Why most innovation labs fail before they launch (and the one behavioural fix that changes everything)

I watched a Fortune 500 company spend £2.3 million on their innovation lab before anyone had worked out what would actually happen inside it.

Beautiful glass walls. Whiteboards everywhere. Sticky notes in every colour. A coffee machine that cost more than most people's cars.

Six months later, it was a glorified meeting room where people held the same PowerPoint reviews they had always held.

This is not unusual. Most innovation labs fail before they launch, and it has nothing to do with budget, leadership support, or the wrong facilitators.

It is because they solve the wrong behavioural problem.

The real problem is not capability, it is permission

Every innovation lab I have seen starts with the same assumption: people need better tools and processes to innovate.

Wrong.

People already know how to solve problems creatively. They do it at home every weekend. The issue is not capability, it is permission.

In most organisations, the unwritten rules are clear: do not rock the boat, follow the process, stay in your lane. Innovation requires the opposite behaviour. Take risks, challenge assumptions, try things that might not work.

You cannot put people in a room with fancy tools and expect them to suddenly ignore every signal their organisation has been sending them for years.

When I worked with NatWest on their innovation programme, the first session was painful to watch. Brilliant people, genuinely committed to change, sitting in silence because they were terrified of suggesting something that might fail.

The behavioural problem was not skills. It was safety.

The one fix that changes everything

Before you design your first workshop or buy your first sticky note, you need to solve for psychological safety at the organisational level.

This means creating explicit permission to fail in service of learning.

Not the usual corporate speak about "failure is learning." Actual permission, with actual protection.

At Disney, we solved this with what we called "innovation insurance." Any project that came out of the innovation lab had automatic protection from performance reviews for 90 days. If it failed, it failed safely.

The behavioural change was immediate. People started suggesting ideas they would never have mentioned in a regular meeting. Not because they suddenly got more creative, but because the social and professional risks had been removed.

Three action steps to create real innovation permission

1. Make failure protection explicit and visible

Do not just say "we encourage risk-taking." Create an actual mechanism that protects people when experiments do not work out.

This could be a formal policy that innovation projects cannot negatively impact performance reviews. Or a "learning fund" that covers the costs of failed experiments without affecting departmental budgets.

The key is making it visible and systematic, not just a nice sentiment from leadership.

2. Start with micro-innovations, not moonshots

Most innovation labs try to solve massive problems on day one. This creates enormous pressure and reinforces the fear of failure.

Instead, start with tiny experiments that take less than a week to complete. A new way to run team check-ins. A different approach to customer feedback. A small process improvement.

The goal is not the innovation itself, it is building the muscle memory for trying new things without catastrophic consequences.

3. Celebrate the learning, not just the success

Create a regular forum where people share what they learned from failed experiments. Make it as visible and celebrated as success stories.

At American Express, we ran monthly "failure parties" where teams presented their most interesting failures and what they discovered. These became more popular than our success showcases because people could be genuinely honest about what did not work and why.

This is behavioural design, not just culture change. You are creating new reinforcement patterns that reward the behaviour you actually want to see.

The innovation lab is not the innovation

The physical space matters, but it is not where innovation happens. Innovation happens when people feel safe enough to suggest something that might not work.

Your innovation lab will succeed or fail based on whether you solve for permission before you solve for process.

Everything else is just expensive furniture.

Innovation is not about having better ideas. It is about feeling safe enough to share them.