A Ratio of Hours to Hours
Utilisation is a quotient with two chosen ends. What the ratio literally is, what it is a measure of, and why the arithmetic is the easy part.
The figure one firm reported to its partners last quarter
78.4%
Numerator: Hours booked to client codes, including hours later written off
Denominator: Contracted hours less approved leave
Published as a single percentage, with no note about either count.
The same quarter, read net of write-offs against attended hours, gives 61.2%. Neither number is wrong and the report did not say which one it was. One firm's own quarter, not a benchmark.
Utilisation is hours divided by hours. A count of time spent on work that can be charged, over a count of time the firm paid for. Both counts are choices, both have several defensible answers, and the arithmetic between them is the only part of the exercise nobody argues about.
The measure discussed in “A Ratio of Hours to Hours” becomes more credible when its source records are timely and reviewable. A professional-services team assessing review the corresponding official page for getting teams to meet deadlines can connect time with projects and work categories, while documenting the definition separately so the dashboard does not silently decide what utilisation means.
Everything difficult about the number is in the two definitions, and almost nobody writes either of them down.
Teams documenting “A Ratio of Hours to Hours” may use QuickBooks business resources as an outside reference. Keep the source record, calculation rule and resulting decision visible as separate steps.
The numerator
Hours on work that can be charged. Which raises the first question immediately: booked, billed, or collected?
Booked is what was recorded. Billed is what reached an invoice. Collected is what was actually paid for. They differ by write-offs and by recovery, and they can be fifteen points apart.
The denominator
Hours the firm paid for. Which raises the second: contracted, paid, or attended?
Contracted ignores overtime and leave. Paid includes overtime. Attended removes leave and sickness. Each is a different statement about what the firm considers available.
The spread
Six defensible pairings of those choices, applied to one person's one week, give answers between about 62 and about 87 per cent. That is not a rounding difference; it is a quarter of the scale.
Which means a utilisation figure quoted without its definition carries no information at all.
What it is a measure of
How much of the time a firm pays for is spent on work it can charge for. That is a real and useful thing to know about an operation.
It is a property of the operating model: how much supervision, business development, administration and training the work requires before anybody touches a client.
What it is not a measure of
Effort, productivity, value delivered, or whether the work was any good. The instrument does not contain any of those and no analysis recovers them.
Most of the damage this number does comes from being read as though it did.
Why it feels like a productivity measure
Because it has a percentage sign, because it is higher when people are busier, and because the alternative to billable time is described, in every firm, as "non-billable" — a word that sounds like an absence rather than a category of work.
The vocabulary does most of the misreading before anybody opens a spreadsheet.
The ceiling
There is one, and it is well below a hundred. Supervision, training, absence cover, business development and administration do not disappear.
In most professional firms the structural ceiling is somewhere between seventy-five and eighty-five per cent of attended hours. A figure above that is usually a recording decision rather than an achievement.
The floor
Also real. Below some level the firm is not covering its costs, and the number is a signal that either demand or pricing has a problem.
Neither of those is solved by the number going up, which is the most common response to it going down.
Why firms measure it at all
Because labour is the cost and chargeable time is the revenue, so the ratio between them is the simplest available statement about whether the operation works.
That is a legitimate reason. It does not license the number to answer other questions.
What this collection is about
Not how to raise it. What it measures, why six definitions give six answers, what happens to a week once anybody is appraised on it, what it cannot see, and what has to be read beside it.
The last of those is the only part that makes the number usable, and it is the part that is almost never published.
The arithmetic on one week
Thirty-five contracted hours. Four on an internal review, three on a proposal that may not land, two on a system that would not open, one on a conversation that was half a client matter and half a courtesy.
Twenty-five hours against a denominator of thirty-five is 71 per cent. Move the ambiguous hour and it is 74. Remove the system failure from the denominator and it is 76.
Three defensible numbers from one unchanged week. The spread is wider than most of the differences a firm takes seriously, and nothing in the week decides between them.
Where the ratio is uncontroversial
At the level of a whole firm over a year, under one definition applied once, the figure is a reasonable description of how capacity was spent.
It becomes contentious when it is cut finer — by person, by month — and asked to support judgements about the people in it.
The measure did not change between those two uses. What changed is how much weight a single arithmetic ratio is being asked to carry, and the fineness of the cut is a good predictor of how much trouble it causes.
The single sentence
A utilisation figure needs its numerator and its denominator written beside it, in words, every time it appears. Twenty words.
Everything in this collection either explains why that sentence matters or describes what goes wrong in its absence.