Hire An Operations Manager, Or Outsource Ops?

The decision you're actually making

You've decided the current arrangement isn't survivable. Too much runs through you, the days are reactive, and the obvious answer — the one every founder in your position reaches for — is a person. Someone senior. Someone who owns the operation so you can stop owning it. You've probably already got the job description half-written in a doc somewhere.

But the question you're posing isn't quite the one you're deciding. You think you're choosing between a salary and a service. What you're actually deciding is whether your business has a capacity problem or a design problem. Those look identical from where you're standing — both feel like overload — and they have opposite solutions. A capacity problem gets fixed by adding a person. A design problem gets worse when you add a person, because you've just added a second human who has to hold context that should have been infrastructure.

Getting that distinction right is worth more than getting the vendor choice right. It's also worth more than the $70-110k you're about to commit to, because the salary is the smaller number in this decision. The year is the bigger one.

What an ops manager does well

An ops manager is in the room. They're in Slack at 9:14am when the client goes weird. They pick up the thing nobody scoped. They notice that two people on your team have quietly stopped talking to each other. That work depends on continuous presence and a place on the org chart, and no outside partner provides it.

They also accumulate. Context compounds inside a person who stays — which clients get anxious in week three, which vendor always misses, which of your promises are aspirational. And they carry authority: they can hold a deadline, tell a team member no, and make a decision at 4pm that would otherwise wait for you.

That's a real category of value, and it's the right thing to buy when your operation already exists and simply needs running. The question is whether it does.

Where the hire breaks down

Here's the failure mode we see most, and it's expensive. A founder with no defined operation hires a strong ops manager, and the manager arrives to find that every process they're supposed to own exists only inside the founder's head. So their first six months become an extraction project: sitting with you, asking questions, writing down what you do. They're doing the exact work you hired them to relieve you of, except now you're paying $70-110k plus benefits for the privilege of being interviewed by your own employee.

Then one of two things happens. Either they're strong enough to build the architecture themselves — in which case you got lucky and hired an operations designer at ops-manager rates, and they'll leave within two years for what that skill is actually worth. Or, far more commonly, they become a very expensive extension of you. They absorb the chaos rather than resolve it. The business feels marginally lighter for a quarter and then settles back, because you didn't remove the dependency, you split it across two people. When you are the system, you can't grow beyond yourself — and two people being the system is just a bigger, more expensive system with the same ceiling.

The tell is when you find yourself saying, six months in, that your ops manager is great but you still can't take a week off. That's not a hiring miss. That's a design problem wearing a hiring costume.

What we build instead of hiring a person

Ops+AI builds the operation itself, so there's something for a person to run. Not a report about it, not a recommendation — the machinery. The CRM and pipeline architecture in GoHighLevel, so a stage means the same thing to everyone and a closed deal produces an onboarding record without a human retyping anything. The delivery system that carries state across handoffs, so work in flight has a location instead of living in someone's memory. The automations that close the seams between your tools — Zapier or Make where the path is simple, n8n or Retool or a small piece of custom code where the logic has real opinions. Telemetry that pushes the yellow into Slack before it turns red. Claude or ChatGPT embedded at the points where a summary should already exist by the time a human opens the record. SOPs and documented decision rights, so your team stops waiting on you to unblock them.

The engagement has a shape. The OPERATE Report ($1,997) maps where your work actually flows, which parts are judgment and which are transport, and what's missing — before you commit to a salary. Build Days ($5K/day) construct a specific thing you can already name. A monthly retainer ($5,000+/mo, three-month minimum, five build credits) is for when the operation is tangled enough that each build reveals the next one and the sequencing itself is most of the value. Custom Builds get scoped and quoted when the thing is genuinely bespoke.

Nobody joins your org chart. That's deliberate, and it's the point rather than a limitation: we're building assets, not filling a seat. What you own at the end is infrastructure that runs whether or not we're still on the invoice — which is the only real test of whether an engagement changed anything.

Then the sequencing that makes the hire work: architecture first, then the person. Build the operation, then hire someone to run it. Do it in that order and the ops manager you eventually hire is cheaper to recruit, onboards in weeks instead of quarters, and is dramatically more likely to stay — because they're stepping into a system with defined ownership and a scoreboard rather than into a fog. They spend their first month running your operation instead of interviewing you about it. The business stops depending on your instructions and starts depending on your infrastructure.

It also changes the size of the hire. Founders regularly finish this work and hire a smaller role, or none, because a meaningful chunk of what they were about to staff turned out to be transport between two systems that nobody had connected. You don't scale by doing more. You scale by enabling more — and a salary is only one of the ways to enable.

How to tell which one you need

Write down the five things you'd hand your new ops manager on day one. Then check each one against three questions: does it have a defined process, does that process live somewhere other than your head, and could someone tell whether it went well without asking you?

If four of the five clear all three, you have an operation and you need an operator. Hire the manager. We'd be redesigning a machine that already works and charging you for the privilege. The same is true if the load is simply volume — well-defined work and not enough humans is arithmetic, and arithmetic wants a body. And if what you actually need is someone accountable in the room at 4pm on a Thursday when something's on fire, hire the person: we're not on your org chart, and no retainer is a substitute for that.

If four of the five exist only as things you personally do, don't hire yet. You'd be paying $70-110k plus benefits for someone to spend nine months reverse-engineering your head, and you'd be their bottleneck the entire time. That's not a hiring miss you can recruit your way out of — it's a design problem wearing a hiring costume, and the tell is the founder who says, six months in, that their ops manager is great but they still can't take a week off.

Your job isn't to be indispensable. But you can't delegate your way out of a system that was never designed — you can only hand it to someone else and watch it become their cage instead of yours. Build the thing first. Then hire someone to run it.