Hire a COO when the business works but every routine decision still runs through you, and fixing that needs someone with the authority to redesign how work moves, not just someone to follow up. In a growing business, that person is usually fractional first: senior, part of the week, with a defined exit. A full-time COO comes once the systems exist and need a permanent owner.
I am a fractional COO, so read this knowing which way I lean. That is also why the first step below costs nothing and involves nobody but you: often the honest answer is a written procedure or an operations manager, not a COO.
Run the one-week test before you hire anyone
Most advice on when to hire a COO lists feelings: overwhelmed, stretched, stuck. Feelings do not tell you which hire fixes it. A week of evidence does.
The decision log, 5 working days
- Keep one running note on your phone. Every time someone asks you something, waits on you, or you step in to fix something, write one line: what it was and who asked.
- At the end of each day, mark every line with one letter. I (information): you knew something they did not. A (authority): they knew, but nobody had told them they could decide. B (behavior): the rule exists and was not followed.
- Add a star to every line that could have been settled by a written rule.
- On day five, count the letters and the stars, and add up the time.
How to read it: mostly B means a follow-through problem, and an operations manager fixes it. Mostly I and A, with stars on most lines, means nobody has designed how the business runs, and that is a COO's job. A short log means you do not need either yet.
These are the same three buckets I use in every Operations Leak Audit. Each one has a different fix: information needs a system that shows the right person the right thing in time, authority needs written decision rights, and behavior needs an owner and a weekly check.
The 7 signs a service business needs a COO
1. Every routine decision routes back to you
Not the strategic ones. Those should. The weekly ones: schedules, approvals, quotes, complaints, refunds. If your team cannot act on things that happen every week without asking you, the context lives in your head, and it stays there until someone extracts it, writes it down and builds it into how work moves.
2. You are still the dispatcher
In field and home services the schedule is the operation. If you build tomorrow's calendar tonight, you are doing a coordinator's job with an owner's hours. At one field-service company I work with, the schedule built on the day gave way to one set two weeks ahead, on routes. How that works is in field service scheduling and dispatching.
3. You hired more people and it got harder
More headcount scales a broken process instead of fixing it. If you hired three people and now have three more people waiting on you, the constraint is structure, not capacity.
4. Good people are doing average work
When I map how work actually moves through a team, the gap between the process the founder describes and what happens is usually large. People are not ignoring the process; there was never a written one. The standard is implicit, so the feedback is "this was wrong" instead of "here is the right way".
5. You cannot take a week off
The clearest test. If you went fully offline for five working days, what would break? "Nothing critical" means you do not need a COO yet. "Several things" means the business depends on you in ways no hire will fix by being busy.
6. You cannot say what a job or a customer earns
Revenue is on the dashboard. Margin by job is not. At one field-service company I work with, crew payroll ran above 80% of revenue. Scheduling by route, assigning work by where each crew member is based and removing a redundant management layer took it to 46% in one quarter. Until then, nobody owned that number.
7. Revenue is growing and the operation is getting harder
Growth should buy room. If the business grows and gets harder to run at the same time, the systems are not scaling with the revenue, and one bad month or one key resignation exposes it.
If two or more of these are true and your decision log is mostly I and A, the constraint is operations, and the question becomes which hire.
Which hire fixes which problem
| Option | Fits when | Watch for |
|---|---|---|
| Nobody yet | The log is short, or the business is before steady revenue | Building structure before you have customers slows you down |
| Fixed-scope project | One process is broken: quotes, invoicing, the schedule | It fixes one thing; nobody owns the weekly rhythm after |
| Operations manager | Processes are written and owned; the problem is follow-through (mostly B) | Without written rules, every question still comes back to you |
| Fractional COO | The business works, depends on you, and little is written down (mostly I and A) | Needs real authority and access, or it becomes a consultant with a longer contract |
| Full-time COO | The systems exist and the operation needs a daily executive owner | Expensive to get wrong; hire after the systems, not instead of them |
The order matters. A full-time COO hired into a business with nothing written down spends the first year doing what a fractional COO would have done in a quarter, at full-time cost. An operations manager hired into the same business ends up asking the founder, and the bottleneck stays exactly where it was.
Worked example: how the test reads in a pest and lawn company
To show a system without client data I use Quillmoor Operations, a fictional operator from my demo workspace. Its fictional client, Briarhollow Pest & Lawn, has 412 recurring accounts, four technicians and an office manager who runs the phones and the schedule. The owner decides pricing and hiring.
Run on Briarhollow, the log reads mostly I and A. Estimates took about three days to go out. The Thursday pest route carried 38 stops while Tuesday carried 22, and the decisions to fix either sat with the owner. Little of it was B: the team followed the rules that existed. There were just not many.
He did not hire a full-time COO. He brought in a fractional operator for 10 to 14 hours a week with three jobs: estimate turnaround, recurring routes and technician scheduling. Pricing and hiring stayed with him. Estimates now go out the same day or the next morning, and the Thursday route is being fixed by moving stops, not by adding a truck.
What a fractional COO should change in 90 days
Judge the hire on numbers that move, agreed before the start. From my own engagements:
- At a six-location seasonal operation, the owner's involvement went from over 10 hours a day to 30 minutes a week in 90 days.
- At a field-service company, crew payroll went from over 80% of revenue to 46% in one quarter, and the owner's daily involvement dropped about 90%.
Ask any candidate the same three questions: what will you build in the first 30 days, which number moves from what to what, and what will my team be able to do without you at the end? For the role itself, read what a fractional COO does.
When not to hire a COO at all
- Before steady revenue. You need customers, not infrastructure.
- When you will not hand over authority. A COO without decision rights and access to the numbers cannot change anything.
- When one process is the problem. Buy the fix for that process. A retainer is the wrong shape.
Questions
At what size does a company need a COO?
Size is a weak signal. A business with eight people can need one if every decision routes through the founder, and a business with forty may not if the operation is already written down and owned. In a growing business, the trigger is usually the founder's time, not headcount: when the routine work of running the operation fills the day, the business needs someone who owns that layer.
What is the difference between a fractional COO and a full-time COO?
Hours, cost and exit. A fractional COO works part of the week, owns a defined set of systems and is measured on whether the team can run them without help. A full-time COO runs the whole operating side every day and makes sense once the systems exist and need a permanent owner. Many businesses need the first before they can use the second.
Should I hire an operations manager or a COO?
An operations manager runs the system that exists. A COO designs the system, decides what gets measured and changes it when the numbers say so. If your processes are written down and the problem is follow-through, hire a manager. If nobody has written down how work moves through the business, a manager will end up asking you, and the bottleneck stays.
How many hours a week does a fractional COO work?
It depends on the scope, and it should be set by the systems to be built, not by a block of hours. I carry two or three clients at once so each gets real time, with daily contact with the team and a weekly review with the owner.
Can a fractional COO become full time?
Sometimes. More often the fractional engagement builds the systems and trains the people who run them, and the next hire is an operations manager, not a full-time executive. Decide at the end of the first 90 days, on what changed.
Not sure your log points to a COO? Bring it to a 45-minute Operations Leak Audit. You leave with a one-page leak map and the one build I would do first, whether or not we work together.
Book the Operations Leak Audit
Wilkins NdwigaI build AI systems and automations for growing businesses in any industry, and I work beside founders as their fractional COO. I build the systems myself. More about me.
