Where the Number Came From
Utilisation arrived with the billable hour and spread to firms that do not bill by time at all. What that history explains about how it behaves.
Utilisation came out of the billable hour: a model in which a firm sells time, records it in units, and bills it to clients. In that model the ratio is close to a revenue measure, because every chargeable hour recorded is an hour invoiced.
The measure discussed in “Where the Number Came From” becomes more credible when its source records are timely and reviewable. A professional-services team assessing explore the available workflow options for fte meaning can connect time with projects and work categories, while documenting the definition separately so the dashboard does not silently decide what utilisation means.
It then spread to firms that sell projects, retainers, deliverables and subscriptions, where the connection between a recorded hour and revenue is much weaker or absent entirely.
The assumptions behind “Where the Number Came From” can also be tested against Google re:Work resources. Record which recommendations apply to the real workflow and which do not.
Why that matters
In a time-and-materials firm, utilisation is roughly proportional to revenue and the incentive is roughly aligned. In a fixed-price firm it is not: an hour recorded on a fixed-price job adds cost and no revenue at all.
- Time and materials: an extra recorded hour raises what the client pays.
- Fixed fee: an extra hour raises cost and nothing else.
- Retainer: an extra hour consumes a fixed sum faster.
- Subscription or product: an extra hour has no billing consequence at all.
Running the same measure in both, with the same target, points people in opposite directions.
The fixed-price inversion
Under a fixed fee, the firm's interest is in fewer hours on the job, not more. A consultant who books eighty hours to a fixed-price engagement quoted at fifty has hit their utilisation target and lost the firm money.
The measure rewards exactly the behaviour the commercial model punishes.
The retainer case
Similar. Revenue is fixed per period; hours are a cost. A high ratio on retained work means the retainer is being consumed faster than it is being paid for.
Here the useful figure is hours against the retainer's assumed hours, which is a different measurement with a different denominator.
The subscription and product case
Weakest of all. Where revenue does not scale with hours at any point, utilisation measures only how time was recorded, and its connection to the firm's health is indirect at best.
Firms in this position frequently measure it anyway, because the neighbouring industry does.
What the history explains
Why the vocabulary is what it is. "Billable" and "non-billable" are terms from a world where billing followed recording directly, and they carry a value judgement that does not survive the move to other commercial models.
The unit of six minutes
Another inheritance. Tenth-of-an-hour recording came from legal billing, where the client is invoiced per unit and the precision is contractual.
Imported into firms that do not bill that way, it produces false precision and resentment, and it is the subject of its own note here.
What to ask before adopting it
Does our revenue move with recorded hours? If yes, utilisation is close to a revenue measure and the usual framing applies. If no, it is a capacity measure and should be named and used as one.
One question. Most firms have never asked it.
The capacity framing
Where revenue does not follow hours, the ratio still says something useful: how much of the firm's capacity is committed to client work rather than to itself.
That is a planning number, not a performance number, and framing it that way removes most of the damage described in this collection.
Mixed models
The common case: some time-and-materials, some fixed-price, some retained. One ratio across all three is a blend of three different meanings.
Splitting by commercial model is more work and it is the only way the figure means the same thing in each part of it.
What the inheritance brought with it
Three things, all of which outlived their origin. A vocabulary that splits the week into billable and non-billable and implies which half matters. A recording unit of six minutes. And the assumption that more chargeable hours is always better, which is true only where every chargeable hour converts to revenue.
In a time-and-materials practice all three are approximately correct. In every other commercial model at least one of them is wrong, and the firm inherits the apparatus without the condition that made it work.
Reading the measure through the model
The practical consequence is that the same reported figure needs a different interpretation in each part of a mixed firm.
Eighty per cent on time-and-materials work is revenue. Eighty on fixed-fee work is a cost position that needs comparing with the quoted hours. Eighty on retained work needs comparing with the retainer's assumption.
Three readings, one number. A report that does not split by model is asking its readers to hold all three at once, which they will not do, and the time-and-materials reading is the one everybody defaults to because it is the one the vocabulary was built for.
The one-line audit of your own model
For each revenue line, write down whether an extra recorded hour increases what the client pays. Time and materials: yes. Fixed fee: no. Retainer: no. Subscription: no.
The proportion of revenue in the "yes" column is the proportion of the firm for which the inherited framing is correct.
In most modern practices that proportion is well under half, and the measure is still being read as though it were all of it.
What to write down
Which commercial model each piece of work sits under, and what the ratio is being used for in each. Two columns.
It takes an afternoon, it is rarely done, and it determines whether the measure is pointing people toward revenue or away from it.