person nobody invites anywhere.

The basic calculation

The obvious version is attendees × duration × hourly rate. That is the right starting shape and it understates the answer, usually by a lot.

Start with the hourly rate. Annual salary divided by 2,080 (52 weeks × 40 hours) gives the naive hourly figure — £60,000 becomes roughly £29/hour. But salary is not what an employee costs. Fully-loaded cost — payroll taxes, benefits, equipment, software licenses, office space, the share of recruiting and management overhead — is conventionally estimated at somewhere between 1.25× and 1.4× salary for most office roles. That is a planning heuristic used in budgeting and consulting, not a law of nature, and the real multiplier varies enormously by country, benefits structure, and whether the organization carries property costs.

So a more honest hourly figure is:

(annual salary × 1.3) ÷ 2,080

On £60,000 that is about £37/hour rather than £29. For an eight-person hour-long meeting the difference between the two methods is roughly £64 — per occurrence, every week, forever.

What the simple number misses

Three costs sit outside the arithmetic, and in aggregate they are frequently larger than the meeting itself.

The fragmentation cost. A one-hour meeting at 2pm does not remove one hour from the day. It removes the hour, plus the wind-down before it, plus the time to get back into whatever required concentration afterwards. It also does something worse: it splits the surrounding time into two blocks too short for demanding work. A meeting placed in the middle of an otherwise open afternoon can neutralise the afternoon. This is why two meetings back-to-back are often cheaper in real terms than the same two meetings three hours apart, and why calendar placement deserves as much thought as calendar quantity.

The preparation and follow-up cost. Pre-reads, slides, notes, actions written up afterwards. For a status meeting this can exceed the meeting duration, and it is borne unevenly — usually by whoever is most junior.

The queue cost. If a decision waits for the weekly sync, the average wait is half a week regardless of how long the meeting runs. For anything blocking other work, latency dominates. A five-minute decision that waits three days cost three days, and no meeting-cost calculator will show that.

Conversely — and this matters, because meeting-cost arguments are usually made in bad faith in one direction — the calculation also ignores everything a meeting is for. Alignment that prevents two people building the same thing twice. A disagreement surfaced in ten minutes that would have taken three email rounds. Someone new understanding why a decision was made. These are real and they are not on the invoice either. A cost figure is one side of a ledger, and treating it as the whole ledger is exactly the error the number is supposed to correct.

Using the number without being insufferable

There is a particular failure mode here. Someone discovers meeting-cost arithmetic, announces the running total mid-meeting, and achieves nothing except making themselves tiresome. The number is an argument for a decision, not a weapon to deploy in real time.

It works in three situations.

Reviewing a recurring invite. This is the strongest use by far. "This is a weekly, eight people, an hour — that is roughly £15,000 a year of salaried time. Is it earning that?" Annualising is the move. A £300 meeting sounds trivial; the same meeting at £15,000 a year invites a real answer, and often the answer is a defensible yes.

Sizing the attendee list. Cost scales linearly with attendees while value usually does not. The eighth person in a decision meeting is frequently there for visibility, and visibility can be served by good notes at a fraction of the cost. Framing the invite as a budget makes the trade-off legible: every added person is a purchase.

Choosing a format. A written update read by eight people at their own pace costs a fraction of eight people synchronised into the same hour. Some things genuinely need the synchronous version — anything contested, anything sensitive, anything where you need to read the room. Status is rarely one of them.

Reading the calculator honestly

The calculator on this page runs the simple model deliberately: attendees, average salary, elapsed time. It uses salary rather than fully-loaded cost, which means it understates the true figure by roughly a quarter to a third. That is a defensible choice for a live counter — salary is the number people actually know — but it is worth knowing which direction the error runs.

It also cannot know whether the meeting was worth it. A £400 hour that prevents a month of misdirected work is the best £400 the organization spent that week. The number tells you what you paid, not what you got, and anyone quoting it as though it settles the question is misusing it.

The lifetime estimator alongside it does something different and less rigorous: it extrapolates your current meeting load across a career. Treat that one as a provocation rather than a forecast. Meeting loads change with seniority, industry and role, usually upward, and no straight-line projection survives contact with an actual career. It is there to make a weekly number feel like a real quantity, which is a legitimate thing for a number to do even when the projection itself is crude.