The ratio is easy. Knowing when your report is ready for it is the actual skill.

Ask a struggling new manager what they did yesterday.

If most of the answer is their direct reports' work, you've found the problem. They got promoted four months ago. They're still operating like the senior IC they used to be, except now with 1:1s stacked on top.

There's a faster version of this diagnostic that Fahd uses in our leadership programs, and it's brutal in a useful way:

If you had to take an unexpected week off work, would your projects, initiatives, and priorities advance in your absence?

Sit with the honest answer. Most managers, especially in their first year, know the projects would stall. And they know it because they've built it that way, one reasonable decision at a time.

‍

The Confusion Underneath It

Here's the thing about the managers who hold onto work. They're not lazy, and they're usually not control freaks. They're confusing two things that feel identical from the inside.

Being essential and being involved.

You need to be more essential and less involved. That sentence sounds like a contradiction until you've watched a good manager operate. Their fingerprints are on the outcome, the standard, and the decision. Their fingerprints are not on the document.

When you justify holding onto work, you're almost always defending your involvement and calling it essentialness. The two are not the same thing, and the gap between them is where a manager's ceiling gets set.

John Maxwell calls this the Law of the Lid: organizations and teams grow only to the capacity of their leaders. You are the lid. Your team cannot outgrow your capacity, so the only way the ceiling rises is if you extend your reach through other people. Which means the work of delegation isn't administrative. It's the whole job.

‍

Three Ways It Goes Wrong

1. ‍The Re-Doer. Manager assigns a task. Direct report ships it. Manager rewrites 40% before sending it on. The manager thinks they're maintaining quality. What the team learns is: don't try too hard, she'll change it anyway. And that lesson compounds. Effort drops on the next task, which confirms the manager's belief that the work needs rewriting, which drops effort further. Six months in, the manager is doing more work than before the promotion and has evidence for why she has to.‍
2. The Ghost. Manager assigns a task, says "let me know if you need anything," and disappears. Direct report ships something that misses the brief by 30%. The manager reads this as a capability problem. It's usually an information problem. "Let me know if you need anything" puts the burden of knowing what they don't know on the person who, by definition, doesn't know it yet.‍
3. The Helicopter. Manager assigns a task, checks in three times on day one, sends two Slack messages with "small thoughts," schedules an alignment meeting on day three. Every touch is individually defensible. Together they communicate that the manager doesn't believe the work will be done right without supervision, which is both demoralizing and, eventually, self-fulfilling.

Different symptoms, and the standard diagnosis is that these managers don't know which part of the task to be in. That's half right.

‍

The Real Failure Is Speed, Not Willingness

Most delegation advice assumes managers hold on because they won't let go. In our work with scaling teams, that's rarely what's happening. The more common failure is moving people through the stages of empowerment too fast, watching it break, and concluding the person can't handle it.

People move through four stages, and each one asks something different of you:‍

1. ‍Controlled. You tell them exactly what to do. This is not micromanagement when someone is genuinely new to a task, and the fear of being called a micromanager makes a lot of managers skip it entirely, which sets everyone up to fail. This is the coach explaining the drill.‍
2. Managed. You set priorities together and give them a checklist. They have space inside a defined structure. The coach has given the lesson and sent them to practise the specific skill.‍
3. Coached. They make their own decisions, and you give feedback on those decisions afterward. Go play the game, record it, and we'll review the tape.‍
4. Empowered. They decide, they tell you what's best, and you support it.

Here's the failure that costs the most. A founder hires their first VP at around 25 people, pays them well, and reasons that a VP doesn't need managing. So they hand over full autonomy on day one. The VP, who has never worked in this company, makes a reasonable call that's wrong for this context. And that first senior hire almost always ends up gone within a year, with the founder concluding it wasn't a good fit.

It wasn't a fit problem. Nobody managed the VP. Seniority changes how fast someone can move from Controlled to Empowered. It doesn't mean you skip the journey.

The same mistake happens at the task level, every week, in smaller ways. You give someone a project at the Empowered level when they're at Managed, it comes back wrong, and you take the next three back entirely. That yank backward is the actual damage. It's harder to recover from than never having delegated at all.

‍

The 10-80-10 Framework

So when should you use 10-80-10? At the Coached stage, mostly, and increasingly as someone moves toward Empowered.

The framework itself isn't ours. It's a delegation heuristic that circulates widely in management circles, popularized by John Maxwell among others. What makes it worth teaching is that it's the most repeatable structure we've found for the Coached stage, and it survives contact with real work.

The first 10% is yours. You scope the task: what outcome, by when, under what constraints.

The middle 80% is theirs. Your job is to be unblockable, not present.

The last 10% is yours again. You review, polish, and broadcast.

That last question changes everything downstream. A report who knows the analysis feeds a hiring decision makes different judgment calls than one who thinks they're producing a document. It's the difference between someone executing a task and someone solving your problem.

Write the brief before the conversation, not during it. If you can't answer all five in writing, you haven't finished the first 10%, and no amount of checking in during the middle 80% will fix that.

‍

The middle 80%: be unblockable, not present

Schedule one checkpoint at the midpoint, in advance. "Let's sync Wednesday at 2pm for fifteen minutes." Then leave them alone.

The checkpoint has to be scheduled at the start, not conjured on day three when you get nervous. A checkpoint you announce mid-flight reads as a loss of confidence, because it is one.

If something's off at the checkpoint, describe what you saw and what it caused, not what kind of person they're being. "The exec summary opens with methodology, and the three people reading it will stop at paragraph two" is usable. "You're getting lost in the weeds" is a character note they can't act on.

‍

The last 10%: polish, don't rewrite

When the draft comes back, two questions before any edit:

Is this changing the substance, or my style? Style edits are how the Re-Doer is born. If the outcome is achieved and it doesn't sound like you, that's not a defect. That's what delegation looks like.

Will this edit teach them something next time? If yes, make it and explain it. If no, and it's cosmetic, leave it.

Then the test: if 80% of the document is unchanged from what your report sent, you delegated well. If 40% carries your fingerprints, you've recreated the founder bottleneck one layer down, at the manager level, which is where it does the most damage because it's least visible.

‍

When Not to Delegate

Every blog on this topic tells you to delegate more. Almost none of them tell you when not to, which makes the advice feel absolutist and easy to dismiss. We teach a decision tree, and it starts with time.

Will this take longer to delegate than to do myself? If no, delegate it. Done.

If yes, ask: is this a repeating task? If it isn't, just do it. A one-off that costs more to hand off than to complete is not a delegation opportunity, it's a task. Do it and move on.

If it repeats, the calculation flips. The teaching cost is paid once and the return arrives every time it comes around again.

Then the question that matters most: can they get it to 70% of your quality?

If yes, delegate it. Seventy percent from them beats one hundred percent from you, because their seventy climbs toward ninety with reps, and your hundred stays exactly where it is while consuming your week. This is the number most managers refuse to accept, and refusing it is the single most expensive habit in new management.

If they can't reach 70%, one more question: is this a core competency of their role?

If it isn't, do it yourself. Genuinely. Not everything is a development opportunity, and forcing someone to slowly fail at something outside their job is not coaching.

If it is a core competency and they can't hit 70%, you've found a teaching problem, not a delegation problem. Walk them through it. Build the skill deliberately. Then delegate it next quarter.

The distinction matters because managers routinely delegate things that should be taught and teach things that should be delegated.

‍

What It Sounds Like When It's Working

David Marquet spent his career on nuclear submarines and wrote Turn the Ship Around! about turning the worst-performing boat in the fleet into the best. His ladder of leadership, developed with Stephen Covey, has seven rungs, and they're all phrased as things a person says to their boss:

Tell me what to do → I see → I think → I would like to → I intend to → I've done → I've been doing

The bottom of the ladder is someone waiting for instruction. The top is someone informing you of work already underway.

The target is rung five. "I intend to."

That's where empowerment is high and communication is still happening, which is the combination you want. Above it, you lose visibility. Below it, you're still making decisions that shouldn't be yours. Marquet's own example: instead of waiting for the order to dive, his officers would say "I intend to dive, we have everything squared away and the crew is prepared," and he'd answer "very well."

The action hadn't happened yet. He could still stop it. But he almost never needed to, and the officer had done all the thinking.

This ladder pairs with the Empowerment Spectrum because they describe the same journey from opposite sides. The spectrum is what you do. The ladder is what they say. Someone at Controlled says "tell me what to do." Someone at Coached says "I would like to" or "I intend to." Someone at Empowered says "I've been doing."

Which gives you a free diagnostic. Listen to how your reports open requests. The language tells you where they actually are, and it's usually more honest than where you assume they are.

‍

The Three Things Managers Say Instead

"It's faster if I just do it." True for one task. Catastrophic across ten. Every task you take back is a rep you stole from someone who needed it, and you'll pay for it in month eight when you're the only person who knows how anything works.

"I don't trust the output yet." Then your brief was incomplete, or you're delegating at the wrong stage. Trust is built by watching someone demonstrate competence, and you can't watch that happen if you keep doing the work. The trust you're waiting for is downstream of the delegation you're avoiding.

"I'll look like I'm not contributing." This is the honest one, and it's usually said quietly. Multiplying your team's output is the contribution. The job changed at promotion. Sometimes the identity takes another year to catch up, and that's normal, but the work can't wait for it.

‍

Try This Monday

Pick one task on your plate this week that a direct report could take to 70% of your standard. Write the five-question brief. Send it in one message. Schedule one fifteen-minute checkpoint at the midpoint. Then leave them alone until then. When it comes back, count what you'd change. If it's under 20%, you delegated well and you now have a rep in the bank. If it's over 40%, go back and look at your brief, because that's almost always where it broke.

Then ask yourself the week-off question again in ninety days. The answer is the only performance review of your delegation that matters.

‍

Frequently Asked Questions: