The coaching session this week put words on a practice I’ve been building from day one of this role: ask before tell, listen long enough to hear what the person is actually carrying, and give counsel that lands because I already know them. I want this essay to be honest about why that practice has been the most important investment I’ve made as a manager — and about what the session genuinely added to it.
What coaching as a daily behaviour looks like for me
I want to be concrete, because the abstract version of “coaching culture” is too easy to nod at.
It looks like opening with a question I don’t already know the answer to. What are you seeing? What’s making it hard? What have you already tried? These three questions, used consistently before I offer anything, do more for autonomy than any delegation framework I’ve used.
It looks like staying in the question longer than is comfortable. The temptation when someone is thinking out loud is to fill the next pause. The pause is where the work is happening. I’m a teller — I love telling stories, I love walking through a problem from beginning to end — and learning to keep my mouth shut long enough for the other person to find their own thread has been a real practice for me. The session reaffirmed it.
It looks like resisting the urge to validate the answer I would have given. When someone arrives at a different conclusion than mine, my reflex is to weigh it, hand-correct it, suggest the path I would have taken. If their conclusion is defensible, even when it isn’t what I’d have done, the coaching move is to let it ride. They’ll learn more from owning their version than from executing mine.
It looks like not protecting people from the small failures that build judgement. The whole reason I’m tempted to step in is that I can see the rock they’re about to hit. The whole reason I shouldn’t always step in is that the rock is the teacher. Coaching is the discipline of knowing which rocks are formative and which are the ones I owe them protection from.
The biggest lesson of management for me
Here’s the thing I want to be honest about. I came into management from eighteen years of being a deeply process-oriented engineer. I love routines. I love rules. I love training a practice until it’s perfect. As an engineer, that’s what made me move forward.
The hardest and most important lesson of these first two years is that what works for me is not what works for them. My pace isn’t their pace. The way I think about a problem isn’t the way they think about it. The framework I’d reach for isn’t the one they need. Coaching is the work of learning how each person on my team works — what’s easy for them, what’s challenging, what they need more of, what they need less of — and meeting them there. That’s the part I underrated as an engineer and have spent the last two years actively learning as a manager.
The team I lead has been generous with me through that learning. They’ve told me when my pace was too fast, when my framework didn’t fit, when my advice was the wrong shape. That’s how I’ve gotten better at coaching: by listening to the people I was supposed to be coaching.
A small story
A senior on my team came to me a week ago with a tricky scoping call. Two teams wanted his work; the deadline only allowed one. He laid it out, listed his options, and asked what I thought.
The teller in me would have answered. The options were clear, my preference was clear, the conversation could have been four minutes long.
The version of me I’ve been working to be asked instead: which one do you think you should pick, and what’s the part that’s making you hesitate? He paused, talked through both options again — this time with the trade-offs articulated more sharply — and arrived at the choice I would have made anyway. But he arrived at it. He owned the reasoning. When that team later complained about the deprioritisation, he was the one who explained the call and held the line. I wouldn’t have had to do that for him, but he wouldn’t have done it as well if he hadn’t owned the decision.
The conversation took eight minutes instead of four. The downstream conversations were much shorter than they would have been. Net throughput won.
When to not coach
I want to name this clearly, because the failure mode of new converts to coaching is to coach when the moment calls for telling.
If someone is about to make a decision I genuinely believe will hurt the team or the work, I tell them. If someone is missing information they cannot reasonably reconstruct on their own, I share it. If we’re in a moment where speed matters more than ownership, I make the call. Coaching is the default, not the rule. Responsive leaders read which moment they’re in.
The line the session crystallised for me: I coach when the cost of the next mistake is learning, and I tell when the cost of the next mistake is the team. That’s the line I’d been operating on by feel, and now I have words for it. I’ll use those words.
Why this is a team effectiveness lever
A team where the manager coaches by default produces three compounding effects.
First, the team grows faster, because every problem is a small rep for someone other than the manager. Over a quarter, that’s hundreds of reps of judgement, scope, framing, decision-making — none of which would have happened if I’d answered the question.
Second, decisions move faster, because they don’t have to route through me. The whole reason a team gets bottlenecked at the manager is that the team has been trained to bring decisions there. Coaching, applied consistently, untrains that pattern.
Third — and this is the one I find most interesting — the team gets better than I am at things, in directions I couldn’t have set. The best manager I worked under as an engineer let me grow into skills she didn’t have herself, and that experience is part of why I’m relentless about coaching now. The people I’m coaching this year can grow skills, frames, instincts that exceed mine. That outcome only happens when I stop being the bottleneck for my team’s thinking.
What the session added
Two things.
The line about cost of next mistake. I’d been operating on it by feel; now it’s in my pocket as a sentence. I’ll use it on myself before harder decisions about whether to coach or tell.
Confidence to keep doing this even when it’s slower in the moment. Coaching is slower than telling in any single conversation, and there are weeks where the slowness feels expensive. The session reminded me — clearly, with data — that the slow conversation compounds and the fast answer doesn’t. That’s the kind of confirmation I needed.
The honest close
I’m not a naturally good coach. I’m a naturally good teller, and the gravity that pulls me back into telling is constant. What I’ve built — and what the session reinforced — is a series of small interrupts that let me default to coaching most of the time. Asking a question before any answer. Counting to five before I respond. Leaving the pause open.
The team I have today is more capable than the team I had eighteen months ago, and most of that growth wasn’t a hire. It was the cumulative effect of the team being patient with me as I learned to ask instead of tell, and the team practising on each other as I modelled it. Coaching as a daily behaviour isn’t a programme. It’s a hundred small choices, made in favour of the person in front of me, until the team I have can do without me when I’m not in the room.
That, in the end, is the test of whether I’m leading them well.