By Andreas Pettersson, Founder, Leaders ADAPT
It is Tuesday afternoon. You handed the supplier problem to your operations lead on Monday. By Tuesday at four it is back on your desk, with a question attached. Can you just call them. You do. It takes twenty minutes. The next week there is another one.
That is the boomerang. It is one of the two reasons most CEOs book a call with me. The other is that they hate one on ones. The two are related. When delegation comes back, every one on one turns into a queue of decisions the owner thought he had already given away.
Delegating as a business owner works when you hand over a function, not a task: the outcome, the numbers, the decision rights, the weekly cadence, and a seat that owns all of it. Hand over less than that and the work comes back. Every time.
Quick answer: Delegating as a business owner fails when you hand over tasks and keep the decisions. Hand over a whole function instead: a written outcome, two or three numbers, clear decision rights, a weekly cadence, and one person who owns the seat. Do it with someone who has run that function before, alongside you, not for you. Plan on about 90 days per function, and start where you are the bottleneck.
Leaders ADAPT places fractional executives from its own bench and stays accountable for each engagement, priced as a fixed fee quoted after the first call. The research on this page belongs to its publishers, is linked with the date I read it, and their page wins if it has changed. No source named here pays us.
Why delegation boomerangs back to the owner
I have been the owner who took the work back. I built Arcules from zero to over 150 people and sold it to Canon, then ran a Canon company as its youngest standalone CEO. In the early years I delegated the way most founders do. I gave people things to do. I kept the decisions. Then I wondered why they kept coming to me.
The work does not come back because your people are weak. It comes back because you handed over the activity and kept three things.
The outcome. They know what to do this week. They do not know what good looks like in a quarter.
The numbers. They cannot tell whether the function is winning without asking you.
The authority. The first time something unusual happens, they have no right to decide, so they ask. Asking is the rational move.
Harvard Business Review published the classic on this in 1999. William Oncken and Donald Wass asked why managers run out of time while their people run out of work. Their answer was that the next move on each problem, the monkey, quietly jumps onto the manager's back. In an owner led company there is no layer to absorb it. Every monkey lands on you.
Gallup studied 143 CEOs on the Inc. 500 list and found that those with high delegator talent averaged three year growth of 1,751 percent, which was 112 percentage points higher than CEOs with low or limited delegator talent (Gallup, April 2015). In the same work, only one in four employer entrepreneurs had high delegator talent. If you find this hard, you are in the majority. It is a skill, not a personality, and it can be built.
Task delegation versus function delegation
Here is the difference I draw on a whiteboard in the first meeting.
| Task delegation | Function delegation | |
|---|---|---|
| What moves | A piece of work with a finish line | A part of the business that runs every week |
| Outcome | Done by Friday | A result for the quarter, written down |
| Numbers | You check the work | Two or three numbers on the weekly scorecard |
| Decisions | Stay with the owner | Written levels, most of them with the seat |
| Cadence | Ad hoc check ins, usually in a one on one | A fixed weekly review of the numbers, not the work |
| When something breaks | It comes back to you | The seat fixes it and reports it |
| Your time saved | An hour | A whole category of questions |
| What it builds | A helper | A second in command for that function |
Task delegation is not wrong. It is where everyone starts. Our how to delegate guide covers the task level well. The problem is that most owners stop there and wonder why they are still the bottleneck at $15M.
A function has edges. Billing and collections is a function. Hiring is a function. Delivery and customer onboarding is a function. Each has an outcome, a cost, a set of decisions and a rhythm. When all of that moves to one person, the questions stop, because the person can answer them.
The decision rights levels
Decision rights are where most handovers die. The owner says you own it, and then overrules the first real decision. After that, nobody believes the word own.
I use four levels. Write them down for every function you hand over, and name a few real decisions at each level.
Level 1. The owner decides. The seat brings information. Capital, pricing outside the band, senior hires, anything that changes what the company is.
Level 2. The seat recommends, the owner decides. The seat brings a recommendation, not a question. You answer inside an agreed number of days.
Level 3. The seat decides and tells the owner before acting. You can object inside an agreed window. If you do not, it goes ahead.
Level 4. The seat decides and reports after. It shows up on the weekly scorecard. This is where most of the decisions in a function should live within 90 days.
Bain calls the same idea RAPID, and the key letter is D, the person who decides. They describe it as the single point of accountability who commits the organization to action (Bain, 2006). The test is simple. For every recurring decision in the function, can you say in one word who has the D. If the answer is me, mostly, you have delegated tasks.
McKinsey surveyed 1,228 respondents and found that those who said decisions were made at the right level, often lower in the organization, were 6.8 times more likely to be part of a winning company (McKinsey, 2019). Our page on delegating authority goes deeper on the ladder itself.
How to hand over a whole function in seven steps
This is the sequence I run with owners. It takes about 90 days per function.
- Pick the function where you are the bottleneck, not the talent. If you are the best seller in the company, do not start with sales. Start where you get pulled in most often for the least value.
- Write the outcome in one sentence. Invoices out within two days of delivery and cash collected on terms. If you cannot write it, you are not ready to hand it over.
- Pick two or three numbers. They go on the weekly scorecard, with a target and a name next to each.
- Write the decision rights. Use the four levels. Ten real decisions, sorted. This page is the contract.
- Name the seat on the chart. One person, one box, on the accountability chart. Not shared, not a committee.
- Run it together for 30 days, then watch for 30 days. First month, the new owner runs the function with someone experienced beside them and beside you. Second month, you read the scorecard and stay out of the work.
- Step out of the weekly review for that function. Month three, the review runs without you. You get the numbers and the exceptions. If something comes back, you send it back with a sentence, not an answer.
Step six is where the outside help matters most. Jesse Sostrin wrote in HBR that one of the hardest transitions for a leader is the shift from doing to leading (HBR, October 2017). His advice includes asking how much of your involvement the person needs. In a function handover, the honest answer for the first month is quite a lot, but not from you. It should come from someone who has run that function before. That is the do it with you part. Not do it yourself in a peer group. Not do it for you and walk away.
Our founder to CEO delegation systems page shows how this sequence repeats across the company once the first function sticks.
The GM hire that failed, and what it taught
Most owners who call me have tried this once already. They grew the company by grind. At some point they hired a general manager or an operations person to take the load. It did not work. They still feel it. Some of them now want to sell rather than try again.
Here is the pattern, as a composite, no names. The GM arrives with a title and a long list of tasks. The outcome is vague. The numbers live in the owner's head. The decision rights are never written, so the GM makes a call in week three and the owner reverses it in week four. From then on, the GM checks everything. The owner sees a senior person who cannot decide anything and concludes that the hire was wrong. Sometimes it was. More often the handover was a task handover with a bigger salary attached.
What it taught me, from my own mistakes as much as anyone's: hire into a function that already exists on paper. Outcome, numbers, decision rights, cadence. Then the person has something to own on day one, and you have something to measure them against that is not your gut.
There is a second lesson the owner usually does not want to hear. No second in command lowers the price of the company. A buyer looks at a business where every function reports to the owner and sees a job, not an asset. Our founder dependency test shows where you stand in about ten minutes, and the GM vs COO vs integrator comparison helps you pick the right seat for the second attempt.
Where AI fits when you hand over a function
Three places it helps, one place it does not belong.
It helps write the function down. Most owners have never documented how billing, hiring or delivery actually work. An AI assistant can turn a recorded walkthrough of how you do it into a first draft of the outcome, the steps and the decisions in an afternoon. You correct it. That is far faster than a blank page.
It helps run the cadence. Scorecard numbers pulled automatically, a short weekly summary drafted for the review, exceptions flagged before the meeting. The new owner of the function spends the meeting deciding, not assembling.
It helps the new owner answer the questions that used to come to you. A company knowledge system built from your past decisions, your pricing logic and your customer history gives the seat a place to look before they knock on your door. For companies on an EOS style rhythm, our AI for EOS companies guide shows where this fits in the quarter.
Where it does not belong: the decision rights themselves. Who decides pricing outside the band, which customer you let go, whether the new owner of the function is the right person. Our page on what not to delegate to AI is the longer version. AI can make a handover faster. It cannot make one real.
What handing over a function does not fix
I say this early, because it saves owners a year.
It does not fix an owner who will not let go. If you reverse the first two Level 4 decisions, the function is back on your desk by month two, whatever the chart says. The work in that case is yours, and it is the first thing we work on.
It does not fix the wrong person in the seat. Clear decision rights make a weak fit visible faster. That is a good thing, but it is not a cure.
It does not fix a leadership team that cannot run a meeting without you. One function handed over is a start. A company that runs without you needs a team that holds each other to the numbers, which is the work in leadership team development.
It does not fix a company with no plan. If nobody knows what this year is for, no function owner can tell whether they are winning.
And a fractional seat is not always the answer. Sometimes the right move is a full time hire, or simply a better operations lead and a written scorecard. Our page on when fractional is wrong says so plainly.
Who is on the bench for this
Chris Geddes is a fractional COO on the Leaders ADAPT bench. Chris is an operator who pairs operations leadership with AI transformation work in owner led companies. When an owner needs to hand over operations and wants the documentation and the scorecard built on modern tools from the start, Chris is the one beside the new function owner for the first month.
Marina St. Cyr is a fractional COO on the bench as well. Marina is detail strong and runs the cadence and the follow through. When the handover fails because nobody holds the weekly review, Marina holds it until the internal owner can.
Steve Shaw takes the seat when the function is finance, or when the operating problem is finance led: cash, margin, pricing and the cadence around them. Steve has twenty years in finance leadership.
Steve Prodger takes the seat when the function is revenue. Steve builds revenue engines, pipeline discipline and sales leadership for owner led companies, which is usually the last function an owner hands over and the one that most needs a structure first.
Every engagement runs through Leaders ADAPT. I take the first call, match the partner, and stay in the room. Our fractional executive services hub shows every seat, and how fractional engagements work covers the shape week by week. Our fractional COO page covers the operations seat in detail.
Delegating as a Business Owner FAQ
Why does delegation keep coming back to the owner?
Because most owners delegate a task and keep the decision. The first time something unusual happens, the person has no authority to act, so the question walks back into your office. Do that fifty times a month and it feels like nobody can do anything without you. The fix is to hand over the outcome, the numbers and the decision rights together, not the activity.
What is the difference between delegating a task and delegating a function?
A task is a piece of work with a finish line, such as sending the invoices this week. A function is a part of the business that runs every week, such as billing and collections, with an outcome, a number, a budget and a person who owns all three. Task delegation saves you an hour. Function delegation takes a whole category of decisions off your desk.
Which function should a business owner hand over first?
Hand over the function that pulls you in most often and costs you the least sleep to give away. For most owners that is operations or delivery, sometimes finance. Do not start with sales if you are the best seller in the company. Start where you are the bottleneck, not where you are the talent, and keep the function you are best at for later.
How long does it take to hand over a whole function?
Plan on about 90 days for one function. The first month is writing the outcome, the numbers and the decision rights. The second month the new owner runs it while you watch the scorecard, not the work. The third month you step out of the weekly meeting for that function. If it takes much longer, the seat or the person is usually wrong.
What if I tried a general manager and it failed?
Most owners who call me have tried once. The usual cause is that the GM got a title and a long list of tasks, but not the outcome, the numbers or the authority to make calls without the owner. That is not proof the company cannot run without you. It is proof the handover was a task handover. The second attempt should start with the function, not the hire.
Do I need a full time hire to delegate a function?
Not always. A fractional COO can own a function one to three days a week, build the cadence, and hand it to an internal person once it runs. That is often safer than a full time hire into a function nobody has defined yet. Leaders ADAPT places that seat from its bench, and the fee is quoted per engagement after the first call.
How do I stop people bringing me decisions I already handed over?
Ask one question every time. What do you recommend, and what level is this decision. If the decision sits at a level you handed over, send it back with a sentence of support, not an answer. It feels slow for three weeks. After that, people stop asking, because the answer is always the same, and the decision rights become real.
Start with one function, not ten tasks
Owners who want to stop being the bottleneck usually make a list of twenty tasks to give away. Do not. Pick one function. Write the outcome, the numbers and the decisions on one page. Then hand it over with someone beside the new owner who has run it before.
Tomorrow morning, do one thing. Count the questions that come to you before lunch. Write down which function each one belongs to. The function with the most questions is the first one to hand over.
What the first conversation looks like
It is a 30 minute call with me, not a sales rep. Bring the list of questions you counted and the name of the person you would like to own that function. I tell you whether the gap is the outcome, the numbers, the decision rights or the seat, and whether a fractional COO or another seat from the bench is the right first move. If the answer is a written scorecard and a better weekly meeting, I say so. Our Fractional Executives and Advisory page shows the shapes we work in.
If we go ahead, the partner joins the second call, we agree the function, the 90 day outcome and the days per week, and the fee is fixed and quoted after that call. I stay in the room for the length of the engagement.
Your step for tomorrow morning: take the free assessments so I can see where the company depends on you before we talk. Then book a fit conversation. Handing over the work is the easy part. Handing over the business is what the call is for.
