Where margin actually goes

Margin rarely collapses in a dramatic blaze of missed deadlines, furious clients and emergency meetings. It erodes quietly, while everything still looks under control.

The client is satisfied, the team is busy, the status report is green, and nobody has raised a dispute. Beneath that surface the commercial position may already be weakening. Extra time gets absorbed, scope expands without a reset, delivery effort increases, and the billing position does not move with it.

The misdiagnosis

When the final number disappoints, the question asked is almost always why delivery lost margin. Delivery ran the work, delivery consumed the hours, so delivery must be where the money went.

That question presumes there was margin to lose. Often the more accurate question is whether the margin was ever properly protected — whether the assumptions in the quote survived contact with the work, whether the discount conceded to close the deal reached the billing rules, whether the effort model reflected how the team actually delivers.

The distinction matters because it changes where you spend the remediation effort. If margin was lost in delivery, you tighten delivery. If margin was never protected, tightening delivery produces a tired team and the same result.

The four places it actually goes

In practice, margin leaves through a small number of doors, and each of them is upstream of the invoice.

Pricing assumptions that were never tested. An effort estimate built on the best version of the team, a rate agreed verbally, a discount described as harmless. None of these is visible as a loss; each reduces the ceiling before work begins.

Handover. What was sold and what delivery believes was sold are two documents, and the gap between them is absorbed as delivery effort rather than raised as a variation.

Scope that grew politely. Small additions nobody wanted to make a fuss about, delivered as goodwill, recorded as nothing. Individually trivial, cumulatively the largest of the four in most firms.

Time capture and approvals. Hours entered late, entered against the wrong engagement, or approved without reference to whether the contract permits billing them. Every one of these makes the WIP position less defensible, and an indefensible WIP position gets conceded rather than argued.

None of those events announces itself. That is precisely why the apparently healthy project is the one to fear, rather than the one everybody already knows is in trouble.

Why month-end cannot recover it

Margin is not protected at month end. By then you are explaining history.

By the time a margin problem appears in reporting, it has usually been leaking for weeks through pricing, handover, change control, time capture and approvals. The close can identify the loss and it can allocate it. It cannot reverse it, because every decision that produced it has already been made and, in most cases, already been communicated to a client who now considers it agreed.

This is also why margin dashboards disappoint. They are accurate and they are late, and accuracy about a position you can no longer change is reporting, not control.

The mechanism

Making margin visible while it can still be acted on means carrying the commercial rules forward with the work rather than reconstructing them at the end.

The engagement holds its pricing basis, rate card and billing terms from the point it is quoted, so the delivery structure inherits them rather than interpreting them. Change is a priced, approved variation attached to the engagement, not a note in a meeting minute — which means unapproved work cannot quietly enter WIP and then require a decision about who absorbs it. Time and cost entries inherit the terms that decide whether they are billable. Approval is a state on the record. Each WIP line traces back to work, approval and contract clause.

The consequence is narrow but valuable: the gap between what was sold, what is being delivered and what can be billed becomes readable during the engagement rather than after it. Whether that gap is visible at all is the difference set out in WIP: register or argument, and the earliest of the four doors is covered on our scoping and quoting page.

Where the effort side of the position is formed — the assignment, the actual shape of the team against the plan — sits on our delivery and time capture page.

What this costs, and where it does not apply

Governing the chain means agreeing pricing rules, approval gates and change control before delivery starts, and then holding to them when a client asks for something small and the commercially comfortable answer is yes. That is a genuine cost, and it is paid in relationship capital as well as time. Some firms will decide the flexibility is worth more than the leakage, and for a small number of them that will be correct.

It does not apply evenly either. A firm running mostly fixed-scope work of short duration, with one billing model and a single approver, will find the ceremony outweighs the recovery. The firms this changes are the ones with concurrent engagements, mixed billing models and change as a normal event rather than an exception.

And it is worth stating what this does not do. It does not improve a price that was wrong when it was set, and it will not make a badly estimated engagement profitable. What it does is make the position legible early enough that someone can choose — reset the scope, raise the variation, or knowingly absorb the cost. Choosing to absorb it is a perfectly good outcome. Discovering you absorbed it four months ago is not.

Green should mean controlled

The status report that says green is making a claim about delivery. It is usually silent about the commercial position, and in most firms nothing is checking whether those two agree.

A useful test costs nothing. Take three engagements currently reporting green and ask, for each, what has changed since the quote and where that change is recorded. If the answer for any of them is a conversation, the margin position on that job is an estimate — and green means busy rather than controlled.