Same number, different conversation:

why the source of your FM data matters more than the figure

By Ernie Melling, Founder & Consulting Partner, EMC Associates

A single figure can mean one thing in the boardroom, another in procurement and something else entirely on site. When contract and procurement decisions rest on numbers nobody has traced back to their source, the risk is not just a poor decision. It is a confident one

I came across a simple graphic recently that has stayed with me. It showed one business target, a cost of goods figure of 19%, travelling down through an organisation. In the business plan it was a percentage. In the team plan it became a question: where is waste happening in what we do every day? In the individual plan it became a commitment: here is the one I am going to fix this quarter. The line underneath read that translation is the work of leaders at every level.

It is a good point, well made. But it made me think about the reverse problem, which I see far more often in facilities management. The number travels through the organisation, and at each level it is not translated. It is reinterpreted. The figure stays the same while its meaning quietly changes, until people are making decisions about entirely different things while believing they are all looking at the same evidence.

One number, four readings

Take a figure that will be familiar to many FM Talk readers: catering food cost at 19% of revenue. To a finance director it is a margin indicator. To a procurement lead it is a contract benchmark, perhaps a target written into the specification. To the contract manager it is a monthly KPI on the supplier’s report. To the catering manager on site it is a daily pressure on portion sizes, menu choices and waste.

None of them is wrong. But each is answering a different question, and very few will have asked what sits inside the 19%. Is it food cost against gross or net revenue? Does it include subsidised meals, hospitality, free staff meals or waste? Is it the supplier’s purchase price or the price after rebates? Is the revenue figure the till total, or the till total adjusted for vouchers and transfers?

The same pattern runs through almost every FM measure.

Why the source matters

Every number has an author, a method and a motive. In outsourced FM, much of the performance data a client relies on is produced by the supplier, under a payment mechanism the supplier is incentivised to satisfy. That does not mean the data is wrong. Most providers report honestly within the rules they have been given. But the rules themselves are often loose, the definitions are rarely tested, and the client side frequently lacks the time or capability to check.

Experience has taught me that the most common problem is not falsified data. It is unexamined data. A PPM compliance figure is calculated correctly against a schedule that omits a third of the assets. A helpdesk KPI is met because jobs are closed when a technician attends, not when the fault is fixed. A cost per square metre is compared with a peer organisation that measures area differently and excludes half the services. Every figure is accurate. The conclusion drawn from them is not.

When these numbers feed contract and procurement decisions, the consequences are real. A client renews a contract on the strength of green KPIs while the asset base deteriorates. A retender is specified around a baseline that understates true spend by 15 or 20%, so the new contract looks like a saving until variations restore the cost. Performance deductions are never applied because the measurement regime cannot support them. A board approves an insourcing or outsourcing decision based on a comparison that was never like for like.

Where off the shelf benchmarks fall short

Benchmarking is often presented as the answer to this problem. Compare your figures with the market and you will know where you stand. The difficulty is that most published benchmark data inherits exactly the same weaknesses, multiplied across many organisations.

An average cost per square metre drawn from dozens of contracts tells you very little unless you know how each contributor defined its area, its scope, its service levels and its exclusions. A hospital, a further education college and a city centre office are not comparable on a single rate. Even within one sector, two NHS trusts can have entirely different scopes hidden behind the same service heading. Off the shelf benchmarks can be a useful sense check. They are a poor foundation for a decision worth millions.

What forensic benchmarking does differently

Forensic benchmarking starts from the other end. Before any comparison with the market, it asks whether the client’s own numbers can be trusted.

In practice that means reconciling what the contract says with what was invoiced, and what was invoiced with what was delivered. It means tracing performance figures back to the underlying job records, schedules and asset data. It means separating core contract cost from variations, projects and pass-through, so that the true cost of the service is visible. Only once the client’s own baseline is clean, consistent and properly defined does an external comparison become meaningful, and then only against comparators that have been normalised for scope, specification and context.

The evidence tells us that this first step often changes the conversation entirely. The question shifts from whether the contract is good value against the market to what the client is actually buying, what it is actually paying, and what it is actually getting. Those are the questions that support good contract and procurement decisions.

At EMC we approach this through a simple discipline we call DICE: Discover what the data actually says and where it came from; Interpret it in the context of the client’s estate and objectives; Challenge the assumptions on all sides, including our own; and only then Execute. It is deliberately unglamorous. Its value lies in the order.

Translation is still the leader’s job

This brings me back to the graphic. Its central point holds: a strategic number is useless until people at every level can see what it means for their own work. In FM, though, translation only works if the number being translated is sound. Cascade a flawed figure through the organisation and it does not become actionable. It becomes a shared misunderstanding, with everyone working hard on the wrong problem.

So the leader’s task is twofold. First, establish that the number is right: its source, its definition, its completeness. Second, translate it so that the board, procurement, the contract manager, the supplier and the site team are all having the same conversation. A target to reduce cleaning cost should become a clear understanding of where cost sits in the specification, which frequencies deliver value and which do not, and what the site team will do differently this quarter.

One thing I’ve noticed is that organisations which get this right tend to share a habit. Whenever a number is presented, someone asks five simple questions:

1. Where did this figure come from, and who produced it?
2. How is it defined, and what is included or excluded?
3. Is it complete, and does it reconcile with other sources?
4. What is it being compared with, and is the comparison like for like?
5. What decision is it being used to support?

None of these require specialist expertise. They require the discipline to ask before acting, and the willingness to accept that a familiar number may not mean what everyone assumed. That is where better decisions begin: from evidence to better outcomes.

Questions for reflection

1. Which single figure most influences decisions about your FM contracts, and when did anyone last trace it back to its source?
2. If your board, procurement team and contract manager were each asked what your key FM metric means, would they give the same answer?
3. Is your view of value for money built on your own reconciled contract data, or on an external average whose definitions you have never seen?

From evidence to better outcomes.