The owner who tells me delegation didn't work almost always handed off the bookkeeping first. Then scheduling. Then the parts of the week that made him tired. He kept the pricing calls, the three accounts that know him by name, and the final word on anything that might embarrass the business. A year later he is working the same hours, he has a manager who approves time sheets, and he has concluded that good people are impossible to find.
Good people were never the problem. The sequence was. The owner-operator to manager transition is not a volume exercise where you push work across a desk until your calendar clears. It is a transfer of judgment, and judgment transfers in a particular order. Get the order wrong and you will hand back everything you gave away, usually within six months, usually in front of the staff.
You are handing off decisions, not tasks
Tasks are the visible part. The real cargo is decision rights — who gets to decide, within what range, without asking you. An owner who hands over a task but keeps the decision has not delegated anything. He has hired an assistant and called it a manager.
This is why the dislike-first approach fails so reliably. The work owners hand off first is usually the work with the fewest decisions in it. Payroll runs. Ordering. Filing. It feels like relief for about a month, and it teaches the new manager nothing about how the business actually makes money. Meanwhile every question with real stakes still routes to the owner, because it always did. The manager is now expensive, idle in the way that matters, and slowly learning that his job is administration.
Then something goes wrong — a customer escalation, a quote that should have been priced differently — and the owner steps in. Of course he does. The manager was never given the context to handle it. The owner concludes that the manager can't handle judgment. The manager concludes that nobody here will let him near anything real. Both are responding rationally to a sequencing error made a year earlier.
Sort by feedback speed, not by distaste
The useful sort is not "what do I hate" or even "what is below my pay grade." It is: how fast and how clearly will I find out if this was done badly, and how expensive is it to fix?
Work that recurs weekly, where errors surface within days and cost little to correct, goes first — even if you enjoy it, even if you are the best in the building at it. Quoting routine jobs. Handling the inbound calls. Scheduling crews against capacity. Running the daily standup. These are the places where a new manager builds judgment at low cost, in public, where you can watch. You want them making forty small decisions a week and getting feedback on all forty, not making four decisions a quarter and waiting for a verdict.
Work that recurs but is slow to show error — hiring, supplier selection, anything where you find out in nine months — goes in the middle, and only after the fast-feedback work is genuinely stable without you.
Work that is rare, irreversible and defines the business goes last, and some of it never goes at all. Where to put the next dollar. Whether to take on debt. Whether to open a second location. Whether to fire the manager's own peer. I have written separately about how an owner should think about the next dollar, and I hold the view that capital allocation is the owner's job for as long as the owner owns the thing. Handing it off early is not delegation, it is abdication with better manners.
The practical implication is uncomfortable. The first serious handoff should usually be a piece of work you like and are good at. That is the whole point. You are handing over something you can evaluate precisely, because you have done it a thousand times and you know what a bad version looks like within thirty seconds.
Hire the first manager into a job that already exists
Most owners hire their first manager too abstractly. They write a job description for the business they want in three years and interview against it. Then the person arrives and finds a job nobody has ever done, with no precedent, no boundaries, and an owner who is still in every room.
Hire into a job you can describe from your own calendar. Take four weeks of your actual time — not your intended time — and mark the blocks you are going to stop doing. That block list is the job. If the blocks don't add up to a real week, you are not ready to hire a manager; you are ready to hire a specialist, or an intern, or nobody.
On the candidate themselves, I weight one thing more than most owners do: has this person been managed well? Someone who has only ever worked for absentee owners has never seen the thing you are asking them to become. They will either micromanage or disappear, because those are the two models they have observed. Someone who has worked inside a well-run operation knows what a reasonable escalation looks like and what a tolerance band feels like from the inside.
And resist hiring in your own image. The common instinct is to look for a younger version of yourself — the same energy, the same instincts, the same willingness to do everything. That person will do everything, which is exactly the pattern you are trying to break. What you usually need first is a strong lead in the function that consumes most of your week, not a general manager in waiting.
Define wrong before you leave the room
Delegation without a tolerance band is a trap you have set for someone else. If you hand over quoting and say "use your judgment," you have not defined the range in which their judgment is final. So the first time they quote something you would have priced differently, you will intervene — and from their side, you moved the line after the fact.
Before you hand anything over, write down three things: what a good outcome looks like, what a bad outcome looks like, and the range of outcomes you will not intervene in even though you would have chosen differently. That third one is the hard one, and it is the one that makes the handoff real. You are pre-committing to live with decisions you mildly disagree with. If you cannot write that sentence honestly, you are not handing off; you are auditioning someone for the privilege of agreeing with you.
Much of this is easier if the recurring answers already exist outside your head. The repeated questions your staff and customers ask every day are the cheapest documentation in the business, and I have argued elsewhere for writing those answers down deliberately. A manager who inherits the answers inherits the standard with them.
The take-back, and the one rule about it
You will take something back. Everyone does. The rule is not "never reverse a decision" — sometimes the decision is genuinely wrong and the cost is real. The rule is about where you do it.
Reverse privately. Never in front of the team, never in the group chat, never in the moment with staff watching. One public reversal tells everyone in the building that the manager's authority is provisional, and from that day forward your people will route around them to you. You will then observe that the manager isn't being respected, which is true, and misdiagnose why.
When you do reverse, own the framing failure as yours. Nine times out of ten the manager made a defensible call inside a boundary you described badly. Say that out loud: I didn't give you the constraint, here it is now. Then write the constraint down so the next person inherits it. A take-back that produces a written boundary is a cheap tuition payment. A take-back that produces a quiet resentment is the beginning of the end of the arrangement.
The quiet months, and what should fill them
Nobody warns owners about the trough. Three or four months in, the handoffs are holding, the business is running slightly differently than you would run it, and you are bored and faintly useless. This is the most dangerous period in the entire transition. It is when owners invent reasons to go back into operations — a special project, a customer who "really needs me," a sudden interest in the stockroom.
Fill the vacuum on purpose, in advance. The work that only you can do is sitting right there: capital allocation, the next market, the relationships that genuinely require ownership, and the long-horizon questions you have been too busy to face. What I do at Legacy Ventures puts me across the table from owners at exactly this moment, and the ones who come through it well have usually decided beforehand what the new job is. The ones who struggle treat the empty calendar as a problem rather than as the product of a year's work. Some of the discipline worth importing here is covered in what I've written about the habits private equity imposes on an operator — chiefly the habit of asking what each hour and each dollar is actually buying.
One more honest note. Not every owner should make this transition. If the business is you — if customers buy your taste, your hands, your name — then converting it into something run by managers may destroy the thing that works. That is a legitimate choice, not a failure of ambition. The failure is making that choice by accident, through a sequence of handoffs that were always going to come back.
Done properly, the measure of our job changes completely. It stops being how much we got through this week and becomes how many good decisions got made without us. That number is harder to feel proud of, and it is the only one that scales.