On time. On budget. Oh dear.

Let us begin with a small thought experiment.

Imagine a project that delivers on time, on budget, and entirely to specification. The launch is smooth. The stakeholder presentation goes well. Someone sends a congratulatory email that uses the phrase great outcome.

Six months later, almost nobody is using what was built. The problem it was designed to solve has not, in any meaningful sense, been solved. The value that justified the investment has not materialised.

Was that a successful project?

You would think the answer was obvious. And yet, in the majority of organisations, it would be recorded as one. Because the metrics said so. Because the project closed on the right date, within the right number, and the RAG status was green throughout.

Bent Flyvbjerg, professor at Oxford's Saïd Business School and perhaps the world's foremost researcher into why major projects go wrong, has spent a career documenting exactly this problem. In How Big Things Get Done, he and his co-author analysed data from thousands of projects across the globe and found that the overwhelming majority failed to deliver their intended value – not because they ran over time or over budget, but because the outcomes they achieved bore little resemblance to the outcomes that were promised. We have, he argues, become extraordinarily good at measuring the wrong things.

Antonio Nieto-Rodriguez makes the same point from a management perspective in Harvard Business Review. "Too many organizations still define project success as 'on time and on budget,'" he writes, "and never ask, Did we actually create the value we intended? If you measure the wrong things, you drive the wrong behaviours."

The wrong behaviours, in this context, are not hard to imagine. Teams optimise for the deadline rather than the outcome. Difficult truths get smoothed over in progress reports because nobody wants to be the person who turns the dashboard amber. Features get delivered because they were in scope, not because they are needed. And at the end of it all, everyone agrees it went rather well, while quietly knowing that it didn't.

The problem is not dishonesty. It is the entirely human tendency to be judged on what can be measured, and to focus, under pressure, on the things you can control. Dates can be controlled. Budgets can be managed. Whether the new system actually changes the way the business operates – that is a much more complicated question, and it doesn't fit neatly into a fortnightly status update.

Nieto-Rodriguez suggests a better set of questions. Leaders in a project-driven organisation, he argues, should be able to answer these at any moment: What value will our current projects deliver – this quarter, this year, over the long term? What is each project's real-time status? Do we have the flexibility to accelerate value creation where it's working, and stop it where it isn't?

Most organisations cannot answer these questions. Not because the information doesn't exist somewhere, but because nobody has made it their job to ask them – regularly, honestly, and without confusing activity with progress.

There is a discipline here that matters enormously and arrives almost entirely too late in most projects. It is the discipline of defining, before anything else, what success actually looks like. Not what will be delivered. What will be different. Not what the system will do. What the organisation will be able to do that it cannot do now.

This sounds straightforward. In practice, it is one of the hardest conversations in project work – because it requires stakeholders to commit to outcomes they cannot fully control, and to be measured against them.

A skilled project manager has this conversation at the start, while there is still time for the answer to shape the work. They return to it at every significant milestone. And when the pressure to simplify everything into a green dashboard builds – as it always does – they are the person in the room who asks the more important, more awkward question.

Did we actually create the value we intended?

Escape Company finds the people who know how to ask it.

And, more usefully still, how to make sure the answer is yes.