Every time I speak at a conference, every time I sit down with a new client’s leadership team, the same question comes up: “If we could only track one metric, which one should it be?” My answer surprises people. It isn’t pipeline velocity. It isn’t portfolio ROI. It’s Cost of Failure.

The question we get every time

The reaction in the room is usually the same. A pause. A few raised eyebrows. Then someone says, “That sounds like the metric you don’t want to put in front of the board.” Which is, of course, exactly the reason it’s the metric you most need to.

Cost of Failure is the total capital your company spent on innovation bets that didn’t reach the next stage. The projects that got shut down, pivoted, or quietly forgotten. The line item nobody puts in the annual report.

Most companies don’t even know what it is for them. That’s the first problem. The second is that, when they do find out, they almost always try to reduce it — which is exactly the wrong response. A healthy innovation portfolio should have a meaningful Cost of Failure number. The work is in making sure you’re failing on the right things and learning from each one.

Why traditional KPIs fail innovation

The KPIs most companies use to measure innovation were built for a different job. ROI, NPV, payback period — these are exploitation metrics. They work brilliantly when you’re measuring the performance of an existing business model. They fall apart the moment you try to measure something that doesn’t have revenue yet.

The classic failure mode is asking a Stage-1 idea to forecast a 5-year NPV. Either the team makes the number up to clear the gate, or the idea gets killed for failing to meet a threshold that nobody — not Bezos, not Musk, not anyone — could have hit with that little evidence.

What Innovation Accounting tries to do is give you metrics that work before revenue lands. Cost of Failure is one of those. It doesn’t require a working business model. It just requires honest record-keeping.

What Cost of Failure actually measures

The temptation is to read Cost of Failure as “money wasted.” It isn’t. It’s the cost of buying information.

Every innovation bet that doesn’t advance has produced something: evidence that this particular hypothesis doesn’t work. That evidence has a value. The question Cost of Failure helps you ask is whether the price you paid for it was reasonable.

A bet that was killed at Stage 1 for €10k after two weeks of customer interviews has produced cheap evidence. A bet that ran for three years and consumed €8M before being shut down has produced expensive evidence — and the team almost certainly knew it wasn’t working at the €500k mark.

The healthy metric isn’t low Cost of Failure. It’s low cost per learning: a portfolio in which bets that should fail, fail early and cheaply.

How to calculate it

The mechanics are straightforward. For every project that exits the portfolio without progressing, sum:

  • Direct spend — cash paid to vendors, contractors, tooling, infrastructure
  • People cost — the loaded cost of the team time spent (salaries plus overhead)
  • Management overhead — review meetings, governance time, executive attention. This one is almost always ignored, and it’s almost always the biggest number.

That gives you Cost of Failure per project. Roll up across the portfolio for the period and you have your number.

What makes the metric useful is what comes next: segmenting it. Cost of Failure by stage gate (how much are you spending on bets that fail at Stage 1 vs. Stage 3?). Cost of Failure by business unit. Cost of Failure by strategic option. The aggregate number tells you whether you have a problem; the segments tell you where the problem lives.

Like this kind of thinking?

One essay like this in your inbox, roughly monthly.

New thinking on growth systems, innovation accounting, AI, and how the best leaders run corporate innovation today. No spam.

Why it changes leadership behaviour

The shift Cost of Failure produces in leadership conversations is the reason I push for it so hard.

When the metric goes up on the board pack for the first time, the immediate reaction is usually defensive: “That number is too high.” The right next question, the one that takes leaders six to twelve months to get good at asking, is: “Compared to what? Are we failing on the right things? Are we failing fast enough?”

Once that conversation becomes routine, the company has crossed a real threshold. Failure stops being a thing teams hide and starts being a thing they report on. The CFO stops treating innovation as off-balance-sheet noise and starts treating it as a venture portfolio. The board stops asking “why does this lab keep losing money” and starts asking “is the cost-per-learning trending in the right direction.”

Healthy innovation portfolios should have a meaningful Cost of Failure number. The work is in making sure you’re failing on the right things, and learning from each one.

Three common mistakes

Three patterns I see when companies first introduce the metric:

  1. Treating it as a stick. If teams know their Cost of Failure will be used to grade them at year-end, they’ll either avoid risk or hide their failures. Both outcomes wreck the portfolio. The metric has to be treated as a system-level signal, not a personal score.
  2. Trying to drive it to zero. A zero Cost of Failure portfolio is a portfolio that’s only investing in bets it already knows will work. That isn’t innovation; it’s sustaining-investment. If the number is too low, the portfolio is too conservative.
  3. Forgetting the management overhead. A €200k team-level failure that consumed 40 hours of executive committee attention has a real cost much higher than €200k. Innovation governance has a real price, and pretending otherwise distorts the picture.

One final thing

Cost of Failure is one of the metrics we built directly into SATORI. We did that because the calculation matters, but more because the cadence matters. A metric you compute once a year is a metric that doesn’t change anyone’s behaviour. A metric on the dashboard every quarter, broken down by strategic option, is a metric your leadership team will start treating as theirs.

The reason it’s the most important Innovation Accounting KPI isn’t that it’s the smartest measurement. It’s that it’s the metric that forces the right conversation. And the right conversation is usually what was missing in the first place.