An MSP has three operating functions at every size. Service Delivery. Sales and Marketing. Finance.
Those three do not change from $2M to $50M. What changes is how many seats carry them.
There is a fourth seat, and it is the one you keep. Strategy belongs to the shareholder.
Here is the number that should get your attention.
About 81% of IT solution providers never cross $10M
Service Leadership’s 2025 industry population data: roughly 81% of IT solution providers are under $10mm in revenue. About 67% are under $5mm. 58.5% are under $3mm.
Sit with that distribution for a second.
The majority of this industry clusters under $3M. Not because those owners are lazy. Not because they lack ambition. Most of them are talented technologists working brutally hard.
They are stuck at a structural ceiling, and structure is invisible until you hit it.
You do not experience an org chart problem as an org chart problem. You experience it as being tired. As every decision routing through you. As a good month followed by a chaotic one. As hiring someone senior who somehow does not reduce your workload. As the vague sense that the business got heavier as it got bigger, which is exactly backwards from what you were promised.
That is what a structural ceiling feels like from the inside. It never announces itself.
For the full roadmap, start with scaling MSP operations.
The broken small-business org chart
Every MSP starts the same way. You did the work. Then you hired someone to help you do the work. Then you hired someone to help them.
Nobody designed that. It accreted.
So by $3M you have a lead engineer who is also the de facto Service Delivery Manager. You are the founder, and also the head of sales, and also the person the team escalates to at 4:45pm on Friday. Finance is a bookkeeper plus your gut.
That structure is not a mistake. It is precisely correct for $3M. Dual-hatting is how you got here, and it worked because at $3M the founder can physically hold the whole system in their head.
Then it stops working, and it stops working suddenly.
The failure mode is specific. A dual-hatted person under load does not split their attention evenly. They collapse into whichever hat is louder. Your engineer-slash-service-manager stops managing the moment a P1 lands, and P1s always land. Your founder-slash-head-of-sales stops selling the moment operations catches fire, and operations always catches fire.
So the second hat is not a role. It is a role that gets abandoned under exactly the conditions that require it most.
That is why revenue plateaus rather than declines. The system self-limits at the exact point where the founder’s attention runs out.
Designing an MSP organizational structure for operational scale
Start by separating two things owners constantly conflate: functions and seats.
Functions are fixed. Every MSP runs three operating engines, whether you have eight people or eighty.
Service Delivery owns the client experience and service gross margin. This is also where technology standards live. Standards are not a separate discipline, they are the instrument that determines your delivery cost, and divorcing them from the team accountable for margin is how you end up with a standards document nobody enforces.
Sales and Marketing owns pipeline generation as a system rather than as a function of the founder’s relationships. This is the function founders hold longest and should release earliest, because founder-dependent pipeline is the most visible de-risking failure a buyer will find.
Finance owns forward-looking analysis, not historical bookkeeping. The difference between a bookkeeper and a finance function is tense. One tells you what happened. The other tells you what is about to.
Seats are not fixed. How those three functions get carried is entirely a function of your scale and your bench.
At $3M, the founder holds all three and calls it being busy.
At $5M, you commonly see three functional heads reporting to a founder who is still CEO and still operating.
At $10M and above, a fourth layer usually appears: a COO or President who owns execution across all three functions, with dedicated heads of Service Delivery, Sales, and Finance beneath them. Sometimes there are more seats than that. Security and compliance may pull out as its own leader. A large project practice may separate from managed services. None of that adds a function. It redistributes the same three.
So do not go looking for the correct org chart. There isn’t one. There are three functions and a set of decisions about how many people carry them at your stage.
Copying a $30M firm’s chart at $6M is not ambition. It is cosplay with payroll.
The seat you keep
Here is where owners get the sequence wrong.
Strategy is not a fourth function you delegate to somebody. It is the seat that stays with the shareholder, and it is the only one that does.
Service Leadership treats Value Creation Strategy as the paramount maturity trait: the shareholder goal, the valuation target, the timeframe, and the incremental steps, revisited annually. It drives target customer profile, tech stack, offerings, and comp plans. Everything downstream is an expression of it.
That is not an operating function. That is an ownership function.
Which is why the COO or President seat matters more than owners realize. Hiring one is not an efficiency move. It is the structural act of separating the operator job from the owner position, so that the founder can occupy the seat that was theirs the entire time.
Owner is an equity position. Operator is a job.
Most founders try to delegate strategy and retain operations, because operations is where they are competent and strategy is where they feel exposed. That is the inversion that keeps $6M firms at $6M.
More on the founder side of this in MSP leadership maturity.
Upgrading direct reports from tasks to outcomes
Naming seats does nothing if you keep managing to-do lists.
Immature organizations manage tasks. The owner knows what everyone is doing this week. Status meetings are recitations. Delegation means assigning work and then checking whether the work got done.
Mature organizations manage outcomes. The Service Delivery leader owns service gross margin and SLA retention, holds a budget, and reports on the metric rather than the activity. You stop asking what they did. You start asking why the number moved.
That shift is uncomfortable, and the discomfort is diagnostic. If you cannot name the metric a leader owns, you have not delegated a function. You have delegated a task list and given it a title.
This is also the test that tells you whether a seat is real. Service Delivery owns service gross margin. Sales owns pipeline and cost per contract won. Finance owns forecast accuracy. If a proposed seat cannot be handed a number, it is not a function. It is a preference.
Service Leadership’s maturity data puts weight behind this. EBITDA improves markedly as a firm approaches and exceeds OML 3.0, and at OML 3.8 and above, best-in-class profitability is typically the norm. Maturity is scored across Strategy, Finance, Sales and Marketing, Service Delivery, Compensation, and Security and Compliance.
The maturity ladder and the org chart are the same object viewed from two angles.
Structure is the de-risking asset
A layered organizational chart is the clearest proof a buyer has that the business is a system rather than a job with employees. It is Value Driver #4, leadership depth and owner independence, and Value Driver #5, operational maturity, made visible on a single page.
When a buyer sees functional leaders who own metrics and hold budgets, they see transferable cash flow. When they see a founder with three direct reports who all escalate to him, they see a highly paid employee with equity.
Those two businesses can have identical revenue and identical EBITDA and receive materially different offers.
The average Infrastructure-Managed Services firm grew from $5.53mm to $8.38mm over the five years ending 2025, an 11.2% CAGR, while approximate company value grew from $3.3mm to $8.2mm, a 25.8% CAGR. Value compounded more than twice as fast as revenue.
That spread is not revenue growth. It is maturity.
One caution, because the sequence matters. Service Leadership’s First Principles are direct: you cannot grow into profitability. Get profitable, then grow. Their 2026 budget data shows bottom-quartile firms budgeting 66.4% EBITDA growth on 11.7% revenue growth, which is a firm planning to fix its margins by selling more. Best-in-class firms, already top quartile, budgeted 12.4% revenue growth and 14.1% EBITDA growth.
The disciplined firms grow slower on paper and end up worth more.
See also building a winning leadership team for MSP success and what PE buyers actually want to see.
Where to start
Not with a reorg. Reorgs announced on a Monday create six months of anxiety and rarely change who decides what.
Start by drawing the three functions and writing your own name in every one you currently carry. Most owners are surprised by the honest version of that page.
Then pick the function whose absence costs you the most attention, and define the outcome it owns before you decide who fills it. Not the tasks. The metric and the budget.
Then release it fully. Half-released is worse than not released, because the person can neither own it nor escalate it.
One function at a time. The business should get lighter as it grows, not heavier.
If you stepped out for 90 days, would the business speed up, or stop?
FAQ
What is the ideal organizational structure for a $10M managed service provider?
Three operating functions: Service Delivery owning client experience and service gross margin, Sales and Marketing owning systematic pipeline, and Finance owning forward-looking analysis. At $10M and above these are commonly layered under a COO or President who owns execution, with the CEO retaining Strategy. The functions are fixed at every revenue level. The number of seats carrying them is what scales.
Should technology and standards be a separate seat on an MSP org chart?
No. Standards determine delivery cost, so they belong to the leader accountable for service gross margin. Service Leadership scores Driving Technology Standards as a Service trait, and Product and Vendor Selection as a Strategy trait. There is no Technology domain in their maturity model. A standards function separated from delivery produces documents nobody enforces.
How do I transition dual-hatted employees into dedicated functional roles?
Define the outcome and metric for each function first, then decide who carries it. Split the hats one at a time rather than in a single reorg. Expect the person to keep collapsing into the louder hat until the other is formally reassigned. A dual-hatted role is not a role, it is a role that gets abandoned under exactly the pressure that requires it.
When should an MSP hire a COO or President?
When the founder is the constraint on execution rather than on direction. The COO seat exists to absorb the operator job so the owner can hold Strategy, which Service Leadership treats as the paramount maturity trait and which does not delegate. If you are still resolving delivery escalations personally at $8M, the seat is already overdue.
Why do most MSPs stall before $10M in revenue?
Roughly 81% of IT solution providers are under $10mm, about 67% under $5mm, and 58.5% under $3mm, per Service Leadership’s 2025 data. The constraint is usually structural rather than commercial. Informal reporting lines require founder presence to resolve, so the business self-limits at the point where founder attention runs out, typically between $3M and $6M.