Once a year your business runs a pay review. Everyone goes in at the same time, and the outcome is presented as fairness.

It is not a fairness exercise. It is the largest discretionary allocation your business makes all year, and it is being made by calendar rather than by strategy.

Here is what the timing does. When every role is considered at once, the money spreads. Nobody wants to explain why one team moved and another did not, so the differences flatten until the outcome is defensible to everyone in the room. Defensible is not the same as correct.

The result is predictable. Roles where supply is loose get more than the market required. Roles where supply is genuinely broken get roughly the same, which is nowhere near enough to hold them. You have paid a premium. Just not for scarcity.

The reason this survives is that a pay review answers a question about process, and retention answers a question about risk. They are different questions, and they have been put on the same calendar.

So separate them. Keep the annual cycle for the obligations it exists to meet under your awards and agreements. Then run a second, quieter conversation, on its own timing, about the small number of roles the business genuinely could not refill. Different question, different owner, different line in the budget.

Budget season is where this gets settled for the year ahead. It is worth settling on purpose rather than by default.

The question worth taking into your next leadership meeting: which roles could we not refill this year, and does our pay position reflect that, or does it only reflect the calendar?

The data lives in the server.

The wisdom lives in your head.

You hold more information about this business than anyone else in the building. Every ledger. Every code. Every reconciliation.

And you still carry a quiet worry into the leadership meeting.

That the things you cannot see are the things that will cost you.

That worry is not paranoia. It is accurate.

Here is the belief that keeps it in place.

If it is not in the management accounts, it is not material.

Almost every CFO believes some version of this, and for good reason. The accounts exist to surface what matters. A function that managed every cost by transaction would drown.

Raising something that is not in the accounts also feels like going below the line.

The accounts were built to make cost legible at the leadership level. Never at the level where it actually leaves.

Three features of a standard set of accounts do the hiding.

The roll-up. Operations. Administration. Other. These began as summaries and became destinations. Once a cost sits in Other, it has left the conversation, and nobody decided to stop managing it.

The missing trend. Your accounts show the period against budget and against last year, rarely against several cycles ago. A line that creeps up steadily, resetting against a fresh budget each year, reads as on plan while it drifts somewhere nobody has named.

The unmeasured. Purchasing card spend. Departmental buying under the approval threshold. Auto-renewing software. Invoices coded to a general expense line. Invisible not because they are absent, but because the framework was never built to see them.

The pattern I see in the engagements I have worked on is simple. The accounts show what was easy to categorise. The leakage lives in what was not.

The CFOs who find the money are the ones who treat the management accounts as a starting point, not as the definition of what is material.

The move is not a redesign. It is a parallel view. A short, separate read of what the accounts cannot hold. Card spend. Untendered recurring services. The detail sitting under Other.

Walk each one and ask one question. Is this being managed, or only being paid.

You already carry the wisdom the server does not hold. Your cost base deserves the same treatment.