A founder I talked with last month was about to make a $180K hire. He runs a home services company doing $2.1M in revenue. He was exhausted, drowning in operations, and convinced he needed a COO. I asked him one question: what would the COO actually do in their first 90 days? He paused for a long time. Then he said, "Probably build the systems I haven't had time to build."
That's not a COO job. That's an operations manager plus two solid VAs, and it costs about a third of the price. The confusion between these two hires is the most expensive mistake I see founders in the $300K to $5M range make. Let's fix it.
The real definition of a COO
A COO runs an operating system that already exists. They don't build it from scratch. They inherit playbooks, KPIs, a leadership team, and departments with managers, and their job is to make the machine faster, tighter, and more profitable. Think of a COO as a driver of a car that already runs. If you don't have the car yet, hiring a driver is strange.
Most founders under $5M in revenue don't have the car. They have a founder who does five jobs, a bookkeeper, maybe a sales lead, and a handful of contractors. There's no leadership team to run. There's no operating cadence to optimize. A real COO walks into that environment and either builds from scratch (which is not what they signed up for) or gets frustrated and leaves in 14 months.
The math is worse than it looks. A COO at $180K base plus benefits, equity, and ramp time is closer to $240K all-in for year one. If they don't fit, you've burned that money and 12 months of momentum.
The test that tells you which hire you need
Write down every task that's currently on your plate. Every single one. Then sort each task into one of three buckets:
- Execution work. Editing videos, sending emails, updating the CRM, scheduling, invoicing, chasing collections, sourcing leads, formatting proposals.
- Management work. Reviewing someone else's output, running a weekly meeting, coaching a team member, holding a vendor accountable.
- Leadership work. Setting quarterly goals, deciding what not to do, hiring senior people, pricing changes, deciding to enter a new market.
Now count. If 70% or more of your list is execution work, you don't need a COO. You need to get execution off your plate. That's a VA problem, or more accurately, an infrastructure problem. If you have a healthy split between management and leadership, and execution is already handled, then a COO conversation makes sense.
Most founders I run this exercise with discover they're spending 60% of their time on execution, 25% on management, and 15% on leadership. A COO cannot fix that mix. Only removing the execution can.
Why "better VAs" is the wrong frame
Founders who've been burned by cheap VAs tend to think the answer is either a better VA or a full executive. That's a false choice. The real issue with most VA experiences isn't the person. It's the infrastructure around them.
Here's what usually goes wrong. You hire a VA from a marketplace. They're skilled. They start. Then you realize you have no SOPs, no way to measure their output, no backup when they get sick, no manager reviewing their work, and no clear escalation path. You end up managing them yourself, which is the exact opposite of why you hired them. Six months later you fire them and conclude that VAs don't work.
What actually didn't work was the model. A single freelance VA plugged into a founder with no systems is a recipe for frustration. What works is a vetted full-time hire supported by SOPs, quality checks, a account manager, and a bench for coverage. That's what we mean when we say infrastructure-backed talent. The person is only part of the product.
When the infrastructure is in place, three good VAs handling video editing, executive admin, and B2B lead gen can replace 30 to 40 hours a week of founder time. That's more leverage than most first-year COOs deliver, at a fraction of the cost.
The right sequence for your first operations hire
The pattern that works for service businesses in the $300K to $5M range is almost always the same. It goes in this order:
- Offload execution first. Get 20 to 30 hours a week of task work off your plate using specialized VAs. This buys you time to think.
- Document what you do. Once you have the time, spend six to eight weeks writing basic SOPs for sales, delivery, and finance. They don't need to be perfect. They need to exist.
- Hire an operations manager, not a COO. Around $70K to $100K. Their job is to run the SOPs, manage the VAs, and own weekly metrics. This person is 80% of the value of a COO at 40% of the cost.
- Consider a fractional COO if you need strategic help. A fractional COO at $6K to $12K per month can help you build the operating cadence without the full-time commitment. Good for the $2M to $5M range where you're not quite ready for a full exec.
- Hire a full-time COO when you have a real team. Usually north of $5M in revenue, with at least three department heads reporting up. Now a COO has something to actually run.
Skipping steps is where it gets expensive. Hiring a full COO at $1.5M in revenue is like buying a forklift when you need a wheelbarrow. The tool is fine. The context is wrong.
What a fractional COO actually does well
Fractional COOs get lumped in with agencies and consultants, but the good ones are different. They typically work with three to five companies at a time, spend a day or two per week with each, and focus on installing operating rhythms. Weekly leadership meetings. Quarterly planning. Clear KPIs by role. Hiring scorecards.
They're most useful when you have a small team of five to fifteen people, revenue between $1M and $4M, and you can feel that the wheels are starting to wobble. Deals are getting dropped. Clients are complaining about handoffs. You're the bottleneck on every decision. A fractional COO for six to nine months can install the operating system, and then you either keep them, promote your operations manager to run it, or eventually hire a full-time COO to own it.
What they're not good at: doing the work. If you need someone to actually run payroll, chase invoices, or manage your calendar, a fractional COO is the most expensive way to solve that problem. That's still a VA job.
A real example of getting the order right
A marketing agency doing $1.8M in revenue came to us convinced they needed a COO. The founder was working 65-hour weeks. Client work was slipping. He'd interviewed four candidates and was about to make an offer at $165K.
We ran the task audit. His week broke down like this: 22 hours on client work he should have handed off, 14 hours on admin and scheduling, 8 hours on sales calls, 6 hours on team management, 4 hours on strategy, and the rest on email and firefighting. Almost 60% of his week was execution.
He paused the COO search. He brought in a video editor and an executive admin, both full-time, both vetted, both with an account manager doing quality checks. Total cost: about $6K a month combined. Within 90 days he got 30 hours a week back. He used that time to write real SOPs and promote his best account manager into an operations lead role at $85K.
Eighteen months later he's at $3.2M in revenue with an operations lead, two VAs, and no COO. He might hire one in the next year. He also might not need to. Either way, he didn't spend $240K to find out.
How to decide this week
If you're staring at a COO job description right now, do three things before you post it. First, run the task audit above and be honest about the split. Second, ask three founders in your network who hired a COO under $5M in revenue how it went. The answers will be more mixed than you expect. Third, price out the alternative: two or three infrastructure-backed VAs plus a $90K operations manager. That stack is under $180K and solves the problem most founders actually have.
The COO conversation is seductive because it feels like the mature move. It sounds like what real companies do. But the mature move is matching the hire to the stage. Get execution off your plate. Build the machine. Then hire someone to run it.