There is a scene that repeats itself in every Portuguese factory we visit. The COO walks into the meeting with a KPI that proves production is running perfectly. The CFO has another figure that says the operating result is stagnant. Nobody is lying. Both are looking at real data. And neither of them can explain why the factory that "produces well" doesn't make money.

The truth is more uncomfortable: when the CFO and the COO use metrics that don't talk to each other, they are building two parallel realities. One sees speed. The other sees cost. One celebrates output. The other laments that the output doesn't convert into margin. And while they argue over which is the "real" KPI, the company optimises itself for neither world.

We argue that the problem is not having too many KPIs. It's having KPIs that don't share the same language. And that failure to communicate costs money — every single month.

The Illusion of the Solitary KPI

The COO loves OEE. It's tangible, visible, immediate. A machine running 16 hours a day at 85% efficiency is a fact you can measure on the screen. The CFO looks at OEE and asks a question nobody likes: "And how much does that 85% cost us?"

Why does it cost? Because an OEE of 85% can mean many things. It could be 85% of a line producing parts with 30% material waste. It could be a machine that stopped 6 times a day, but each stoppage was 5 minutes long — the OEE "recovered" because the line ran fast the rest of the time. It could be a production run that met the plan, but of a product that will sell at a 15% discount because the customer rejected it in a quality audit.

The COO is right: OEE matters. But the CFO is also right: OEE on its own doesn't tell you whether the factory is making money. And this is where most Portuguese industrial companies get stuck. They optimise for the metric they can measure quickly, and forget to ask whether that optimisation is serving the business.

A KPI that doesn't connect to the financial result is just a pretty number. Full stop.

Why the CFO and the COO Speak Different Languages

The COO thinks in terms of speed, volume, continuity. Their pressure is daily: "Did I meet the production plan?" The CFO thinks in terms of margin, liquidity, return on assets. Their pressure is monthly or quarterly: "How much profit did I make from what I produced?" These two perspectives are not enemies. They are complementary. But they require a bridge — and that bridge has a name: data analysis that links operations to results.

Let's take a concrete example. A garment manufacturer in the Vale do Ave with 120 employees. The COO says: "We have 94% labour utilisation — the seamstresses spend 94% of their time in production." The CFO says: "The labour cost per unit produced rose 7% compared to last month, despite volume growing 3%." Who is right?

Both. What's missing is the next question: "If utilisation went up and volume grew, why did the cost per unit increase?" The answer may lie in three places that no one sees without integrated data. First: a change in the product mix — we started making more complex parts, which take longer per unit, and OEE doesn't distinguish between a simple part and a difficult one. Second: an increase in absenteeism or turnover — new, less productive seamstresses are inflating the total hours. Third: unaccounted rework — parts that go back to the machine because they're defective, which the production stopwatch doesn't separate from "good" production.

None of these answers appears if the CFO only sees aggregate figures and the COO only sees the production stopwatch. They need the same photograph — taken from different angles.

The Pattern of the Companies That Get This Right

There is a common temptation: to create a single KPI that "solves everything" and that the CFO and COO both agree on. Usually it's something like "Operating result per production hour" or "Gross margin per OEE achieved." It sounds elegant. It rarely works, because it forces both to abandon the metrics they genuinely need to see.

The pattern we see work is different. The companies that achieve alignment don't eliminate the COO's or the CFO's KPIs. They create a layer above — a BI dashboard that simultaneously shows what the COO needs (OEE, stoppage rate, material waste, plan fulfilment), what the CFO needs (production cost per unit, margin per product, stock turnover), and — this is the essential part — the link between the two. When the CFO clicks on a cost KPI, they can see which machine stopped, which product it was, what time of day it happened. When the COO can see that the product they think is "easy to produce" is the one that leaves the least margin.

When both see the same dashboard — and understand how the numbers on one side affect the other — the discussion changes. It stops being "My KPI is better than yours" and becomes "If we do this here, what happens there?"

Integrated Data Is Not a Luxury

Five years ago, we believed alignment came from training. That if we explained the business properly to the COO and operations properly to the CFO, they would naturally converge. We were wrong. What really changes things is when both can see, in real time, the translation of their numbers into each other's language.

This only happens with integrated data. Having an ERP that stores everything is not enough. The shop-floor information — minute by minute — must talk to the financial information — also in real time. Production data captured by a MES system must reach the BI dashboard the same day. Material cost, labour cost and overhead cost must be allocated to the specific product that was produced, not to an aggregate average.

According to 2025 data from INE, only 53.7% of companies in Portugal were using an ERP — and many of those that do still have parallel systems of Excel, paper and informal notes. Without an integrated core, BI works on scattered data. And scattered data creates parallel realities. Five years ago, we thought this was a "nice to have". Today we know it's the difference between a factory that optimises for the wrong metric and one that optimises to make money.

The CFO and the COO don't need to agree. They need to see the same reality.

What to Do in the Next Meeting

If you're a CEO and you see that the CFO and the COO disagree about how the factory is doing, don't ask IT to "create a KPI that pleases everyone." Ask for this in three steps.

First: list the KPIs each one uses today. Not to criticise them. To see where they don't touch each other. If the COO measures "plan fulfilment" and the CFO measures "margin per product," there's a gap: no one is measuring whether the plan being fulfilled is profitable. If the COO measures "stoppage rate" and the CFO measures "labour cost," there's another gap: no one is translating a machine stoppage into its impact on cost.

Second: ask what data is missing to answer the most important question. "Are we making money on the production we do?" If the answer requires data spread across different systems — production on the shop floor, purchasing in the ERP, results in the financial system — then you have an integration problem. It's not a KPI problem. It's a data architecture problem.

Third: once you have that data integrated in a dashboard both can use, the discussion changes. The CFO stops saying "OEE doesn't matter." The COO stops saying "Cost is the CFO's problem." Both see the same story — told in two different ways. And when they see the same story, decisions stop being political. They become operational.

This isn't sophisticated technology. It's logic. But logic requires data that talks. And data that talks requires a system that brings it into the same place — integrated, up to date, accessible.