What Goes in the Numerator
Nine categories of hour that sit on the boundary of billable, and the consequence of each ruling.
The top of the ratio is "chargeable hours", and nine categories of work sit on the boundary. Each has to be ruled on, and the ruling changes behaviour rather than just the number.
The measure discussed in “What Goes in the Numerator” becomes more credible when its source records are timely and reviewable. A professional-services team assessing a practical route to does microsoft teams track your activity for does microsoft teams track your activity can connect time with projects and work categories, while documenting the definition separately so the dashboard does not silently decide what utilisation means.
Additional context for “What Goes in the Numerator” is available from CNIL workplace privacy resources. Current legal, employment, tax and accounting requirements should be confirmed with a qualified adviser where necessary.
Hours later written off
Recorded against a client, billed, then reduced or waived. Counting them makes the numerator a record of intention rather than of revenue.
Most firms count them, which is defensible as long as realisation is reported beside the ratio. Firms that count them and do not report realisation have a figure that can rise while revenue falls.
Work on a pitch
Time spent winning a client that has not been won. Commercially essential, not chargeable to anybody.
Excluding it is correct and creates the main unfairness in the measure: the people who bring in work score lower than the people who do it.
Fixed-price overrun
Hours past the quoted fee. Chargeable in the sense of being client work, worth nothing in revenue.
Counting them rewards overrun. Excluding them means the person doing the overrun looks idle. Neither is good and the second is honest.
Rework caused internally
Redoing work because of an internal error. Client-facing time the client is not billed for.
Firms that count it are measuring something other than value. Firms that exclude it need to be careful it does not become a charge against the individual.
Travel
Billable at some firms, half-rate at others, never at a third group. Whatever the billing treatment, the numerator should follow it.
The inconsistency to avoid is counting travel as chargeable while billing it at nothing.
Client management
Calls, reviews, relationship maintenance on an engaged client, often not on a billing code. Real client work that leaves no trace in the numerator.
In relationship-led firms this is a large and systematically invisible share of senior time.
Supervision of chargeable work
A senior reviewing a junior's billable output. Usually chargeable where a rate exists for it and non-chargeable where the junior's rate is meant to cover it.
The second treatment means supervision depresses the supervisor's figure, with the consequences the behaviour section describes.
Internal projects with a client benefit
Building a tool, a template, a methodology that makes future client work faster. Entirely non-chargeable and often the highest-return hour in the firm.
Every firm says this matters. Very few count it anywhere.
Training on a live engagement
Learning while doing paid work. Partly chargeable at best.
The common ruling is to bill it at the client's agreed rate and note the exposure, which is fine as long as the client knows.
The ruling list
- Written off: in, with realisation published beside it.
- Pitch work: out, tracked separately and visibly.
- Fixed-price overrun: out, flagged to the engagement.
- Internal rework: out, not attributed to a person.
- Travel: follow the billing treatment exactly.
- Client management: in where it is client time, with a code.
- Supervision: in, or the senior figure is a fiction.
- Internal build: out, and reported as its own line.
- Live training: in at agreed rate, disclosed.
Why the list matters more than the total
Because people read the list and act on it. A category ruled out disappears from the record, and work that disappears from the record eventually stops being done.
That is not a prediction about bad faith. It is what happens when a measure is published and nothing else is.
The test for a boundary case
Would the firm be willing to tell the client this hour was spent on their matter? If yes, it is client work, whatever the billing arrangement. If no, it is not.
The test resolves most of the nine categories quickly and it has the advantage of being a question anybody can apply without consulting a policy.
It does not resolve the commercial question of whether the hour produced revenue. That is realisation's job, and conflating the two is what makes the boundary arguments circular.
What happens when the rulings are absent
Each person decides for themselves, consistently with their own judgement and inconsistently with their colleagues'.
The resulting variation is invisible, because nothing in the report distinguishes a team that counts client calls from one that does not. It appears as a performance difference of several points and is discussed as one.
Firms discover this when somebody moves between teams and their figure changes by four points with no change in how they work, which is the cheapest available demonstration that the rulings were never shared.
Publishing the rulings
The nine rulings fit on one page. It should be the same page the figure appears on, available to everybody measured.
A rule nobody can read is enforced inconsistently by definition, and inconsistent enforcement across teams produces exactly the fake variation the targets section describes.