If you've led a business for any real length of time, you already know the meeting I'm about to describe.
Call it the weekly. Or the monthly. Or the quarterly review. Everyone's in the room, everyone's got their numbers, and everyone presents. And here's the thing about numbers — anyone with a little experience can make them tell a favorable story. So the dashboard is green. Everyone's on track. Everyone's winning.
And everyone in the room knows better.
Because you also know what's sitting behind the green. The goal that quietly got missed and recategorized. The budget that got blown. The process that broke in week two and never really got fixed. You know somebody stayed until nine three nights running to make that number land. You know somebody covered for somebody else. You know the target got hit on the back of one person's grit — not because the system produced it, but in spite of the system.
That's the gap I want to talk about. The distance between what the numbers say and what everyone in that room actually knows.
For a while, that gap stays invisible. And the thing that fills it — the thing that hides it — is your people.
Great people hide broken systems. So does luck. So do heroics. So does plain determination. A good enough team will carry a bad enough process for years, and from the outside — from the dashboard — it looks like the process works.
It doesn't. It never did. They did.
And you don't find that out on a good day. You find it out when the person who quietly held three things together finally leaves. Or when volume doubles and the workaround that got you through at ten can't survive at fifty. Or when you grow — because growth has a way of finding every crack you'd papered over with effort.
Nothing broke. That's the part that catches people off guard. The system didn't suddenly stop working. It was never working. Your people were working, and the system was riding on them the whole time.
Which brings me to the thing it took me far too long to really understand.
Why people are almost never the problem
Start with something I've found to be true in nearly every company I've ever looked at: people want to do good work.
They do. Very few people show up hoping to be mediocre. They want to be competent, to be trusted, to go home feeling like the day meant something.
So what goes wrong? When expectations are clear, good people do good work. When expectations are unclear, those same good people start to guess. They're not being lazy — they're filling a vacuum. Someone has to decide, and if the system hasn't decided, the person will.
But my guess and your guess aren't the same guess. Multiply that across a team and a few months, and you don't have a team anymore. You have a dozen slightly different versions of how things are supposed to work, all running at once, all sincere.
That's inconsistency, and it's invisible until it fails. Everything looks fine right up until two of those guesses collide in front of a customer. Then someone gets blamed, everyone resolves to try harder, and nothing changes — because the bottleneck was never the person. The blame just buried it deeper.
Let me make it concrete, with one I'm in the middle of right now. A business has to answer a simple question: do we have the staff to actually run the appointments we budgeted for? Book more than you can staff and you set people up to fail. Staff for more than you book and you've got people sitting idle. Either way you want a real answer — and a guess won't do.
There was never a real system for tracking it. Years ago someone built a workaround — a way of counting that worked well enough for the moment — and then everyone moved on and left it running. It goes quiet for long stretches. It only rears its head when someone isn't hitting their number. Then the capacity question comes roaring back, and instead of a straight answer it turns into a guessing game and a round of finger-pointing — because the data everyone's arguing over is weeks out of date.
And every time, the explanation is the same four words: “That wasn't up to date.”
Sit with that sentence, because it's the tell. “That wasn't up to date” is what a system failure sounds like when it's wearing a person's clothes. Nobody was careless. The number could stay current — nothing keeps it current. So it costs real hours every week and every month just to patch back together, and it goes stale about as fast as anyone can update it.
And here's the part that should worry any owner. That workaround lives in two or three people's heads. If one of them takes a new role or walks out the door, the workaround goes with them — and there's no system underneath to train the next person on. You wouldn't just lose the answer. You'd lose the only way anyone knew how to reach it.
That's the gap — not a people problem, a missing system — and it has to be built before the people holding it together are gone. It's exactly the kind of thing we're building a system for right now; it's still in development. But the fix was never “try harder.” It was always going to be one place that keeps the score current on its own, so the answer is the same no matter who's asking.
Every company has two versions of reality
Go back to that review meeting for a second, because it was really about two versions of reality sitting in the same room. Every company has them. There's the data — what the numbers say happened. And there's the story — what people know actually happened, and why. The green number on the slide is the data. “Someone stayed late three nights to make it green” is the story.
Most advisors pick a side. The finance-minded ones worship the data and dismiss the story as noise. The people-minded ones trust the story and wave off the data as cold. Both are half-blind.
Because data and story each know something the other doesn't. Data tells you what happened. It doesn't care about anyone's ego, which is exactly why people resist it — but it can't tell you why. The story tells you why. It's where the real texture lives: the workarounds, the history, the unspoken “that's just how we've always done it.” But the story is also where people protect themselves, where memory bends toward the comfortable version.
Data tells you what happened. The story tells you why. The truth lives in the seam between them.
Neither is enough on its own. The data without the story is a number with no cause. The story without the data is a hunch with no proof. That's the actual work — not the software, not the AI. Finding the truth that lives in the space between what the numbers say and what people know.
How we find a bottleneck: the Obsidian Method
Over enough of these, I noticed I always move in the same order. It's not a checklist I designed on a whiteboard — it's just what the work honestly requires. But it's consistent enough to name.
Until you've done the first five, any tool you buy is aimed at a problem you haven't actually found yet — and a tool aimed at an unnamed problem is just an expensive guess.
Why software doesn't fix a broken process
Companies feel the pain, and pain wants a solution now, so they buy one. A platform. A subscription. A dashboard. And it doesn't help, because technology reinforces whatever process you already have. Automating a broken process just lets you produce the wrong result faster and more consistently. That's not a fix. It's an efficient way to make the same mistake at scale.
Software makes a healthy system faster. It makes a confused system confused at speed.
Get the system right first. Then ask what technology could make it faster, cheaper, or impossible to get wrong. In that order — never the reverse. Which is why, before you evaluate a single tool, it's worth sitting with three quiet questions. Do your people actually understand the mission — not the words, the point? Are your core values alive in how decisions get made, or framed on a wall and contradicted by Tuesday? Does everyone know who owns each process, or is ownership one of those things that's assumed right up until the moment it fails?
What AI is actually for
The anxious version of the AI conversation is about replacement — what it takes away, who it makes redundant. That framing misses the point, and it certainly misses the opportunity. The real value of AI isn't that it replaces people. It's that it gives leaders their time back — and time is the one resource a leader can never buy more of.
Say building a sales certification takes a good manager six hours. Research, structure, writing, formatting. Now say the right tool produces a strong first version in five minutes. The mistake is to call that six hours saved, as if the point were subtraction.
It isn't six hours saved. It's six hours moved — out of document formatting and onto the floor, coaching a struggling rep, sitting in on a hard call, noticing the quiet employee who's about to quit. The tool doesn't do the leadership. It clears the underbrush so leadership can happen at all.
The report nobody should be building by hand
There's one bottleneck hiding in nearly every company, and it's almost always worse than the owner thinks. Reporting. Somewhere in your business, right now, a capable and well-paid person is stitching numbers together by hand — pulling from one place, pasting into another, rebuilding the same report they built last week and will rebuild next week. It feels like work. And it's usually the exact report that shows up green in the meeting.
But the assembly isn't the value. The thinking about what the report reveals is the value — and the assembly is eating the time that thinking needs. Put a number on it. Three hours a week is roughly a hundred and fifty hours a year, from one of your sharpest people, spent copying and pasting.
So ask the real question. What would you do with a hundred and fifty hours? That's what a bottleneck actually costs. Not just the hours — everything those hours could have been.
The companies that win
The companies that win over the next decade won't necessarily have the best products. They won't have the biggest marketing budgets. They won't even have the most people. They'll have the clearest systems.
Because systems create consistency. Consistency builds trust — the customer who knows exactly what they'll get, the employee who knows exactly what's expected, the leader who knows the green number is actually green. And trust, compounded over years, is what builds a great company. Everything else is weather.
Every business has bottlenecks. That's not a flaw; it's just physics. The difference is that the best businesses find theirs before they become expensive — while it's still a quiet leak, and not yet the thing everyone in the room already knows and no one says out loud when the dashboard is green.
So the next time everything comes up green, ask the harder question. Not “did we hit the number?” but “what did it cost us to hit it — and would the system have hit it without the heroics?”
Stop asking who's slowing you down. Start asking what is.