Founders ask Collective 54 this 9 times in our records, and 5 of those were in 2026. The word actually is doing work here: most founders can read the number and cannot explain it.
The unit of measure of profit for a healthy boutique firm is the project. A firm's financial performance is the sum of its projects.
This sounds like an accounting preference and is actually the whole answer to the question. A firm-level margin number tells you that something changed. A project-level view tells you what. Averages hide the two things you most need to see: which work is subsidizing which, and how much variance sits underneath a stable-looking blended figure.
Capital 54 has seen the diagnostic value of this directly. In one diligence on a commercial photography firm, the unit of measure was cash flow per project, built from fee, hours per staff member, fully loaded cost per staff member and allocated overhead. Some projects produced strong cash flow and others produced negative cash flow. The volatility was the disqualifying finding, because it meant the delivery model was not standardized and therefore not scalable.
Note what that firm's blended margin would have shown: something acceptable. The problem was invisible at the firm level and obvious at the project level.
Two related distinctions matter while you are at it. Cash flow is not net income and is not EBITDA, and firms run on cash. Look at cash flow per partner and cash flow per project, not only at reported margin.
Yield is average fee per hour times average utilization rate. At a $400 average fee and 75 percent utilization, yield is $300 per hour. On the standard assumption of a forty-hour week and forty-eight weeks, that is 1,920 hours per employee and $576,000 of revenue per head.
That formula is worth writing down because it tells you where the room is. Most boutique firms past the start-up stage have already optimized utilization; they would not have survived otherwise. The point of diminishing returns has been reached, and further improvement there means asking people to work on Christmas Day.
So the lever is fees. And raising fees is not the same as raising prices. Competitive markets push fees down, so the route is to become more valuable rather than to push harder, and the reliable route to that is specialization across five dimensions: industry, function, segment, problem and geography. A firm specialized on three to five of those commands a fee a generalist cannot.
Benchmarks worth measuring yourself against: average utilization above 85 percent, senior above 70, midlevel above 80, junior above 90; average fees above $400, senior above $750, midlevel above $500, junior above $250.
Leverage is the ratio of non-partners to partners. Thirty employees and three owners is 10:1, and 10:1 is the floor for a firm that intends to scale.
Leverage drives margin through a simple mechanism: when owners perform work that could be delegated, project profitability falls, because owners are expensive labor. Under-delegation is not a time-management failing, it is a margin event, and it repeats on every project.
The second-order cost is worse. Junior staff who never get real work do not develop, which raises turnover, and a firm turning over employees cannot hold its delivery quality or its capacity. The margin damage compounds through recruiting and rework long after the original decision.
The constraint on leverage is the type of work. Engagements requiring high skill that cannot be proceduralized keep leverage low by nature, because juniors genuinely cannot do them. What breaks firms is taking on that kind of work while assuming the economics of routine work.
The fix is replication, and it has a specific form: certify knowledge, meaning practical understanding of your domain, and skills, meaning the ability to do particular things such as running an executive interview properly. Build the exam from post-mortems of a representative sample of engagements, broken down to the task level. Grade people 101, 201 or 301. Then staff from the record rather than from memory.
The third driver is the one that does not appear as a line item anywhere, which is why it survives.
Delivery is where revenue is either converted into EBITDA or quietly lost. Scope creep, margin leakage, poor bench management and weak handoffs do not arrive all at once. They accumulate, and by the time the financials reveal the damage the work is done and the profit is gone.
The term for this is trapped profitability: profit the firm has already earned but failed to capture because delivery execution broke down. It hides in an unchallenged scope change, a misstaffed project, a late time entry, a missed utilization target, a handoff that lacked clarity. Individually harmless. Collectively, they are the difference between your target margin and your actual one.
This is also the driver most responsive to instrumentation, because the failure mode is a small problem noticed late rather than a missing insight. Continuous monitoring of engagement health, real-time detection of scope creep, ongoing cost-to-complete forecasting and enforcement of delivery methodology are exactly the work humans perform badly and machines perform without fatigue.
One organizational condition has to accompany it. Whoever owns delivery needs the authority to say no: to deals that cannot be delivered profitably, to scope changes without economic justification, to staffing plans that break the utilization math. Without that authority, saying yes is rewarded, saying no is treated as obstruction, and margin erosion is absorbed as the price of client satisfaction.
There is a reason founders ask this question rather than answering it themselves, and it is not a lack of attention.
Most boutique firms have historically managed in units of effort rather than units of economics. They track hours, utilization and capacity. What they have not tracked systematically is what those hours cost, what they returned, and whether they were allocated well. That was a technical limit rather than a philosophical one: attaching dollars to every unit of work was too slow and too error-prone to be practical, so finance stayed downstream of delivery, recording outcomes without shaping behavior.
The consequence is a specific kind of blindness. An analyst spending 25 hours on a task is a utilization statistic. The same 25 hours is also a $2,500 delivery cost. That translation is what turns a margin question into an operating decision: should this be automated, shifted, done offshore, handled by a more junior person, or is this exactly where senior expertise belongs.
The related blindness is benchmarking. Founders who believe a 50 percent gross margin is strong, or that a 25 percent EBITDA margin is impressive, are usually comparing to nothing. Without an external reference point every decision is made in a vacuum, and finance can report what happened without saying whether it was good, bad or fixable.
Diagnose before acting, then work in this order.
First, get project-level visibility. Nothing else can be prioritized without it, and the variance you find will usually tell you which of the three drivers is binding.
Second, stop the leak. Delivery discipline is the fastest and least disruptive improvement, because the profit already exists and is being lost rather than never earned. It requires no price conversation and no reorganization.
Third, fix leverage. Slower, because certification and delegation take quarters rather than weeks, but structural. This is also what breaks the link between revenue growth and headcount growth, which is the difference between growing and scaling.
Fourth, raise yield. Last, not because it matters least but because specialization is a strategic change and a price increase lands better from a firm that can already deliver reliably.
If your margin problem is concentrated in one or two client relationships rather than spread across the book, this framework is overbuilt. Fix or exit those engagements and re-measure before redesigning anything.
If you are genuinely an intellect firm hired for never-before-seen problems, low leverage is your business model rather than a defect, and pushing a 10:1 ratio onto work that cannot be proceduralized will produce the quality failure you are trying to avoid. In that case yield and delivery discipline are your only two levers, and selectivity about which engagements you accept matters more than either.
And if the firm is growing quickly, expect margin to dip during the build. Adding capacity ahead of revenue compresses margin temporarily and is often the correct decision. The distinction worth holding is between margin you are investing and margin you are losing, which is another thing only the project-level view will tell you.
Change the unit of measure first, because margin is created and destroyed at the project level while your financials report at the firm level, and an acceptable blended number can sit on top of projects that range from strongly profitable to cash negative. Then decompose it into three drivers. Yield, which is average fee times utilization, where utilization is usually already optimized and the remaining room is in fees, earned through specialization across industry, function, segment, problem and geography rather than through pushing harder on price. Leverage, where the 10:1 floor matters because owners doing delegable work is an expensive labor decision repeated on every project, and where the fix is knowledge and skills certification graded 101, 201 and 301 so staffing runs off a record rather than memory. And delivery leakage, the trapped profitability lost through unchallenged scope changes, misstaffing, late time entry and vague handoffs, which is the fastest thing to recover because the profit already exists. Fix them in that order, and give whoever owns delivery the authority to say no, because without it the leak reopens.
Because the unit of measure is wrong. The unit of profit for a healthy boutique firm is the project, and the financial performance of the firm is the sum of its projects, so firm-level reporting tells you that something changed without telling you what. Capital 54 saw this directly in diligence on a commercial photography firm: measured per project using fee, hours per staff member, fully loaded cost and allocated overhead, some projects produced strong cash flow and others produced negative cash flow. The volatility was the disqualifying finding, and the blended number would have looked acceptable.
Three. Yield, which is average fee per hour times utilization; at $400 and 75 percent that is $300 an hour, or $576,000 per head across 1,920 hours. Leverage, the ratio of non-partners to partners, with 10:1 as the floor for a firm that intends to scale, because owners doing delegable work is expensive labor charged to every project. And delivery leakage, the trapped profitability lost to unchallenged scope changes, misstaffed projects, late time entry and unclear handoffs, which accumulates invisibly and is only visible in the financials after the work is done.
Get project-level visibility first, because nothing else can be prioritized without it and the variance usually reveals which driver is binding. Then stop the delivery leak, which is fastest and least disruptive since the profit already exists and is being lost rather than never earned, and it requires no price conversation. Then fix leverage, which is slower because certification and delegation take quarters, but structural, since it breaks the link between revenue growth and headcount growth. Raise yield last, because specialization is a strategic change and a price increase lands better from a firm that already delivers reliably.
Because most boutique firms manage in units of effort rather than units of economics. They track hours, utilization and capacity, but not what those hours cost or returned, which was a technical limit rather than a philosophical one. An analyst spending 25 hours on a task is a utilization statistic and also a $2,500 delivery cost, and that translation is what turns a margin question into an operating decision. The second blindness is benchmarking: founders who think a 50 percent gross margin is strong or a 25 percent EBITDA margin is impressive are usually comparing against nothing.
Sources: Greg Alexander, The Boutique: How to Start, Scale, and Sell a Professional Services Firm (Advantage, 2020), chapter 14 for yield as average fee per hour times average utilization rate, the worked example at $400 and 75 percent producing $300 an hour and $576,000 per employee across 1,920 hours, the point of diminishing returns on utilization, the utilization and fee benchmarks by seniority, and the five forms of specialization by industry, function, segment, problem and geography; chapter 11 for leverage as the ratio of non-partners to partners and the 10:1 floor, and for the way the type of work sets the ceiling on leverage; chapter 16 for the project as the correct unit of profit measurement, for under-delegation putting expensive senior labor on delegable work and lowering project profitability, for the turnover consequences of junior staff who never develop, and for knowledge and skills certification graded 101, 201 and 301; chapter 12 for cash flow as distinct from net income and EBITDA, for cash flow per partner and cash flow per project, and for the Capital 54 diligence on a commercial photography firm whose project-level cash volatility revealed a delivery model that was not standardized and therefore not scalable; chapter 21 for decoupling the rate of revenue growth from the rate of employee growth. Greg Alexander, The AI-Native Boutique Firm (Advantage Books, January 2027), specifically The AI Delivery Manager for delivery as the point where revenue is converted into EBITDA or quietly lost, for trapped profitability accumulating through unchallenged scope changes, misstaffed projects, late time entries, missed utilization targets and unclear handoffs, for the Era 3 division in which continuous monitoring, scope-creep detection, cost-to-complete forecasting and methodology enforcement run without human stamina, and for the requirement that the delivery function hold the authority to say no to sales, scope and staffing; and The AI Finance Manager for firms managing in units of effort rather than units of economics, for the translation of 25 analyst hours into a $2,500 delivery cost and the operating decisions that follow, for activity-based costing informing service design, pricing, scoping, staffing and capacity, and for the benchmarking blindness in which founders judge a 50 percent gross margin strong and a 25 percent EBITDA margin impressive with no external reference point.
Collective 54 is the private community for founders and executives of boutique professional services firms between $5M and $50M in revenue. Members work these answers against their own numbers.