The number almost nobody knows
Ask ten knowledge workers how many hours they have for project work this week. Nine will answer with their contracted time — 40 hours, maybe 38.
As a basis for planning, that number is useless. It describes attendance, not available working time. For most people there is more than a third of the week between the two.
The difference is not a detail; it is the cause of a recurring pattern. Commitments are made against 40 hours and delivered against 25. The gap shows up as delay, as overtime, or as the quiet backlog carried into Friday. And because nobody knows the number, people look for the fault in their discipline when it is in the arithmetic.
Disclosure: we build Balane Plan, an app that keeps this calculation running. The method below works in a spreadsheet.
The calculation
contractual hours
− fixed meetings (including preparation and follow-up)
− administration, email, coordination
− recovery after meetings
− context switching losses
− absence (pro-rata holiday, sickness)
= plannable weekly capacityApplied to a typical week with 40 contractual hours:
| Item | Hours |
|---|---|
| Contractual hours | 40.0 |
| − fixed meetings (8 meetings, averaging 45 min plus prep and follow-up) | − 8.0 |
| − email, coordination, administration | − 5.0 |
| − recovery after longer meetings (10–15 min each) | − 1.5 |
| − context switching (approx. 8/day × 15 min × 5 days) | − 5.0 |
| − pro-rata absence (30 days holiday + sickness ≈ 13%) | − 5.2 |
| Plannable weekly capacity | 15.3 |
Fifteen hours. That looks low, and it is nonetheless an honest calculation for somebody with eight meetings a week. Fewer meetings lands you at 22 to 28 hours. A leadership role frequently lands under 12.
The three items that surprise people most:
Preparation and follow-up. A 60-minute meeting rarely costs 60 minutes. It costs 15 minutes of preparation, 60 minutes of meeting, 10 minutes of follow-up and 12 minutes before your head is back in the actual work. Realistically just under 100 minutes. A weekly that really costs 90 minutes must not be budgeted as 60.
Context switching. Five hours a week for nothing but changing gear is the norm, not the outlier. The arithmetic is unpleasant but checkable.
Pro-rata absence. 30 days of holiday is 12 per cent of the year. Making annual commitments against 52 full weeks means having sold six weeks too many.
Utilisation: the number behind the number
Once capacity is established, utilisation becomes interesting:
utilisation = committed work ÷ plannable capacityAnd here is the second uncomfortable finding: the right target is not 100 per cent.
A system running at 100 per cent utilisation has no response time. Every disturbance — and there is one every week — pushes everything behind it, and with no slack it never catches up. The backlog grows monotonically.
Workable targets:
| Utilisation | Effect |
|---|---|
| under 70% | Reserves unused, usually an unrealistically low estimate |
| 75–85% | Target range: disturbances absorbed, commitments hold |
| 85–95% | Deadlines hold only when nothing goes wrong |
| over 100% | The backlog grows; the only question is where it becomes visible |
The practical consequence: with 15 hours of plannable capacity you commit 11 to 13 hours, not 15. The remaining hours are not idle time — they are the reason the committed hours hold.
How to get your real numbers
The example above uses experience values. Your own take two to three weeks:
1. Count the meeting load. Readable from the calendar, five minutes of work. Do not forget preparation and follow-up, or the number is systematically low.
2. Sample your administrative time. Track for three days how much goes to email, coordination and organising. Most people underestimate this item by a factor of two.
3. Count context switches. A tally sheet is enough. Multiply the count by 15 minutes.
4. Calculate from measurement, not from estimate. If you are already measuring work sessions — say, for your correction factor — you have plannable time as a by-product: it is the sum of the focus time actually worked.
That is exactly the route Balane Plan takes: because every work session is measured server-side and every pause carries a reason, weekly capacity comes out of observation rather than assumption — including an eight-week forecast and a what-if simulation before a project is moved.
What to do with the number
A capacity figure is not an end in itself. It is the basis for three conversations that were not possible before.
The commitment conversation. "Can we still fit that in?" is a feeling without a capacity figure. With one it is arithmetic: 13 hours committed, 4 hours new, 15 hours available — so only if something else gives. The question "what should wait for it?" is uncomfortable and considerably cheaper than answering it in four weeks' time.
The meeting load conversation. If eight meetings a week cost five hours of plannable time, cancelling two meetings is worth more than any productivity technique. That calculation is persuasive in a team because it makes no claim about anybody's motivation.
The prioritisation conversation. Capacity is what makes prioritisation both necessary and possible. Without a ceiling, prioritisation is academic, because everything appears to fit.
The most common objection
"If I calculate it like that, I can barely commit to anything."
That is the right observation and the wrong conclusion. You could not commit to more before either — you simply did it anyway, and the difference showed up as delay, as evening work, or as quiet backlog. The calculation does not create the scarcity, it makes it visible.
What changes is the order of disappointment: it happens at commitment time instead of at delivery time. That is more uncomfortable in the moment and better in every other respect — for the person receiving the work as much as for the person doing it.
In short
Weekly capacity is the number standing between a task list and a promise that holds. Knowing it costs two weeks of measurement; ignoring it costs the difference between 40 and 25 hours, permanently.
And it is the precondition for day planning being anything more than a politer way of spreading the same overload across five days.
Common questions
How many hours a week are realistically plannable?
On a 40-hour week, 22 to 28 hours of plannable working time is a realistic figure. The rest goes to meetings, coordination, administration, context switching and recovery. In leadership roles it is often under 15 hours.
Is low plannable time a sign of poor organisation?
No. Coordination, leadership and administration are work — just not plannable project work. The problem only starts when commitments are made against contractual hours instead of against plannable time.
How do I handle holiday and sickness in capacity planning?
Holiday is subtracted as known absence, sickness as an annual experience value — usually between 3 and 6 per cent. Ignoring both means planning a year that only exists for healthy people who never take leave.
How often should capacity be recalculated?
Once a week for the coming week and once a quarter for the outlook. More often does not pay, because the numbers move slowly; less often is too coarse, because meeting load varies noticeably from week to week.