What Deloitte Spain's Innovation Team Shows Its Board Instead of ROI

Every innovation leader has been asked the question mid-pilot: is this worth it, and when do the returns show up?

Since there’s often no revenue yet, the team reports what it has available: workshops delivered, pilots launched, projects killed. Leadership stays unconvinced — and when the budget gets cut, priorities get reshuffled, innovation gets pushed into unrealistic claims, and often the program just gets shut down.

Hismael Doval, innovation program lead at Deloitte Spain, has seen this many times. That's why the internal incubation program he leads is built to make disciplined calls in the messy middle, the stage where an idea looks promising but has no revenue yet — and to prove those calls to leadership.

The program treats early innovation as a decision system: every idea starts as a set of assumptions, and each one gets a small, cheap test. The result triggers one of four calls: kill, extend, pilot, or scale. The pass mark is set before the test runs, so evidence only decides the call.

The program earned this arrangement over time. For its first twelve years it reported revenue the way finance wanted, and proved a 3.6x return per euro invested. That track record bought it the credibility to change the terms three years ago: revenue came off the top of the report, and evidence of learning and disciplined decisions took its place. Leadership agreed to that — inside a firm that still reviews its financial results every two weeks.

The same four-part system is one any innovation leader can run: a decision loop that turns assumptions into calls, five desirability checks that stand in for revenue, pass marks set before the results land, and a board update that reports decisions instead of activity.

An activity report shows a team is busy, but does not prove whether uncertainty is shrinking or whether the money is being spent in a sensible order. When leadership cannot see how the money is controlled, it assumes the program is not managed well. And then, the budget gets cut.

The alternative to that report, is a decision loop: four stages that repeat — hypothesis, experiment, evidence, decision. An initiative begins as a set of assumptions that all have to be true for it to work. Each assumption gets its own small and cheap test, designed to give a clear signal about whether the assumption holds. That signal has to be something observable: how clients behave, whether they will pay, how they use the product. Internal agreement does not count. The signal then points to one of four calls: kill, extend, pilot, or scale. Each call is made against a threshold the team set before the test ran.

Leadership recognizes this pattern: put small money in to buy evidence, put larger money in only once the evidence lowers the risk. It is how they already allocate capital. And because every call is tied to the evidence that drove it, the whole program can be audited before it starts earning.

Deloitte Spain's innovation program runs five groups of desirability indicators:

Together the five turn "is this worth it?" into a question the team can answer with evidence.

Every idea gets scored against a set of criteria before it earns more funding. When a team sets those pass marks after the results come in, people just argue the score up or down. This is a core issue in many innovation teams. A mark set too late gets “bent" to fit the answer people already want, and the winner is usually the most senior person.

The fix is to set each criterion's passing mark up front, before a single result exists. Once the marks are locked, an idea's results either meet them or fall short, and that outcome decides whether it gets funded. The decision then holds, even when a senior sponsor wants to fund an idea the evidence rejected.

For Deloitte Spain, that means using a problem solution fit scorecard. The team uses the scorecard below after building a working version of the solution, to see whether clients value it.

The scores give the team a first answer. Then the coaches, team, and sponsor meet to discuss it and make the call. Since the pass marks were set before anyone saw the results, the discussion is solely about what the evidence means. No one can move the bar to save a project they like.

The board update is built straight from the loop the team already runs: each entry is a hypothesis that got tested, the evidence it produced, and the decision that followed. Nothing extra gets invented for the board. Practically, it looks like an investment memo with five lines: what the team tested, what the evidence showed, what changed, what was decided, and where the money went.

What makes the update stick with the board is how it handles bad news: a test that kills an assumption is reported as a good result, because it stopped a weak bet before it cost real money. This is what finance almost never hears from an innovation team — and it is what makes the update sound like careful investing.

This did not happen on day one. The incubation program has run for twelve years; only the last three have reported this way. For its first years the program played by finance's rules and reported revenue, proving a 3.6x return on every euro invested. That track record is what earned it the credit to change the reporting terms. Three years ago it did: revenue came off the top of the report, and learning and evidence-based decisions took its place — inside a company that still reviews its financial results every two weeks.

The switch was not welcomed at first. What changed leadership's mind was sitting in the decision meetings a few times and watching each call trace back to the evidence behind it. The team can now walk into a budget review without a revenue number and keep its funding, because the evidence does the work ROI used to.