"She has handed in her notice. What do we need to match?"

The operations manager at a Melbourne distribution business asked it of the CFO, and the CFO did what most CFOs do. Within the week there was a counter-offer on the table, a retention payment attached, and a higher base in next year's budget for a role that had not changed.

The senior systems analyst who keeps the warehouse software running had asked for nothing. Very few people in the city could do her job, she was paid at the grade she was hired into, and nobody had thought to look.

That is how most mid-market retention budgets get spent: on whoever is holding a resignation letter. It is the retention math done backwards. A counter-offer is paid at the moment of maximum pressure to the person who started looking for reasons that are rarely only money, and it tells every observer that the way to get paid is to threaten to leave, which the quietly scarce people will never do.

Underneath, the labour cost line is running three markets at once. Loose supply, tight supply that moves easily, and a few roles where supply is genuinely broken. A counter-offer treats them as one.

Three months on, the operations manager's replacement was hired at a premium, and the analyst got a recruiter's call.

The fix is a decision made in a calm month. Name the handful of roles the business could not replace inside a year, price each against what replacing it would really cost, and settle their pay on purpose. Leadership will flinch at saying out loud that some roles matter more than others. The flinch is the expensive part.

Settle it before anyone has a letter in hand.

When did anyone last ask what this category would cost if you bought it on your own?

The category manager at a Brisbane building-supplies wholesaler had the buying group's price list open at the renewal meeting. Same list as last year, same supplier, a small uplift he described as "market". "We're in the group," he said. "That's the best price going."

"The P&L is a shadow of the business, not the business."

A group price is the same kind of number: a shadow of a negotiation run for the average member, which is a business nobody actually runs.

His confidence has real grounds. A buying group pools volume from many members and negotiates once, which in plenty of categories beats anything a mid-market buyer could get alone. Choosing the group is also the choice nobody has to defend. If the price later turns out to be ordinary, it was the standard channel, and the standard channel is never blamed.

So nobody is asked to defend the group price, and nobody tests it.

A group negotiates for its average member. The price reflects the blended volume, the supplier's interest in keeping the channel full, and the supplier's pricing across the rest of its book, none of it cut to one buyer's footprint, terms or leverage. For a member sitting close to the average that is fine. For a member whose volume, concentration or discipline sits well above it, the group price leaves money on the table that a tailored negotiation would take, and the group, which earns part of its income from the suppliers it places business with, has no particular reason to go looking for it.

The useful question is narrower than whether to leave the group. For each material category bought through it, what is the gap between the group price and what you could negotiate alone, given your own volume and your own terms? In most categories the gap will be small and the group earns its place. In the few where your profile is plainly better than average, it will be large enough to pay for a direct conversation.

The CFOs who hold the best prices are the ones who treat the group rate as the floor, not the ceiling.

Nobody has ever been blamed for the group price, which is how it came to be nobody's job to check it.