Chris, I love you, but you’re a roadblock.

The CEO stopped in my doorway in the middle of the afternoon and looked at the line. Four people, standing against the wall outside my office, each holding something. A printout. A laptop. A question that couldn’t wait. He watched them for a moment, stepped inside, closed the door, and said, “Chris, I love you, but you’re a roadblock.”

I remember being genuinely confused. Not hurt. Confused.

Because I’d read that line of people as a good sign. Look how essential I am. Look how much gets decided in this room. My door was always open, my answers were usually right, and people kept coming back for more of them, which I took as evidence the whole system was working. The busier the doorway, the better I was doing my job. That was the scoreboard I’d built for myself, and by that scoreboard I was having a great year.

He looked at the same hallway and read an outage report.

I’ve thought a lot since then about the word he picked. Roadblock. He didn’t say controlling, or micromanager, or any of the character words, the ones that start arguments because the only way to dispute them is to defend your personality. A roadblock is infrastructure. It’s a thing that got put somewhere, probably for a reason that made sense at the time, and now the road needs it gone. He wasn’t diagnosing me. He was diagnosing the layout. I couldn’t hear the difference that afternoon, but that word choice is most of why I eventually could.

It took me longer than I’d like to admit that he was right, and longer still to work out what exactly he was right about. It wasn’t a performance problem. Nobody in that line was underperforming, including me. I was working hard, they were working hard, the answers were good. It was a design flaw. I had built, mostly by accident and partly out of vanity, an organization where the fastest path to a decision ran through one office. Mine.

Think about what that line actually was. Every person standing in it had a manager. Most had a manager between them and me. And every one of them had decided, correctly, that the efficient move was to skip their own leader and wait in my hallway. Correctly is the part that should have scared me. They weren’t wrong to be there. I had made the hallway the best answer available. Somewhere along the way I’d taught the whole organization that a decision wasn’t real until it had passed through me, and everyone had simply adapted to that fact, the way people adapt to a road closure. Traffic finds the open lane. I was the open lane.

I want to be clear that I delegated constantly, because for years that was my defense. My calendar was full of things I’d handed off. But I was handing off tasks, and a task comes back. A task shows up in your doorway holding a printout, asking if this is what you meant. What I almost never handed off was ownership. Ownership doesn’t come back to check. Ownership comes back with an outcome.

The difference sounds subtle. It isn’t.

When you delegate a task, you keep the decision and lend out the labor. The person doing the work is executing your judgment, and they know it, so at every fork in the road they pause and file back into the hallway. When you delegate ownership, the judgment goes with the work. The person at the fork picks a direction, because it’s their road now. The first model feels safer, and in any given week it probably is. The second model is the only one that builds anything. Leaders who are never allowed to own a hard call don’t develop the muscle for hard calls, and then the day comes when you need a bench and you discover you’ve got a waiting room.

The cost isn’t only development. It’s speed, and it’s morale, and those two show up in places you don’t think to look.

Speed first. An organization where decisions pool at the top can only move as fast as the top person’s calendar. That’s not a metaphor. It’s arithmetic. If forty meaningful decisions need to happen this week and thirty of them are waiting on me, the company’s clock speed is my Outlook. I used to stay late and feel heroic about it. I was the bottleneck congratulating itself for widening by an hour.

Morale is quieter. For a long time I read our employee engagement scores the way you’d read a report card, a grade the organization earned, and when a number dipped I’d think about what the team could do differently. At some point I started reading them as a diagnostic instead. Not a grade for them. An instrument panel pointed at me. One year a team’s autonomy numbers slid, and instead of scheduling a morale event I traced the actual decisions that team had tried to make that quarter. A pattern showed up fast. Their calls were moving through me. I’d been getting pulled into tactical choices, staffing questions, sequencing questions, things three levels below my job, and every time I touched one I taught that team their judgment was provisional. The engagement survey wasn’t measuring their attitude. It was reporting my behavior back to me, with a delay, in a format I’d been misreading for years.

Then there was the conference. Years later, different company, and by then I’d have told you I had this figured out. At our annual customer conference I presented five sessions in a single afternoon. Five. I walked off the last stage with my shirt stuck to my back, feeling capable and useful and, honestly, a little proud. The pride lasted maybe a minute. Because standing in the hallway afterward, what I actually thought was: that was the roadblock again. Wearing a lavalier mic this time, but the same design flaw. Five sessions meant five topics where the person on stage should have been one of my leaders, building their own reputation with our customers, becoming the person the audience emails afterward. I had just spent an afternoon being impressive in five slots where somebody else needed the reps.

A senior director had joined us a few weeks earlier. I found her and asked her to hold me to something: next year, my leaders present. Not most of the sessions. The sessions. I asked her precisely because she was new. She hadn’t learned to route around me yet, and she had no reason to let me off politely.

Underneath all of this was one question I had to physically replace. The question I’d been asking my whole career, the one that got me promoted, the one that felt like leadership from the inside, was: how do I solve this? It’s a good question. For an individual contributor it might be the question. But asked from my chair it has a side effect. Every time I answer it, somebody else doesn’t.

The question I use now is: who can grow by solving this?

Nobody warns you about the transition, though, so I will. For the first few months, refusing to answer feels like negligence. Someone brings you a real problem, you can see the answer, seeing the answer is literally why they pay you, and you deliberately don’t say it. You ask whose call it should be instead. People walk away unsatisfied, and a few of them, for a while, will conclude you’ve checked out. One of my directors more or less asked me if everything was okay at home. The organization has been trained on a certain shape of you, and it will keep pressing the old buttons long after you’ve rewired what they do, and you have to survive that stretch without caving, because caving once teaches everyone the hallway still works.

It’s slower. I’ll be honest about that. The first question resolves the issue by Friday. The second one means a rougher draft, a wobblier first customer call, a decision I’d have made ten percent differently. For a while you’re trading crisp answers for developing judgment, and the trade feels bad almost every individual time you make it. Then it compounds. The person who owned the wobbly call owns the next one without you, and teaches the one after that, and a year later a whole category of problem simply stops reaching your hallway, not because you closed your door but because there’s nothing left to bring.

The line outside my office is gone now. When somebody does show up with a question their own leader could answer, I’ve learned not to answer it. I ask who they’d take it to if I were on a plane, and then we walk there together, because the walk is the message.

And if you’re at our conference next year and I’m on stage five times, you’ll know the roadblock came back. It brought slides.