The best don't make fewer mistakes. They waste fewer.

Prefer to listen in your podcast app? Apple Podcasts · Spotify

In 1942, Britain was losing a war it could not see. German U-boats were sinking Allied ships across the Atlantic faster than the shipyards could replace them. Convoys crossed with destroyers for protection, and still the submarines slipped inside the escort screen and picked off ship after ship. Every convoy commander had experience. Every captain had a theory. None of them had an answer.

So the Royal Navy tried something that looked nothing like fighting a war. Instead of sending another ship to sea, it painted a large grid on the floor of a room in Liverpool.

This was the Western Approaches Tactical Unit, set up at Derby House under a captain named Gilbert Roberts and staffed largely by young women from the Women's Royal Naval Service, known as the Wrens. Escort officers walked the grid on foot while the Wrens moved model ships and submarines by hand. Every move stood for two minutes of a real battle. Convoys that had already been attacked were rebuilt piece by piece, slowly, until the room could finally answer a question the sea never gave anyone time to ask.

Not “Who made the mistake?” but “What actually happened?”

And patterns began to surface. The submarine wasn't appearing by magic. It was exploiting the same blind spot, the same seam in the escort screen, over and over again. Once you could see it laid out on the floor, you could plan for it.

The breakthrough mattered. What happened next mattered more.

Every week, a new group of escort officers came through Liverpool. They learned the latest tactics on the grid and carried them back to sea. Reports came back from real convoys, the simulation absorbed them, and the teaching changed. The next group left knowing a little more than the last one did. The unit turned hard-won lessons into named tactics and taught them on repeat, so a discovery made by one crew in the Atlantic became standard practice for crews who had never met them.

Britain's advantage in that room wasn't that it suddenly had the smartest people. It was that it had built a faster learning loop.

Like many World War II stories, that story has stayed with me, because we tend to misunderstand what makes an organization exceptional. We like to picture success going to the most talented team, the biggest budget, or the boldest strategy. Sometimes it does. More often, I think it goes to the organization that learns just a little faster than everyone else.

Not because it makes fewer mistakes. Because it wastes fewer of them.

There's a real difference between the two. Some organizations hit a problem, solve it on the spot, move on, and the fix dies with the moment. Others hit the same problem and do something more deliberate. They capture what happened, share what they learned, adjust the system, and spare the next dozen people from repeating it. The event itself lasts an afternoon. The learning compounds for years.

I've watched this play out across healthcare my whole career. One clinician finds a clearer way to teach a complex skill. Another builds a workflow that catches errors before they reach a patient. Someone at the front desk changes how appointments are booked and the no-show rate falls. Small discoveries, but real ones.

One example that resonates with this is from my own experience. Much of our work was with refugee families, which meant many patients arrived with no prenatal records at all, not for this pregnancy and not for the ones before it. Their prior health experiences on the other side of the world, in health systems nothing like ours, walked in with them and appeared nowhere on paper. Pregnancy runs on a countdown, and every week spent getting a patient onto the right path is a week you never get back—we knew we had to leverage the first encounter in a new way.

We ran it what felt like every possible way. We changed who led it and what it covered. We folded and unfolded it. We stretched it to ninety minutes, then pulled it back to forty-five, then pushed it out again to 60. We split it into two, we condensed it and re-worked it—we shook every tree in the orchard.

What we finally learned was the two things that mattered most: the timing, and the questions. So we stopped treating the first visit as an intake and started treating it as a launch. Labs, screens, and an ultrasound on the same day, with results that reflexed straight into treatment, monitoring, and specialist referrals instead of waiting for the next appointment. And we rebuilt the questions themselves, crafted to draw out a full health story rather than tick off data points, so we could take events from another country and system and find the patterns hiding inside them.

That re-design ultimately saved lives, not because anyone worked harder, but because we found the math underneath it. Time + a rapid deployment of resources + a careful choreography of access, information, planning, and action.

And to get there, we had to walk our own floors, the way those officers walked the grid in Liverpool. The patterns we needed to change were so baked into routine prenatal care that the better options weren't just hard to reach. They were hard to see at all, and harder still to build in.

The stakes weren't the same, a prenatal visit format against convoys crossing the Atlantic. But the mechanism was identical, and a discovery like that is only ever as durable as the organization's memory of it. Most improvements never leave the spot where they were born. They stay one person's trick, one person's memory, one person's spreadsheet. Then that person takes a new job, and the organization doesn't just lose an employee. It loses everything the employee had learned.

We talk about turnover as if we're swapping one set of hands for another. Sometimes we're erasing institutional memory and calling it a staffing change. Those are not the same kind of loss.

The organizations that actually get better aren't the ones with the fewest problems. They're the ones that are almost impossible to surprise twice. They notice, investigate, adjust, and then make the adjustment available to everyone else. The second time the problem shows up, the whole organization responds differently, because the first time wasn't wasted.

Every organization has a learning speed. Intelligence and talent, yes. But learning speed is worth some focus. You can almost measure it as a series of questions:

Those gaps are where cost hides. Every unnecessary week, every time we look past or over or around or through a problem, is another week spent paying tuition for a lesson the organization already bought.

It's why the complaint about too many meetings misses the point, even though most of us give far too many minutes to far too many meetings. The number of meetings was never the real question. The real question is whether learning ever leaves the room. If the same insight gets rediscovered by five teams, you don't have five discoveries. You have one discovery and four failures to pass it on.

The fastest-improving organizations aren't obsessed with collecting information—that’s Standard Operating Procedure and baked in. They're obsessed with moving it. Discovery is only half the loop. Distribution is the other half, and it's the half most places forget.

That's the real lesson of the floor at Derby House. The simulation wasn't the advantage. The weekly teaching was. Learning had been turned into infrastructure, something the organization owned rather than something each officer had to rebuild alone. The next officer never had to start where the last one started.

Neither should the next employee. Or the next clinician. Or the next founder.

Experience is valuable. But shared experience is what actually changes an organization, and it may be one of the most durable advantages any organization can build. Not because it guarantees success, but because it lets every lesson learned once be used many times over.

So here's what I'll leave you with. What is something your organization keeps learning over and over again, because no one has built a way to remember it?

Seen this pattern where you work? Hit reply and tell me. That's the harvest.

Want a bit more? Listen on Spotify or Apple for One More Signal about this topic.