You Pay for Every Hour and Sell About Half of Them

A firm of thirty people has a payroll it pays every month without fail, and a revenue line that depends entirely on how much of that payroll reached a client invoice. Between the two sits a word everybody uses and nobody defines the same way.

Ask three partners what the firm's utilisation is and you will get three numbers. None of them is lying. They are describing different quantities and calling them the same thing, which is worse than disagreeing, because a disagreement can at least be resolved.

Six things that all get called an hour

There are six distinct quantities here and they descend in a strict order. Each one is smaller than the one above it, and something specific falls out in between.

Hours paid is the whole payroll. Hours available is what survives the calendar. Hours worked is what people actually spent on something. Hours chargeable is the part that belongs to a client engagement. Hours billed is what survived the review of the draft invoice. Hours collected is the only one that ever paid a salary.

Firms quote the second or the fourth. Banks and buyers care about the sixth. The gap between them is not a rounding difference, and it is not evenly distributed across the practice.

Six rungs, one word

Every firm quotes the rung that flatters it

1. 1. Hours paid

   Everything on the payroll. The number the firm is committed to whatever happens next.

   Lost before the next rung

   Leave, public holidays, sickness, notice periods, and the weeks between a joiner's start date and the day they are useful.

   Who owns that loss

   Nobody argues about this one, and nobody subtracts it either. It is HR's number and it never reaches an operational report.

2. 2. Hours available

   What is genuinely there to be spent, once the calendar has taken its share.

   Lost before the next rung

   Training, internal meetings, admin, proposals that were not won — and the hours nobody recorded at all, which are the largest part.

   Who owns that loss

   Everybody, and therefore nobody. This is the only gap on the ladder with no named owner, which is why it is the one that grows.

3. 3. Hours worked

   What people actually spent on something, as far as anyone can tell from what was written down.

   Lost before the next rung

   Internal work, rework, and time spent on an engagement that had no code open to receive it.

   Who owns that loss

   The engagement manager, who is also the person under the most pressure to keep the client happy this week.

4. 4. Hours chargeable

   The part of the work that belongs to a client engagement and could, in principle, be invoiced.

   Lost before the next rung

   Write-offs made at the draft invoice, discounts given to keep a relationship, and the line nobody is willing to put in front of this particular client.

   Who owns that loss

   The partner signing the bill — a decision made in a minute, on work that took a fortnight.

5. 5. Hours billed

   What survived the review of the draft and went out on an invoice.

   Lost before the next rung

   Disputes, credit notes, and debt that ages until somebody quietly stops asking.

   Who owns that loss

   Whoever chases, which in most firms is a person with no authority to escalate and no mandate to stop the work.

6. 6. Hours collected

   The only rung that paid a salary. It is also the one nobody quotes.

Which is why the meeting goes nowhere

Two people can both be right about utilisation and be four rungs apart. Naming the rung before quoting the number costs nothing and settles most of the argument — and a firm that cannot produce all six is not measuring utilisation, it is measuring recording.

Utilisation as a target produces recorded utilisation

Here is the mechanism that defeats most attempts to manage this, and it is not a people problem.

The moment utilisation becomes a target that a person is measured against, the cheapest way to hit it stops being doing more chargeable work and starts being recording more chargeable work. Nobody decides this. It happens gradually, in the small judgements that make up a Friday afternoon of catching up on the week — the hour that could reasonably be coded either way, the meeting that was arguably about a client, the round number that is easier to type than the honest one.

So a firm that pushes hard on utilisation gets exactly what it asked for. The number goes up. Realisation does not, because the extra hours were never billable and the partner writes them off at the draft. If you have read realisation rate and write-offs, this is the supply side of that problem: the write-offs are not evidence of bad pricing, they are the system correcting a number that was inflated three rungs earlier.

The tell is simple. If utilisation rises and realisation falls by roughly the same amount, nothing changed except the coding.

The bench nobody can see

Every consultancy has people who are not on anything this week. Very few can name them on Monday morning.

The reason is that a firm records where hours went, and the bench is defined by where they did not go. It is an absence, and an absence does not appear in a timesheet system — it appears as a person whose week added up to a smaller number than usual, which is exactly the observation nobody wants to make out loud about a colleague.

So the bench is discovered at month end, in a report, after the month is over and the hours are gone. And it is discovered in the aggregate — utilisation was low in July — rather than in the only form that could have been acted on: this named person had no work on the week of the fourteenth, and the pitch that would have used them was sitting unstaffed in the pipeline.

Capacity planning answered backwards

Then there is the version of this that costs the most, and it happens in a sales conversation.

A proposal goes out. The client accepts. Somebody looks at the delivery date and says yes, we can start in three weeks. Nobody in that conversation could see what was already committed for those three weeks, because commitment lives in engagement letters, in a partner's head, and in a project plan that has not been updated since kick-off.

So the month fills up twice. Not through carelessness — through the complete absence of a forward view. The firm knows precisely what it did in June and has no shared record at all of what it has promised for October.

That is the wrong way round. Recorded time is history and it cannot be changed. Committed time is the only thing a decision this week can affect, and it is the one thing most practices do not hold anywhere. The honest measure is capacity committed against capacity available, forward, by week — and it is a different report from utilisation in every respect except the units.

Grade mix, and the success that is a failure

One more distortion, and it is the one that survives longest because it looks like everything is fine.

A partner does a senior's work because the senior was busy and the client wanted it Thursday. Every utilisation report in the firm records that as a good outcome — the most expensive person in the building was fully occupied on chargeable client work. It is a margin failure, because the hour was sold at a rate that assumed a cheaper person would do it, or worse, absorbed into a fixed fee that was priced on a mix that never happened.

Read the last column on its own. Every row hides something that is decided forward, and every row is measured backward. That is the whole failure in one shape, and no amount of precision in the left-hand column fixes it.

What to make a vendor show you

On a live system, with real people and real engagements, not on slides.

1. Produce all six quantities for one month — paid, available, worked, chargeable, billed, collected — and show the gap between each pair.
2. Show next month's committed hours by person and by week, from signed engagements, before anybody has recorded a timesheet against it.
3. Show who has no work scheduled for the week after next, by name, today.
4. Take one engagement and show the grade mix it was priced on against the grade mix that actually delivered it.
5. Add a new engagement of a known size and show which weeks it makes over-committed, before it is accepted.
6. Show utilisation and realisation for one team on the same screen, over the same twelve months.
7. Show what a person recorded last week and what it cost the firm, without asking anybody to reconcile two systems.

Item 2 decides it. Everything else on this list is a report over history, and history is a solved problem — every timesheet product built in the last twenty years can total the past. A forward view of committed capacity requires the system to hold the engagement, its plan and its people as one thing, and that is a different piece of software from a timesheet with a nice chart on it.

Item 6 is the honest one. A vendor who will show you the two lines together is telling you something about their own confidence, because those two lines put side by side are where a firm finds out whether it has been managing utilisation or manufacturing it.

The short version

Six different quantities get called an hour, they descend in a fixed order, and only the last one paid anybody. Naming the rung before quoting the number costs nothing and settles most of the argument.

Utilisation as a target produces recorded utilisation. If the number rises and realisation falls by about the same amount, nothing improved — the coding moved. And a firm-wide average is the least useful figure in the building, because it averages the three people carrying the year with the four nobody has staffed.

The bench is an absence, so it does not appear in a system built to record presence. And capacity planning answered from last month's report is answered backwards: recorded time is history and cannot be changed, committed time is the only thing this week's decision can touch.

What that looks like when the engagement, the plan, the people and the ledger are one record is on the professional and business services page, and the system is ServX. The neighbouring pieces are timesheets that people actually fill in, realisation rate and write-offs, and client profitability. If the reports you already have are the thing you no longer trust, why your ERP dashboard is lying to you is the piece on that.