Delegation: the leadership competency managers fake
Almost every manager thinks they delegate well. Almost none do. What real delegation looks like in behavior — and how to find out which one you are.
On this page
Delegation is the competency with the widest gap between self-image and reality of any on the list. Ask managers whether they delegate well and nearly all say yes. Watch what actually happens and you see something else: tasks handed over with the judgment kept, work quietly redone because it "wasn't quite right," problems that a report brought in and the manager somehow left holding. Everyone believes they delegate. The behavior tells a different story, and because the belief is so sticky, delegation is less a competency people lack than one they fake — often without knowing it.
That gap is the whole reason to break it into behaviors. "Delegate more" is advice every over-loaded manager has heard and nodded at and not acted on, because it names a virtue rather than a move. This is the pillar's deep-dive pattern applied to the competency where the self-image is furthest from the truth: what does real delegation actually look like, moment to moment, and how would you find out if yours is the real thing or the performance of it.
What delegation is actually made of
For a leader, delegation decomposes into a handful of observable behaviors — observable to the person on the other end of it, even when they're invisible to you. As the pillar's manifesto argues, that's the only level at which the competency can be developed or judged.
Handing over the decision, not just the task. The faked version delegates execution and keeps the judgment: do this, this way, and check with me. That's assignment, not delegation. Real delegation hands over the outcome and the authority to decide how — which is the part that feels risky, and the part that actually develops the other person. The observable difference: does the person get to make calls, or just do steps?
Not taking the monkey back. In their HBR classic Who's Got the Monkey?, Oncken and Wass describe the exact moment delegation reverses: a report brings you a problem, you say "let me think about it and get back to you," and the next move — the "monkey" — has just jumped from their back to yours. Do that across a team and you end up personally holding everyone's next steps while they wait on you. The behavior to watch is brutally concrete: after someone brings you a problem, whose calendar has the next action on it? If it's yours, you didn't delegate — you collected.
Delegating with context and outcome, not instructions. Handing over a task with a list of steps guarantees you'll get exactly what you'd have done, minus the thinking. Handing it over with the why and the what good looks like lets the person find a route — possibly a better one. The weak version calls the instruction-list "being clear" and gets compliance instead of ownership.
Tolerating it being done differently — and, at first, worse. The redo reflex is the quiet killer. Someone does the thing at 80% of your version, you fix the last 20% yourself "just this once," and you've taught them that ownership is fake and taught yourself that delegating doesn't save time. The behavior is letting the 80% ship when 80% is enough, and coaching the gap instead of closing it silently.
Following up on the outcome, not hovering on the method. Real delegation still checks — on results, at agreed points. Faked delegation either abandons (hands off and vanishes) or hovers (asks about method daily). The observable middle is whether your follow-up is about what happened or about how they're doing it the way you would.
Why you can't tell, from the inside, whether yours works
The self-image problem here is unusually strong because delegating feels like generosity while you're doing it — you gave them the project, didn't you? What you don't feel is the redo you did afterward, the decision you quietly reserved, the daily "how's it going" that reads to them as distrust. You experience your intent to delegate; they experience whether they actually got to own anything.
This is why delegation sits squarely among the skills you can't develop alone: the evidence is entirely in whether your team feels trusted with real decisions, and that lives in their heads. It's also why the manager self-assessment puts its sharpest questions here — what did you delegate and then secretly redo? — because the honest answer is the one your self-image is built to hide.
How to actually develop it
You develop delegation through the loop, not a resolution to "let go more." Pick one behavior. If your gap is the monkey, the one behavior might be: when someone brings me a problem, end the conversation with the next action on their side, not mine. If it's the redo reflex, it might be: ship their version at 80% and coach one thing, instead of fixing it myself.
You pick one because "delegate better" is exactly the kind of advice that's been failing you — it names the destination, not the step. "Hand the next move back in this week's 1:1s" is a rep you can run and, more importantly, one your team can feel, which means it generates the feedback that tells you whether it's working.
A worked example: from "I delegate a lot" to actually letting go
Say you're sure you delegate — your calendar is proof, you hand out work constantly — and a round comes back with your team rating your delegation low, with notes about not feeling trusted. The instinct is bafflement: you delegate all day. Find the behavior instead: you delegate the tasks and keep the decisions, so people execute a lot and own nothing, and they can feel the difference even if you can't name it.
The one behavior becomes: on the next real project, hand over the outcome and one genuine decision — and don't reserve the call for myself. The cue is any assignment big enough to matter. It's uncomfortable; the first time someone decides something differently than you would have, you'll want to step in. You re-ask a couple of months later whether they feel trusted with real ownership. If the read moves, the competency moved — through one behavior, not a philosophy of empowerment.
See where yours actually stands
The starting move is the same as for any competency: honest self-answers on these behaviors, then the same questions answered by the people you delegate to. The delegation and trust template is built for exactly this — you answer first on how well you think you hand off, send the same questions to your team, and read the columns side by side. The gap between "I delegate a lot" and "I never get to actually own anything" is the whole finding, and it points straight at the one behavior to change.
That same decompose-then-measure approach works for any competency. Delegation is just the one where the self-image is most flattering and most wrong — which is exactly why checking it against the people on the other end is worth the discomfort.
Delegation, in one place
- Almost every manager thinks they delegate well; almost none do. The belief is sticky, so delegation is a competency people fake rather than lack.
- The behaviors: hand over the decision not just the task, don't take the monkey back, give context and outcome instead of instructions, tolerate it being done differently (and worse at first), follow up on results not method.
- Delegating feels like generosity while you do it, which hides the redo, the reserved decision, the hover. Only your team knows whether they got to own anything.
- Develop it through the loop: pick one behavior, attach a cue, run it in real handoffs, re-measure with the people you delegate to.
- The fastest tell: after someone brings you a problem, whose calendar has the next move? If it's yours, you collected — you didn't delegate.