Integrator vs Project Manager in an Agency

Integrator and project manager get confused for a specific reason: both titles sound like “the person who makes things actually happen,” and an agency hiring in a hurry reaches for whichever word it heard most recently. I offer both seats, so declaring the bias matters: twelve years running delivery operations for marketing agencies and 92 completed engagements are the basis for what follows, not a guess.

What is the actual difference between an integrator and a project manager?

Ask what each person is holding, not what their job title says.

• Project manager. Holds a project. It has a client, a scope, a start date and an end date, and the PM is judged on whether it ships inside that box.
• Integrator. Holds the plan. Not one project’s plan, the owner’s plan for the whole business this quarter, including everything on it that never had a client attached in the first place.

That second half is the part agencies miss. A quarter’s plan is rarely only client projects. It usually has a hire that keeps slipping, a pricing change nobody has actioned, a new service that has been “next quarter” for three quarters running, a process rebuild, a vendor decision that keeps getting punted. None of that is a client project. All of it is still the plan, and the integrator owns whether that whole list moves, not whether any single item on it has a status update.

What does a project manager own that an integrator does not?

Everything with a client’s name on it. Scoping the work, running the weekly rhythm on that job, catching risk before it becomes a client-visible surprise, and reporting status whether or not there is good news in it. What a PM’s week looks like in more depth is covered in the comparison against an operations manager, and the shape of the seat itself is on my outsourced project management page.

The boundary is simple. If it does not have a client, a scope and an end date, it is not on a project manager’s desk. That boundary is exactly where the integrator’s job starts.

What does an integrator own that a project manager does not?

Whatever the owner decided mattered this quarter, tracked whether or not a client is watching.

• Turning the plan into owned work. The quarter’s priorities get named tasks, named owners and dates, the same discipline a PM applies to a client project applied to the business itself.
• The meeting rhythm for the whole business. Not one project’s stand-up, the cadence where the owner’s actual priorities get checked against what has actually moved.
• Chasing what stalls with no client to notice. A slipping client deliverable gets caught because the client asks. A stalled hire, an unbuilt process, or a service line that never launched has nobody asking, which is exactly why it stalls.
• Reporting the unflattering version. Not “on track,” the actual state of the plan, including the parts that have not moved in six weeks.
• Holding the owner accountable too. Half of what stalls a plan is the owner’s own follow-through, and an integrator is one of the few hires whose job includes saying so.

The term comes out of the Entrepreneurial Operating System, and I have written up that origin, and the seat it sits next to (a COO and an operations consultant), in the three-seat comparison. You do not need to run EOS to hire one. The job of turning a plan into shipped work predates the vocabulary. What the week looks like day to day is on my fractional integrator page.

Which symptom points to which?

Signs you are missing a project manager

• Clients chase you for status instead of the other way around.
• A deadline slips and the first you hear of it is the client, not your own team.
• You are the only person who can say where every live job actually stands.

Signs you are missing an integrator

• Client projects ship fine, and the business looks the same as it did last quarter anyway.
• The same three items sit on every quarterly list, unshipped, quarter after quarter.
• Nobody but you is checking whether last quarter’s priorities actually happened.
• The plan lives in your head, or in a document nobody reopens between meetings.

If the first list is the one that runs long, projects are your bottleneck. If the second list is the one that runs long, projects were never the problem, and that is where the expensive mistake happens.

Why hiring a project manager when you needed an integrator leaves you exactly as stuck

This is the mistake worth naming plainly, because it is the expensive one. An owner watches client work ship reasonably well, notices the business itself is not moving the way they want, decides they need “more project management,” and hires a project manager. The PM does the PM job well. Client work gets a little tighter, status gets a little cleaner. Six months later the founder is exactly as stuck, because the projects were never the bottleneck. The hire added capacity to an axis that was already fine and left the actual gap, the owner’s own plan for the business, exactly where it was: unowned, except by the owner, who is now paying for a seat that was never aimed at the real problem.

The reverse mistake exists too, and it is worth being honest about since I am selling both seats. Hire an integrator when your actual problem is that client projects slip, and you get a sharper quarterly plan and a founder who still fields every client escalation personally, because nobody is running the day-to-day coordination inside those projects. The plan gets held to account. The clients still chase you.

So what order do you hire them in?

Start with whatever is client-visible. If clients are chasing you, if deadlines surprise you before your own team catches them, hire the project manager first. Coordination pain is loud and immediate, and it costs you client relationships while you wait. Fix it before anything else.

Hire the integrator first only when client work already ships without drama and the gap is everywhere else: the hire that never happens, the offer that never launches, the plan that gets stated once in a meeting and forgotten by the next one. That gap does not announce itself the way a missed client deadline does, so it survives far longer than it should, and a project manager cannot touch it, because none of it is a project with a client attached.

In a small agency, one person often ends up doing both, and the trap is predictable. Whichever half has a client attached wins the week, so the plan half quietly stops happening while client work looks fine. If that is your situation, protect the plan half on purpose: a named block of time and a written list, reviewed on a set date. Otherwise you are paying for an integrator and receiving a project manager.

Both seats run the same way if you take the fractional route: roughly 10 to 15 hours a week, from $3,000 a month, with a 60-day minimum, because two months is the shortest window either job can actually prove itself in.

Where to start if you cannot tell

The Agency Ops Audit exists for exactly this fork: $1,500 fixed, two weeks, a written 90-day roadmap and a 60-minute readout call, built to tell you whether your bottleneck is projects or the plan before you commit a quarter’s budget to the wrong one. My basis for the read is the same twelve years running delivery operations for marketing agencies and 92 completed engagements behind everything above. If the honest answer is neither seat yet, that is a finding too, and the roadmap is yours regardless.