The 1% rule that transformed British cycling - and why NZ leaders should apply it to AI

In 2003, Sir Dave Brailsford took over as performance director of British Cycling. A team that had won just one Olympic gold medal in nearly a century, and whose riders had never won the Tour de France. What he did next wasn't dramatic. There was no revolutionary training programme, no overnight transformation. Instead, he introduced what he called the aggregation of marginal gains: the belief that if you improved every single element of what it takes to win a bike race by just one percent, those gains would compound into something extraordinary.

They won the Tour de France within three years. They dominated the Olympics for a decade.

I've been thinking about this framework a lot when it comes to how NZ enterprises approach AI. Because right now, most are waiting for their Tour de France moment, the single transformational breakthrough that justifies the investment and silences the board. And in doing so, they're walking straight past the marginal gains that are sitting right in front of them.

The chicken and egg problem.

Here's what I hear most from senior leaders when I ask about AI strategy: "We're not sure where to start." And I get it, it's a genuine challenge. How do you set a clear AI direction when you're still figuring out what the technology is actually capable of? How do you build a strategy around something that's evolving as fast as this?

The answer most organisations land on is to wait. Wait for more clarity. Wait for better tools. Wait for a competitor or other company to go first. And meanwhile, the gap between organisations who are moving and those who aren't is widening fast. PwC's 2026 AI Performance Study found that nearly three-quarters of all AI economic value is being captured by just one-fifth of organisations. The rest are watching.

Part of what's driving this paralysis is the hype. The media narrative around AI has been dominated by enormous numbers, millions of jobs automated, entire industries disrupted, productivity gains that sound almost impossible. When the bar is set that high, a 5% improvement in how your finance team processes invoices doesn't feel worth celebrating. But it absolutely should.

What marginal gains actually looks like in practice.

The organisations making real progress aren't the ones with the most sophisticated AI strategy on paper. They're the ones that have got intentional about experimentation, identifying specific problems or opportunities, running focused tests, measuring actual productivity gains rather than tool adoption metrics, and then making a clear call: scale what works, celebrate what doesn't, and move on.

One problem or opportunity at a time. That's it.

In practice, this looks like the customer service team finding a way to resolve routine queries 20% faster. The marketing team halving the time it takes to produce a first draft. The finance team cutting report preparation from a day to an hour. The HR team using AI to surface insights from engagement data they never had time to properly analyse before. None of these is a moonshot. Each one, on its own, is unremarkable. But stack ten of them across ten departments, and you're not looking at a 10% improvement. You're looking at a compounding gain that reshapes what your organisation is capable of.

Brailsford didn't try to find one thing that would make his riders 10% faster. He found a hundred things that made them 1% better. The maths worked out the same, but the second approach was far more achievable, far more sustainable, and far harder for competitors to replicate.

Stop waiting. Start experimenting.

The most important shift for NZ leaders right now isn't a technology decision, it's a mindset one. It's moving from "we need a strategy before we can act" to "the experiments are how we build the strategy."

Safe-to-fail experiments are the discovery mechanism. They're how you find out where AI genuinely helps in your organisation, and where it doesn't. They're how your people build confidence and capability with new tools. And they're how you start accumulating the marginal gains that compound into something significant, without betting the house on a single transformation programme that may or may not deliver.

The key word is intentional. This isn't about encouraging everyone to play with ChatGPT and see what happens. It's about identifying the highest-potential problem or opportunity areas, designing a focused experiment, measuring what changes, and making a disciplined call on what to scale. Then doing it again in the next department. And the next.

But experiments alone aren't enough.

Yes, targeted experimentation will unlock the marginal gains, no doubt about it. But it won't tell you the full picture. It won't surface whether your leadership team has the strategic clarity and capability to sustain AI-led change. It won't reveal whether your culture supports the psychological safety required to experiment and fail. It won't show you whether your performance frameworks are set up to measure the outcomes that actually matter, or whether your people infrastructure is ready for the roles AI will create and change.

That broader picture, across leadership and strategy, performance, culture, people, and technology, is exactly what the AI Compass is designed to surface. It's not a technology audit. It's an organisational readiness assessment that shows leaders where they genuinely stand across the five dimensions that determine whether AI transformation succeeds. And from there, we work with you on a plan to address it.

Brailsford didn't just improve the bikes. He improved the nutrition, the sleep, the psychology, the team culture, and the race strategy. The marginal gains worked because they were applied across everything, not just the obvious stuff.

The same is true for your AI transformation.

At Five NZ, the AI Compass helps leaders understand where their organisation stands across the five dimensions that determine whether AI transformation succeeds: Leadership & Strategy, Performance, Culture, People, and Tech.

Start with our free AI Readiness Check — it takes less than five minutes and gives you a clear picture of where to focus first.

Take the free readiness check at aicompass.io →

#MarginalGains  #AIStrategy  #AITransformation  #LeadershipNZ  #AICompass  #FiveNZ

‍