You Recorded a Hundred, Billed Seventy, and Called It a Full Job
· 7 min read · Written by Faceela Research & Editorial Team
Reviewed by Ahmed Hassan Algammal — Founder and Enterprise Systems Consultant
Ask a partner how a job went and you will hear about the deliverable. It went out on time, the client was happy, there were no queries. That is a real answer to a real question, and it is not the answer to whether the firm made anything.
The arithmetic that answers that has a name. The hours recorded on the engagement, at the rate card, are its standard value. What eventually reaches the bank is the cash. Everything in between is a series of reductions, each of which is a sentence about the firm — and in most practices not one of those sentences is written down anywhere.
The bridge, and what each step confesses
Standard value is the softest number on the list, and it is worth saying so at the top. It is not revenue. Nobody has agreed to pay it. It is your own rate card multiplied by your own timesheets, which makes it entirely an internal opinion — useful precisely because it is the same opinion applied to every job, which is what makes the comparison between jobs mean something.
From there it comes down in named steps, and each of them admits something different.
The shape of it
From what the hours were worth to what the bank received
Start
Standard value of time recorded
Every hour on the engagement at the rate card. It is not revenue and nobody has agreed to pay it — it is the only figure in the bridge that is entirely your own opinion.
less
Fee agreed below standard
The discount, the fixed fee that came in under the estimate, the rate concession given to win the work.
What it admits
A pricing decision. Made before the work, deliberately, by somebody with the authority to make it. Nothing went wrong here.
less
Time that was never chargeable
Training, internal work, a proposal that did not land, the hour the partner spent on a favour. Only deductible here if it was recorded as non-chargeable at the time it happened.
What it admits
An investment or an overhead, depending on which it was — and the firm can only tell the difference if the entry said so on the day.
less
Written down at invoicing
The hours were real, the client will not wear them, and somebody quietly reduces the bill on the way out of the door.
What it admits
A delivery failure. The job took more hours than it was worth, or than it should have taken, and this is the only line in the bridge that says so.
equals
Fee billed
The number the client sees. It is also the number most firms treat as the beginning of the story rather than the middle of it.
less
Credit notes and post-invoice concessions
Raised after the argument rather than before it.
What it admits
A dispute the invoice itself caused — usually about work nobody agreed to in writing.
less
Written off as uncollectable
The invoice was correct, the work was done, and the money is not coming.
What it admits
A credit decision made at the point of accepting the client, coming home eighteen months late.
equals
Cash collected
The only figure on this list that pays anybody's salary.
Two of these lines are the same arithmetic and opposite confessions
A fee agreed below standard and a fee written down at invoicing reduce the identical number by the identical amount. One is a price you chose in advance and would choose again; the other is a cost you absorbed because the job went long. A firm that posts both to one account has one number where it needs two, and can never answer the only useful question — which of its recovery problem is pricing and which is delivery.
The two lines to look at are the second and the fourth, because they are the ones the whole article turns on.
A fee agreed below standard is a pricing decision. It was made in advance, deliberately, by someone with the authority to make it, in exchange for something — winning the work, a multi-year commitment, a client the firm wanted on its list. Nothing went wrong. It may have been a bad decision, but it was a decision.
A fee written down at invoicing is a delivery failure. The hours happened, they were real, and either the job took longer than it should have or it was priced on an assumption that turned out to be false. Somebody absorbed the difference on the way out of the door, usually alone, usually in the last twenty minutes before the invoice run.
Both reduce standard value by the same amount and both land in the same recovery percentage. A firm that codes them identically has one number where it needs two, and it will spend years arguing about whether its problem is pricing or delivery with no way to settle it.
The write-off with no reason code
Here is the specific, expensive habit.
At the point of invoicing, somebody reduces a line. There is a reason — there always is — and the reason lives in that person's head for about a day. What reaches the system is a smaller number.
Nothing is wrong with the invoice. The client is charged fairly. The accounts are correct. But the firm has just spent real money and learned nothing, because a write-off without a reason is indistinguishable from every other write-off, and the whole value of the figure was in the distinction.
The reason codes worth having are short and specific, and each of them points at a different fix:
| What the write-down actually was | What it tells you | Where the fix is |
|---|---|---|
| The job took more hours than it was worth | Scoping or estimating | The engagement letter and the work plan |
| Work the client never agreed to pay for | Scope discipline | The point the request arrived |
| Rework after an internal review | Quality at the grade doing it | Supervision and resourcing |
| Time recorded on the wrong engagement | Data | The timesheet, not the job |
| A junior doing a job at a senior's rate | Grade mix | Resourcing, decided at planning |
| Goodwill, granted deliberately | Nothing is broken | Nowhere — record it and move on |
Notice the last row. A knowing concession is a legitimate commercial act and should not be dressed up as a failure. But it has to be named as a concession, or it disappears into the same bucket as the rework and makes the rework look smaller than it is.
Notice the fourth row too. It is the one nobody expects, and in firms whose timesheets are reconstructed weekly it is a meaningful share of the total — a write-off caused by an entry that was on the wrong engagement in the first place. That is not a delivery problem and no amount of supervision will fix it. It is the timesheet problem arriving disguised as something else.
Four ways to cut it, and only one of them changes anything
Realisation can be reported by client, by partner, by service line and by staff grade. All four are easy to produce and they do not do the same work.
By client is the one that gets attention and the one that mostly produces an argument. It confounds too many things at once — a difficult client, a badly scoped engagement and a junior team all show up in the same low number, and the partner who owns the relationship will defend it with reasons that are individually true.
By partner is the one nobody wants to publish and the one that moves the number. Not because it identifies who is bad at billing, but because it is the only cut where the person who could act on it is the person being shown it. A partner looking at his own recovery against his colleagues' will investigate. A partner looking at the firm's average will nod.
By service line is the strategic one and the slowest to act on. It answers whether a whole category of work has become unprofitable — which is usually a pricing conversation for next year rather than something you can fix this quarter.
By staff grade is the diagnostic one, and it is the one that catches the resourcing mistake. Work delivered at a grade above the one it was priced at will show a poor recovery with no delivery failure anywhere in it. The job was done well. It was done by the wrong person, and that decision was made in a resourcing meeting months earlier.
If a firm can only maintain one of the four, it should be by partner, and it should be visible to the partners. Every other cut is a report. That one is a mirror.
What to make a vendor show you
On a live system, on real engagements, not on slides.
- Show standard value, billed fee and cash collected for one engagement, on one screen, with the deductions between them named.
- Record a write-down at invoicing with a mandatory reason code, and refuse the invoice without one.
- Show discounts agreed at the point of pricing separately from write-downs made at invoicing — in the same report, never netted.
- Show realisation by partner, by client, by service line and by staff grade, from the same underlying data.
- Show what a credit note raised after the invoice does to the engagement's recovery figure.
- Reconcile the billed fee on that screen to the revenue in the accounts.
- Show last year's realisation for a client beside this year's fee proposal, on the screen where the fee is set.
Item 3 is the one that decides it. Almost every practice system will produce a realisation percentage, and almost all of them arrive at it by netting everything above the invoice into one deduction. A system that keeps the pricing decision and the delivery failure apart all the way to the report is answering the only question that leads anywhere, and it is a design choice made in the data model rather than a filter added to a report.
Item 7 is the quiet one. Realisation reported after the year is history. Realisation shown at the moment next year's fee is being set is the only version of the number that has ever changed a price.
The short version
Standard value is your own opinion. Cash is the fact. Between them sits a bridge of named reductions, and the name is worth more than the number.
Two of those reductions are arithmetically identical and mean opposite things. A discount agreed before the work is a price the firm chose. A write-down made at invoicing is a cost the firm absorbed because something in the delivery did not go as planned. Coded the same way, they cancel into a single percentage that cannot tell anybody what to do.
A write-off without a reason code is money spent to learn nothing. Six codes are enough, they take a second each, and they are the difference between a recovery report and a recovery decision.
And publish it by partner. The other three cuts inform; that one is the only one where the reader can act.
The value that never reaches the bridge at all is unbilled work in progress; the work that quietly enlarged the job is scope creep; and what all of it means client by client is client profitability. If the numbers underneath are reconstructed on a Friday, start with the timesheet.
Why a well-built report can still mislead is why your ERP dashboard is lying to you. What this looks like as one system is the professional and business services page and ServX.