How many meetings is too many is a question that wants a number back. Something like “more than fifteen hours a week”, so you can hold your calendar up against it and be told you are right.
The number does not exist. Not in a form you should quote, and not in a form that would help you if it did.
Here is what to use instead: two ratios you compute from your own calendar, and a threshold you set yourself against the work you are actually accountable for. It is less satisfying than a benchmark and considerably more useful in a conversation with a manager, because your manager cannot argue with your own numbers the way they can argue with an average.
Why there is no public number for how many meetings is too many
Search the question and you will find figures. Weekly meeting hours, hours lost to unproductive meetings, percentages of the working week. They appear everywhere, they are quoted confidently, and they almost all originate from the same kind of place.
The prominent ones — Fellow’s meeting-statistics roundup, Flowtrace’s State of Meetings reporting, the benchmark posts published by scheduling vendors like Reclaim — are produced by companies selling meeting and calendar software. That does not make them dishonest. It makes them marketing surveys with self-selected samples, aggregated from customers who already believed they had a meeting problem, published by a party with an interest in the number being alarming.
Cited by name and year, they are fine evidence that a lot of people think they are in too many meetings. They are not evidence of what a healthy load is. When a post tells you “studies show” and links to a SaaS blog, the study is the SaaS blog.
There is a second reason the average would not help you even if it were rigorous. Meeting load is not a quantity, it is a fit. A recruiter, a staff engineer, an EA and a VP of sales doing identical hours are in four completely different situations, and three of them are fine. Averaging across them produces a number that describes nobody.
So stop looking for the threshold. Compute the two ratios.
Ratio one: load against your obligations
Meeting hours ÷ contracted hours. Four weeks of data, preparation and follow-up included. The calendar audit walks through getting the export and counting the hidden preparation time, which is the part people leave out and the part that changes the answer.
Now the part that makes it mean something. Write down, in one line each, the things you are actually accountable for delivering this quarter. Not your responsibilities — your deliverables. Most people can list three to five.
Then estimate the focused hours each one needs per week. Be pessimistic; you are bad at this and so is everyone.
Add the two together. If meeting hours plus required focused hours exceeds your contracted hours, you do not have an opinion about your meeting load, you have an arithmetic problem, and it is one you can put in a sentence:
“The three things on my plate this quarter need about eighteen focused hours a week between them. I’ve got about eleven. I’d like to work out which meetings come off, or which deliverable moves.”
That sentence works because it is not about meetings being annoying. It is about a commitment that cannot be met, which is your manager’s problem as much as yours. The version that fails is “I’m in too many meetings” — that one invites a conversation about prioritisation skills.
If the arithmetic comes out fine and you still feel buried, your load is not the issue. Go to ratio two.
Ratio two: the longest unbroken block
For each working day, find the longest continuous stretch with no meeting in it. Count a gap of under 30 minutes as no gap at all, because it is not one — a 25-minute window between two calls is a waiting room, not working time.
Take the median across your twenty days. Not the average. One empty Friday drags an average up by an hour and tells you nothing about a normal Tuesday.
Then compare it against what your work needs. This is the part you have to answer for yourself:
- Work that is mostly responsive — support, coordination, account management, most EA work — genuinely runs on short blocks. A median of 45 minutes may be entirely fine.
- Work that requires holding a complicated thing in your head — writing, design, analysis, engineering, anything where you lose the thread when interrupted — needs a median block long enough to get in and get something done. If yours is under 90 minutes and your job is this kind, the number is your problem regardless of what your total load says.
A moderate load with a shattered day feels far worse than a heavy load with two clear mornings, and it is the single most common reason someone audits their week and finds “normal” hours they cannot survive. Fragmentation is its own diagnosis with its own fixes, and declining meetings is not the main one.
The third number, if you want a target
Listen share. The proportion of your meeting hours where you are not expected to speak. Categorise each meeting as run, contribute, or listen and total the hours.
This is not a diagnosis, it is a budget. Whatever your load, the listen hours are the ones you can most plausibly recover, because getting out of them requires nobody to agree that the meeting is bad — only that you specifically are not needed in it. That is a far smaller concession to ask for.
If your listen share is above about a third, you have a straightforward first move and it is not a policy change. It is one message to one organiser, which is what the declining scripts are for.
If your listen share is near zero and your load is still too high, you are in a harder position: every hour is an hour someone expects your input in. That case is real, it is common for people in coordination roles, and the fix is upstream in how the meetings are designed rather than in whether you attend them.
How to set your own threshold
Do it once, write it down, and revisit it quarterly.
- List the deliverables you are accountable for.
- Estimate the weekly focused hours they need. Add 20% because you underestimated.
- Subtract that from your contracted hours. What remains is your meeting budget.
- Compare your actual meeting hours to it. The gap is the number you take into a conversation.
- Note your median unbroken block alongside it. If the budget balances but the block is too short, the problem is placement, not volume.
Your threshold will not match anyone else’s, and that is the point. It is defensible in a way an industry average is not, because it is derived from commitments your manager already agreed to.
Write the number down somewhere you will find it. In three months, when your load has crept back up — and it will, because freed calendar time gets colonised faster than anyone expects — you will want the original arithmetic rather than a fresh argument from scratch.
Common questions
Is there any load that is objectively too many? There is a shape that is objectively broken: when no unbroken block exists on any day of the week, sustained across a month. Nobody’s job requires that, and it produces work that has to be redone. But that is a shape, not a number of hours.
What about a rule like “no more than half your week in meetings”? It is a reasonable personal policy and a bad benchmark. Half a week is an enormous load for an individual contributor whose output is written work and a light one for a manager whose output is decisions. If you find the rule useful as a personal ceiling, keep it — just do not cite it as though it were established.
My manager says everyone is at this load. That may be true, and it is not a rebuttal. The counter is your arithmetic: these are the deliverables, this is the focused time they need, here is the gap. Everyone being underwater is a description of the department, not a reason your specific commitments can be met.
Can I use a tool to compute these? Most calendar analytics tools will give you the load ratio. Few compute a median unbroken block, and none of them know your deliverables, which is the half that makes either number mean anything. The spreadsheet takes 90 minutes once and then twenty a quarter. Start there.