When someone gets it right, how does anyone else learn from it?
When someone breaks a rule, makes a mistake, or misses a procedure, you have a form. What happened. How it happened. How to make sure it never happens again. The finding goes into the rule, the rule goes into the training, and everyone learns what not to do.
When someone gets it right on a brutal day, short-handed, behind schedule, with every good reason to cut the corner, there is no form. Nobody asks how they did it. Nobody asks what they were reading in the room, or what they knew that was never written down. The success leaves no record and teaches no one.
So how do you repeat what worked, when nobody captured the human capability that made it work? How do people follow best practices with no account of how those practices survive contact with a real day? And how does what your best people know ever reach anyone else, if the only thing your system is built to capture is failure?
Every organization has a machine for learning from what went wrong. Almost none has one for learning from what went right.
Making the connection
In juvenile corrections, kids do things most adults can't. They read each other. They name what is happening in someone before that person can name it themselves. They track agency, regulation, and patterns in themselves and in the people around them.
They do it because it is tied directly to their success. Noticing where those capabilities show up in others is part of the treatment. They report on it. They use it to explain what broke and what worked. It is tied to their release.
Then they go home, where nobody ties those same capabilities to their success. More than seven in ten come back.4
That connection is not something you can teach as an add-on. It has to be built into the process.
See the training that came out of it →
From knowledge transfer to experience integration
For a century, training has meant knowledge transfer. Content moves one direction, from the expert, to the material, to your people, who click through and sign. It produces teams that can pass the quiz. Check-the-box teams. The box matters. It was just never the point.
The other way runs a generative loop. The same training that teaches the procedure asks your people where they have already done it well, under real conditions. Their answers come back in. Their answers become the training. That loop is what we call experience integration.
One direction produces teams that check the box. A generative loop produces teams that get better at the work, and better at understanding and working with each other.
It sounds like a small shift. It changes what the training is for.
The best teams learn from and about each other
Humans didn't become expert hunters from lessons alone. It was the story that brought the lesson to life: why the decision got made, how they knew, what it felt like when it went sideways. After the hunt the questions were simple. What worked? How did you do that? The story was always part of learning, until we took the human out of the loop.
The best teams never took it out. They mine their successes. They hold each other accountable for greatness, not just compliance. They learn from and about each other. From each other's wins, so they have more wins to learn from. About each other, so they know who to ask on the day it matters. Even teams that do nothing more than run a debrief outperform teams that don't by roughly 25 percent, across 46 samples in a peer-reviewed meta-analysis.1
No procedure has one right way through it. It has as many ways as you have people who have done it well on a bad day. The best teams know what those ways are, and who carries which.
Locking out isn't the choice. How you make sure it happens on a loud day is. Three answers of the kind this collects:
Which of these sounds most like something you do, and what's your own version?
⟨ + add your own ⟩Today, we don't need hours to sit around a campfire, and we don't need years of development to build the understanding of the best teams. We need the right questions and the right systems.
Today, we can replicate the practices that have always made the best teams, in the processes we currently run.
Why now
Your biggest source of learning is walking out the door. What your most experienced people know was never in the materials, and no new hire and no content library brings it back once it's gone.
The industry built to answer this moves more than $350 billion a year.2 And a year after the training, in the largest survey of its kind, only about a third of people were still using what it taught them.2 More content has not moved that number. Content was never the scarce half.
The scarce half was capturing exactly how your people succeed, reading hundreds of their accounts, and feeding the pattern back into the training. Two years ago that was impossible at any sane cost. It isn't now.
AI can do everything except be there. What happens there can finally be kept.
Three moves
So we built it. This is Finding Good: not another concept, a loop that runs inside the work. It happens in three moves.
First, we read the work. Before we ask anyone anything, we take the rules, procedures, and best practices you already have, or, where nothing was ever written down, what your best people carry in their heads, and we identify the human capabilities each piece of the work actually depends on. Not the steps, the steps are already written down. The judgment, the noticing, the willingness to stop the job. Nobody ever wrote those down. They are what the procedure is silently asking for, and they are what it fails without.
Take one line from a real procedure: lock out the machine before clearing a jam. The step is clear. What it silently asks for is the willingness to stop a line everyone needs running.
Then the training asks. Sometimes that is a question layer riding inside the training you already run. Sometimes we build the training itself, out of your own procedures. Compliance, onboarding, technical, leadership, whichever it is. Either way it ends in the same place: the training itself changes. Where the procedure says lock out the machine, the training asks: when have you stopped a job everyone needed running, and what told you it was time? The answer carries the capability, in their words.
Something else happens in the telling. Most people cannot see their own capability until someone asks for the story of it. Asked, they hear it out loud, and what they have always done becomes something they know they do. And the way a team tells them, taken together, shows how ready the work is. Call that a readiness read: a read on the group, always with the stories underneath it. It points one direction. The read exists to make the next run of the training better.
Then it comes back. Every run of the training produces new accounts. Every account makes the next run deeper, more specific, and more in your people's own voice.
The first two moves run today. Closing the third, so that every account feeds the next run on its own, is what we are building now, with the first few partners.
The assessment is the intervention. The loop is the product.
What actually transfers
We don't collect stories. The story is the packaging. What transfers is the experience inside it, and most of an experience is noise.
We've built the models that find the signal: the handful of capabilities that decide, on the evidence, whether the work actually gets done. We know what to look for in your rules and procedures. We know how to ask for it, with questions built for the job. And we know how to get what we find back into the training, so the loop keeps running.
Safety proved the pattern first. How people see those capabilities, in themselves and in each other, predicts whether the work gets done. That is the whole logic of the Bradley curve:3 a workforce's safety culture, read through how people see each other and the work, tracks with its injury rate. The curve's top rung is interdependence: people watching out for each other. Safety has organized itself around that model since 1995. It has never built the training that produces the top rung. And the pattern was never only about safety. It holds for any work that runs on judgment.
We use the same pattern to decide what the training teaches next.
Put the right experience in front of the right person at the right moment, and it stops being a story about the job. It becomes the other half of institutional knowledge: the half most companies lose without ever knowing they had it.
For the company and for the people
The company gets three things training has never given it. First, the readiness read: where the work is ready to move and where it needs building, grounded in real accounts instead of a dashboard number, and early enough to act on. It also tells you where to start, which is the question nobody can answer for you: read across a few hundred accounts and one part of the work comes back ready to move, because somebody in it is already working the way you would want the whole crew to work. You start there, with them. Second, the knowledge that used to leave with the veterans, kept and teachable. Third, the thing every leader asks for and cannot mandate: people connected to the work and to each other, because being heard is built into the format, not bolted on beside it.
The people feel it differently. Not fewer challenges, more capacity to meet them. Everyone who runs it leaves with three things of their own: a check they will run on their next job, a question they will ask someone, and a person they will involve. They wrote all three. And it moves in both directions: the person who tells the story leaves with more than they brought, and so does the person who asked. One client, after a season of the work:
"You've given me a vocabulary and a vernacular… something that I've done in the past, but I didn't know how I did it."
That is also where buy-in comes from. Nobody here gets convinced of anything. People commit to what they helped build, and this is built from what they did.
Connected, not convinced.
Closing the shelves this page creates
A soft-skills workshop teaches connection off to the side of the work, and what is taught off to the side leaves with the person who learned it. Nothing here happens off to the side. The question rides inside training your people already take.
It is not a survey. Nobody is asked how they feel about the company. They are asked how they did the work, and the answer becomes the training.
Nothing gets ripped out, either. Your LMS stays. Your binder stays. The veteran showing someone the ropes stays. What changes is the training itself: what it asks, and what happens to the answer.
And the AI your LMS vendor will offer you this year is good at content: generating modules, summarizing material, answering questions about it. It cannot run the loop, because the loop is not a feature. An add-on can generate more of what you already have. It cannot send back what only your people know.
The filter at the door
This isn't for everyone, and we'd rather say so here than on a call.
It fits if you already know the check-the-box model isn't working, if you can name what it's costing you, and if you don't want to build the fix yourself.
It doesn't fit if you need completion rates to go up and nothing else. If your training exists to satisfy an auditor and you're at peace with that, or if what you want is a platform you can staff and run internally — those are real needs. They're just not this.
Your world
The shift is the same everywhere. What it touches first is not. For a manufacturer it starts with procedures and the readiness of the floor. For a field-service crew it starts with what the senior tech does that never made it into any manual. For onboarding it starts with the gap between day one and actually useful, and with the veterans who carry what the manual never said.
Most of what we do arrives as learning: the loop, running inside your training. Some of it arrives as coaching, building the practice that keeps the loop alive. Some as consulting, when the loop is ready to grow across more of your world.
We are choosing the next partners by fit, not by volume, and not from one industry. The vertical pages carry the full, honest state of the evidence.
Bring one piece of the training you already run.
In an hour, we find the capabilities underneath it together, live, and build the questions that draw them out.
You leave with that question set and a one-page pilot plan. Yours either way, whether or not we ever work together.
We are early, and we are choosing partners rather than stacking a pipeline. The next few shape what this becomes: how the loop closes, and what it looks like when your own leaders are the ones finding the good. If you want your fingerprints on it, this is the window.
Reach out to Brian to see how you can turn your successes into a source of learning, connection, and growth.