Put two companies side by side today and you would struggle to tell them apart.
Same revenue. Same market. Close to the same headcount. Both have spent the last year putting AI into the business, and both are proud of it. Walk the floor of either one and you would see the same thing. People moving faster than they did a year ago. Tools humming quietly in the background of every workflow. A leadership team that feels, for the first time in a while, like it is keeping up.
Stand close enough and the two companies are identical. That closeness is exactly what hides the thing that matters.
Because three years from now they will look nothing alike, and the reason will have almost nothing to do with which one bought more of the technology.
I have sat with both kinds of company. I want to describe them honestly, because the difference between them is quiet enough that most leaders never see it coming.
Call the first one Northline. Good company. When their leadership decided to get serious about AI, they did the sensible, energetic thing. They pushed it down into the work. Marketing started producing three times the volume. Engineering closed tickets faster and shipped more often. Support resolved cases in half the time. Finance closed the books in days instead of weeks. Every one of those gains is real. I am not going to pretend otherwise, and neither should you.
But watch where the direction of all that work comes from. The strategic calls, the ones that decide which market to chase, which product to bet on, which customer to walk away from, still get made the way they always were. A few people in a room, some numbers on a screen, a quarterly gut check, and a decision. The engine underneath the company got faster. The steering wheel did not change hands. Northline is now executing its existing strategy at a sprint.
The second company, call it Meridian, did something that looked slower and less impressive at first. Before they worried much about speeding up the work, they put AI into the handful of decisions that set direction. Not the thousand small operational choices. The ten or twelve big ones a year that actually determine where the company ends up. Which segment. Which bet. Where to concentrate, where to retreat, when to move.
If you visited Meridian in that first year you might have come away unimpressed. Their teams were not obviously faster than Northline’s. Some of them were slower. They were spending real effort on decisions that, from the outside, looked like they were taking too long to make.
They were not faster. They were pointed better. And being pointed better does something that being faster does not. It compounds.
Here is the part worth slowing down for, because it is the whole game.
Speed is a one-time gain. You get faster, and then you are faster, and that is the end of the story. It is a step up, and you keep the step, but it does not build on itself. A team that ships twice as fast this quarter ships twice as fast next quarter. Valuable, banked, done.
Direction is different. A slightly better strategic call this quarter does not just pay off this quarter. It changes which decisions you even face next quarter. Get the segment right and your hiring starts from better ground, your roadmap starts from better ground, your pricing and your partnerships and your next three bets all start from better ground. Then the quarter after that, you are making your next good decision from a position the other company never reached. The advantage does not add. It multiplies against itself. Three years of that is not a bigger version of the same company. It is a different company standing somewhere the first one cannot get to from where it is.
Now let me give the other side its due, because there is a real argument against everything I just said, and I have made it myself.
The argument goes like this. Markets move fast and reward execution. Strategy is mostly guesswork dressed up as rigor, and the honest move is to run experiments in the real world and let results tell you where to go. Getting AI into every team is concrete and measurable. You can see the throughput climb. Fussing over “where the decisions are made” is the kind of abstract executive talk that sounds wise in a keynote and does nothing on a Tuesday. Speed, this argument says, is not the enemy of direction. Speed is how you find direction, by trying more things faster.
I believed a version of that for a long time. I have walked into companies, seen operational chaos, and told the founder the fastest way to clarity was to fix the machine first and worry about the map later. Get the work flowing, then think about where it should flow. It felt practical. It felt like the responsible thing to tell someone who was drowning.
And it is not entirely wrong. Execution does teach you things planning cannot. The trouble is what happens when you get very good at moving before you get good at aiming. You do not find direction faster. You just cover more ground in whatever direction you were already facing. I have watched a company I advised execute its way, cleanly and impressively and under budget, into a market that was quietly shrinking underneath them. Every dashboard was green. The velocity was beautiful. They arrived early, and they arrived somewhere they never should have gone. That one stayed with me. I stopped telling founders to fix the machine first.
That is the trap, and it is a cruel one, because it feels like the opposite of a trap the entire time you are in it. Faster execution feels like progress. It feels like winning. Northline will feel productive for three straight years, right up to the moment they look up and realize they got somewhere very efficiently and it was the wrong somewhere.
The divergence does not announce itself. That is what makes it dangerous. It does not arrive as a bad quarter or a missed number you could point to and correct. It accrues, one slightly better decision at a time, most of them too small to notice on their own. And then there is a second effect, further out, that almost nobody plans for. The better-pointed company starts setting the terms. It gets to the good markets first, prices with more room, hires the people who want to work somewhere that seems to know where it is going. The faster company finds itself reacting. Responding to moves instead of making them. Its best people begin to sense the drift before the leadership will name it, and some of them leave, and that is its own kind of compounding, running the other way.
I have watched both versions up close, and the part that stays with me is how alike they looked at the start. Same size. Same tools. Same pride. You could not have picked the winner in the room. You could only have asked one question, and the answer would have told you everything: not how much AI does either company have, but where did they point it first.
In three years the tools will be everywhere. Everyone will have them, and having them will mean nothing, the way having electricity means nothing now. The only thing that will still be scarce is intelligence placed where the direction gets set.
You can be faster, or you can be pointed right. In three years, only one of those turned out to be a strategy.
At Kennesys, the fractional CIO and CTO work I do is about that first placement: building the strategic layer where a company decides its direction, before the rest of the machine learns to run.