Coach everything and your team drowns while you ask thoughtful questions.
Direct everything and they stop thinking entirely.
Both failure modes are real, both are common, and both come from the same mistake: picking one mode and calling it your style. Coaching and directing aren’t personality types. They’re tools, and the actual management skill is knowing which moment calls for which.
Here’s a working filter you can run in five seconds, plus the traps on both sides.
The Two Failure Modes
The all-coach manager read the right books. Every question gets a question back. Every decision becomes a development opportunity. It feels enlightened, and it works right up until something is on fire and an employee needs a call, not a journey. Ask their team privately and you’ll hear it: “I just need an answer sometimes, and I can’t get one.”
The all-direct manager is faster. Every problem gets an answer, every decision stays at the top. It feels like leadership, and the short-term results even back it up. Then the manager goes on vacation and the team stalls, because nobody has practiced deciding anything in two years. We wrote about that trap in How to Coach Employees Instead of Giving Them Answers: every answer you give is a rep you took away.
Same root cause. One mode, applied to every situation, regardless of what the situation needs.
What Each Mode Actually Is
Worth defining, because the middle ground is where most managers actually live, and it’s worse than either pole.
Directing is deciding and communicating. What needs to happen, why it matters, when it’s due. Clean, honest, and owned: you made the call, and everyone knows you made it.
Coaching is developing their thinking. They keep the problem, you supply the questions, and they leave with their own plan.
The mushy middle is the suggestion disguised as a question. “Have you considered maybe looping in legal first?” That’s a directive wearing a costume. Your employee knows it, plays along, and learns exactly one thing: guess what the manager wants. It has the slowness of coaching and the dependency of directing. If you’re going to direct, direct. If you’re going to coach, mean the questions.
The Four-Question Filter
Run these in order. The first “yes” usually tells you the mode.
1. Is the cost of a slow or wrong decision severe and immediate? Real emergencies, real deadlines, real safety or client-loss stakes: direct. A learning rep is never worth a burned client. But be honest about the word “real.” If everything is urgent, nothing is, and permanent urgency is usually a directing habit wearing a deadline.
2. Is this a knowledge gap or a judgment gap? Facts, processes, history, the compliance rule: tell them. Facts aren’t coachable, and making someone guess what they cannot possibly know isn’t development. It’s hazing. Judgment calls, tradeoffs, priorities, people problems: coach.
3. Does this person have enough context to reason with? A first-week hire can’t be coached into a decision that requires a year of context they don’t have. Direct now, coach in month three. Calibration is per person and per task: the veteran engineer might need directing on stakeholder politics and coaching on architecture. Same person, different gaps.
4. Will they face this situation again? Repeating situations are where coaching pays compound interest. A one-off weird edge case? Just answer it and move on. The fifty-times-a-year situations, prioritization, pushback, scope creep, are the ones worth building their judgment on.
Five seconds, four questions, and most moments sort themselves cleanly.
The Same Problem, Two Right Answers
Here’s the filter working on a real situation, twice.
Monday, 4pm. Your analyst says: “The client found an error in the report we sent this morning. What do I do?”
Run the questions. Severe and immediate? Yes: a client is holding a wrong number with your firm’s name on it. Mode: direct. “Call them in the next hour. Own it, no hedging. Tell them the corrected version arrives by nine tomorrow, and copy me on the follow-up note.” Clean call, clear reasons, done. Nobody needed a growth journey at 4pm on client-error day.
Now the same analyst, three weeks later: “I keep finding small errors in the reports before they go out. I think our review process is broken. What should I do?”
Run the questions again. Immediate stakes? No: nothing is burning, this is process improvement. Knowledge or judgment? Judgment. Context? They found the pattern themselves, so plenty. Recurring? Every week, forever. Mode: coach. “Good catch. What do you think is breaking down? What have you seen other teams do? Sketch me a fix by Friday and walk me through it.”
Same employee, same subject matter, opposite modes, both correct. That’s the whole idea. The situation picks the mode, not your personality, not your workload, and not which management book you read last.
Notice one more thing: the directing moment earned you credibility for the coaching moment. A manager who can make a crisp call under pressure is a manager whose questions get taken seriously later. The modes aren’t competing. They fund each other.
Directing Well Is a Skill, Not a Failure
Somewhere along the way, “directive” became an insult in management writing. It shouldn’t be.
Your team wants you to make some calls. It’s why the role exists. What they resent isn’t direction. It’s direction done badly: decisions without reasons, reversals without explanation, hovering after the handoff.
A clean directive has three parts. The what: specific and unambiguous. The why: one or two sentences of reasoning, because people execute better when they understand the logic, and because the why is itself development. The when: a real deadline, not “soon.”
Then step back. Directing the outcome and micromanaging the method are different things, and the second one cancels the first.
One more mark of directing done well: it comes with an expiration date. “I’m deciding this one because we’re new to each other. In six months, calls like this are yours.” Now even your directives are development, because they carry a map of where the authority is heading.
Coaching Well, in Brief
The full toolkit is in the coaching article, but the short version: real questions, real silence, their plan, their ownership. If you talked more than a third of the time, it was a lecture with pauses.
One addition specific to mode-switching: coaching only works when they know that’s what’s happening. Which brings us to the most useful habit in this whole area.
Name the Mode
The simplest upgrade available: say which mode you’re in.
“I’m going to just make this call, it’s time-sensitive. Here’s what we’re doing and why.”
“This one’s yours. I have opinions, but you’re closer to it than I am. Walk me through your thinking.”
Naming the mode does three things. It kills the guessing game, because your employee isn’t left wondering whether your question is a real question. It makes directing feel fair instead of arbitrary, because it’s visibly an exception with a reason. And it builds a shared vocabulary: eventually your people start asking for the mode they need. “Can you just decide this one?” and “Don’t tell me, let me work it” are both signs of a team that understands how development works.
Quick test: would your team say your questions are real questions? If you’re not sure, ask them. That conversation is worth ten management books.
The Calibration Nobody Talks About
The filter isn’t static, because people aren’t.
The goal over time is a steady handoff: situations you directed last year become situations you coach this year become situations you don’t hear about until the update. If the ratio isn’t shifting toward their ownership, something is stuck. Either you’re holding decisions out of habit, or the development conversations aren’t happening.
Watch for drift in the other direction too. Managers under pressure quietly slide back into full directing, because it’s faster this week. Your calendar will tell you: if every decision is routing through you again, the slide already happened. Name it, apologize for nothing, and hand the decisions back.
And notice which people you never coach. It’s usually the strugglers, on the theory that they can’t handle it yet. Flip that. The person you keep directing is the person who most needs reps, with lower-stakes situations if necessary. Permanent directing isn’t protection. It’s a verdict, and they can feel it.
Where to Build Both Modes
The judgment between modes, and the coaching skills that make one of them work, is exactly what our Leadership Coaching Mindset and Leadership Coaching in Practice courses build. Both are part of Tandem Academy: $1,000 a year for all nine courses, an AI coach for weekly practice on situations like these, and live small-group coaching where you can bring the employee you can’t figure out. The coaching is built in, not sold separately. Not ready for that? Start with the free book on the Academy page.
Direct when the moment needs an answer.
Coach when the person needs a rep.
The job is knowing the difference.