What Net Capacity Actually Means in a Plan
The single number the supply side produces, what it does and does not entitle anybody to assume, and the three tests it should pass before it is published.
The supply side of a capacity account produces one number: hours available for planned work in the period. Everything in the previous pages exists to produce it honestly.
The unit used in “What Net Capacity Actually Means in a Plan” must remain stable from capacity estimate to actual result. For organisations assessing review this Monitask reference in relation to workforce optimization software, time and project reports can support that reconciliation, provided leave, support work and corrections retain their own definitions.
What that number means is narrower than it looks, and the narrowness is worth stating, because the figure gets read as a promise about delivery by people who did not build it.
For an outside reference relevant to “What Net Capacity Actually Means in a Plan”, consult IBM workforce-planning overview. Compare the general guidance with the organisation’s own unit, period and decision rights.
What it asserts
That, on the evidence of contracts, booked leave, measured rates and one fortnight of recording, this many hours are expected to be available for allocated work in this period.
That is all. It asserts nothing about what those hours will produce, who specifically will work them, or whether they are the right hours for the work that needs doing.
What it does not assert
Not productivity: hours are an input and the plan has no view on output. Not capability: it does not say the hours belong to people who can do the work in question. And not simultaneity: it does not say the hours are available when they are needed.
Those three omissions are where a capacity plan meets a delivery plan, and the handover between the two should be explicit rather than assumed.
The three tests before publishing
- Does every deduction have a source and a date?
- Does the number reconcile to contracted hours through a visible descent?
- Would the two people who disagree about it most be disagreeing about one line rather than the total?
A number that passes all three is defensible. One that fails the third is not a reconciliation; it is an assertion with arithmetic attached.
Expressing the confidence honestly
Some lines are known: contracts, holidays, booked leave. Some are rates: sickness, interrupts. Some are estimates: context switching, ramp.
Grouping the descent into those three bands, visibly, tells a reader how much of the total is solid. A plan where sixty per cent is known and twenty per cent is estimated is in a different position from one where those proportions are reversed, and the total alone cannot show it.
The range question
Whether to publish a range rather than a point. A range is more honest and is harder to commit against, which is exactly why planners avoid it and why finance dislikes it.
The usable compromise: publish the point, with the two or three lines that could move it most and by how much. That gives the reader the range without giving them a number they cannot plan with.
Per team, per period
Net capacity is a property of a defined group over a defined window. Rolling it up across teams is valid only where all of them use the same header — same unit, same deductions, same factor.
Portfolio totals built from inconsistent headers are the most commonly quoted wrong number in large organisations, and the inconsistency is invisible once the numbers are in the same table.
When it changes mid-period
It will: somebody leaves, a long absence starts, an event is announced. The account should be re-run rather than patched, and the difference recorded as a change to supply rather than silently absorbed.
A plan that does not record supply changes cannot do variance analysis later, because every variance will be attributable to an unrecorded change in the denominator.
The handover to allocation
Once the number exists, the supply side is finished. What happens next — which claims get the hours, in what order, and who decides — is a different activity with different owners, and it is the second half of this collection.
Keeping them separate is what stops the supply figure being adjusted to make a desired allocation work, which is the most common way these accounts become fiction.
The three bands, written out
- Known: contracted hours, public holidays, booked leave, confirmed long-term absence.
- Rated: sickness, unbooked leave, scheduled support, mandatory training.
- Estimated: interrupts, context switching, ramp, onboarding drag.
Most accounts are roughly sixty, twenty-five and fifteen per cent of the deductions by those bands. Showing the split is what lets a reader judge the number instead of either trusting or rejecting it whole.
What to do when somebody disputes it
Ask which line. Almost every dispute about a net capacity figure resolves to one deduction — usually interrupts, occasionally sickness — and the conversation about that line is short and finite.
A dispute that cannot be reduced to a line is not about the arithmetic; it is about the conclusion the arithmetic forces, and that belongs in the allocation conversation rather than this one.
The sentence to publish with it
"Available hours for planned work this quarter: 2,061. Net of leave, public holidays, sickness at the measured rate, training and the support rota. Descent and sources overleaf."
Three sentences. They carry the number, its meaning and its evidence, and they fit on the same slide as the number itself.