Nobody Signed Off That £1.2 Million

One and a quarter million pounds. Call it £1.2 million and round down,

because the precise figure doesn’t matter and I’ve changed it anyway.

FM Talk | FM Contract Advisory |

That’s the annual subsidy on a workplace catering and hospitality contract I looked at recently. A global financial institution. Serious people. Serious governance. A procurement function that could take a stationery agreement apart at the seams and hand you back the pieces in order of value.

Here’s the thing I keep coming back to.

Nobody there ever approved that £1.2 million. Not once. There’s no minute, no paper, no signature. Nobody sat in a room, looked at that number, and decided it was a reasonable price to pay for feeding a building.

It just happened. A bit at a time, over fifteen years, in the gaps between decisions.

That isn’t a catering failure. I want to be careful here, because the reflex in our industry is to reach for the supplier and the reflex is wrong. The caterer has delivered faithfully, competently, profitably exactly what the contract asks them to deliver. They have not overcharged. They have not underperformed. They have done the job as specified.

The specification is fifteen years old.

Five things that should have opened the file

In the period since that contract was last substantively reviewed, five things happened. Each of them, on its own, is a reason to take the document off the shelf.

The estate changed shape. Different building, different floorplate, different circulation, different footfall. The resourcing model was built on assumptions about how people move through a space that no longer exists.

Significant capital went into the facilities. New equipment, new counters, a better offer. What didn’t change was the labour model, the opening hours or the service points that the investment was supposed to make more efficient. You can spend seven figures improving a kitchen and change nothing at all about what it costs to run.

The supplier was acquired. The contract is now held and interpreted by an organisation that never sat at the original negotiating table. Nobody at either end of that relationship was in the room when the terms were agreed.

Hybrid working arrived. Covers fell and never came back. The resourcing model, priced against pre-2020 headcount, did not move.

Indexation carried on. Fifteen years of annual uplift, applied compound, to a base that was correct in a different decade.

Five triggers. Five opportunities. The file stayed shut.

Why nobody noticed

This is the part that actually interests me, because the answer isn’t incompetence and it isn’t inattention. These are capable people.

The answer is that the subsidy sits in a cost centre, not in a contract.

It gets reported monthly, as a variance against budget. And a variance report is a beautifully effective way of making a large number invisible. Nobody is ever asked to approve £1.2 million. They’re asked to note that catering came in 3% over budget, again, and the budget was set by taking last year’s number and adding a bit.

That’s not governance. That’s arithmetic with a covering note.

Fifteen years of it, and the number nobody approved becomes the number nobody can explain. Ask most FM or finance teams to derive their catering subsidy from first principles covers, cost per head, labour model, hours, waste, the lot without reference to last year’s subsidy, and watch what happens. It’s not a gotcha. Almost nobody can do it. The number has become self-referential.

The pitfalls, named

If you’re reading this thinking that’s not us, good. Check it against these anyway. I see the same eight traps in almost every long-tenure service contract I’m asked to look at.

1. Benchmarking mistaken for review. You compared your rate to the market. Fine but the market has also stopped thinking. Benchmarking tells you whether you’re paying a normal price. It tells you nothing about whether you’re buying the right thing. A well-benchmarked contract for a service you no longer need is still a waste; it’s just a competitively priced one.

2. Risk transfer nobody registered. Since Covid, workplace catering has moved decisively to cost-plus and management fee models. The contractor operates the service at cost and takes a fee. Which means volume risk the empty Friday, the half-term dip, the team that’s gone permanently to three days sits with you. Entirely. Ask yourself who bears the cost of a quiet building in your contract. If the honest answer is “we do”, then your subsidy is a variable you own and are not managing.

3. Capital investment without model reset. New kit doesn’t reduce cost by itself. It reduces cost when you change the labour model, the hours or the service points around it. Otherwise you’ve bought a nicer version of the same expense.

4. Relocation without re-specification. Footfall in a new building bears no relation to footfall in the old one. The resourcing model does not transfer. It gets carried across because moving is chaotic and nobody wants to renegotiate the catering in the middle of a move.

5. Change of control treated as administration. Your supplier was acquired. Somewhere in the file there is a change-of-control clause. In most cases it generated a letter, a novation and a new set of logos on the invoice. It did not generate a renegotiation. Given how much consolidation this market has seen in the last three years, a lot of readers are one acquisition away from being in exactly this position — or already in it and unaware.

6. Indexation on autopilot. An annual uplift applied without argument is a rounding error in year one and a structural problem in year twelve.

7. No performance framework worth the name. Most catering KPIs measure whether food was served. Very few measure whether the subsidy bought anything participation, retention, the actual reason the organisation decided to feed people in the first place.

8. Evergreen by default. Rolling extension because a proper review is a project, projects need an owner, and everyone is busy. This is the one that does the most damage, because it’s the one that lets all the others compound.

Why going to market is usually the wrong first move

Here’s where I’ll lose some people.

The standard advice, when a subsidy looks wrong, is to re-tender. It feels decisive. It generates competitive tension. It produces a number you can take to a board.

It’s also, most of the time, the wrong first move because if you go to market with the specification you already have, you will re-buy the same service from a cheaper supplier, book the saving, and be back in this position in three years with a supplier who has less margin to absorb the next surprise.

A tender prices a specification. It does not interrogate one. And in the case I’ve described, the specification is the problem. Not the price. The bank could have run a flawless competitive process at any point in the last decade and still ended up with a £1.2 million subsidy, because every bidder would have been pricing a service designed for a building the organisation no longer occupied.

Fix the specification. Then decide whether you need to go to market at all. Quite often you don’t a properly reset operating model and a renegotiated commercial mechanism will recover more than a tender would, faster, and without the mobilisation cost and service disruption that comes with changing supplier.

What a real review looks like

Not a benchmark. Not a procurement exercise. Roughly this:

Start with demand, not cost. How many people are actually in the building, on which days, doing what. Real data, not the headcount on the HR system.

Then the operating model. Hours, service points, labour, offer rebuilt from that demand, not inherited from the last version.

Then the commercial mechanism. Who carries volume risk, how the fee works, what indexation applies to and what it doesn’t.

Then performance. What is the subsidy for? Write it down. If nobody can answer, that’s your finding.

Only then, if it’s warranted, the market.

You need finance, FM, HR and someone from the business in the room. Not because it’s a committee, but because a catering subsidy is a people decision that finance pays for and FM administers, and if those three functions don’t agree what it’s for, it will drift again.

The reframe

A catering contract isn’t a procurement asset that gets refreshed on a cycle. It’s an operational relationship with a very long memory, sitting inside an organisation that changes constantly, and it needs somebody minding it.

Three questions, and you can answer them this week:

When was your contract last opened, as opposed to renewed?

Can you explain your subsidy from first principles, without reference to last year’s?

If your supplier were acquired tomorrow, what would actually happen?

If the third answer is “we’d get a letter”, you already know what the first two are.