Founders ask Collective 54 this 9 times in our records. It is usually asked as an org chart question and is an execution ownership question.
When a boutique firm stalls, the diagnosis is usually wrong. Founders assume the market changed, differentiation weakened or competition intensified, and they revisit positioning. In most cases the strategy is viable. What is missing is execution ownership.
The symptoms are recognizable. Decisions get made but not enforced. Plans are articulated but not operationalized. Priorities are declared but not protected. Initiatives start and rarely finish. Decisions resurface because no one enforced them. Meetings produce agreement without momentum. Accountability diffuses across leaders who already have full-time jobs. And the founder becomes the escalation point for everything.
This is not failure by collapse. It is failure by drift, and none of it shows up on a strategy slide.
The critical distinction: execution ownership is a role, not a personality trait. One of the most damaging beliefs in professional services is that the right founder or the right hire will simply make things run smoothly. That belief collapses under scale. If no role carries explicit responsibility for converting decisions into action, enforcing priorities, maintaining operational rhythm and protecting momentum, no amount of talent compensates.
The standard response is to hire a COO or head of operations. Across the boutique lifecycle that role breaks down in three predictable ways, and it is worth knowing which one you are in.
Small firms need the role acutely and cannot afford a genuinely capable one. The founder absorbs the burden knowing exactly what is missing.
Mid-sized firms can afford it and do not know what they are buying. They confuse senior project management with execution ownership, operational support with operational leadership, and process optimization with decision enforcement. The result is overqualified hires who outgrow the role, underpowered hires who cannot enforce decisions, fractional arrangements with unclear authority, and churn.
Larger firms almost always have an operations leader, hired to keep the firm running rather than to institutionalize execution. That person was not developed to hold institutional memory, enforce strategic intent or serve as a credible successor. When the founder wants to exit, execution still flows through the founder, and the exit stalls or discounts heavily.
The COO model assumes conditions boutique firms do not have: stable repeatable operations, clear functional separation, mature management layers and the scale to absorb inefficiency. In a boutique firm the systems are still being invented, often daily.
The design principle that makes the role work is a clean separation between leadership and governance. Execution ownership fails when humans are asked to perform continuous governance.
Humans should own strategic interpretation, translating founder intent into priorities when trade-offs arise; judgment under ambiguity; escalation and intervention when execution breaks in ways systems cannot resolve; people leadership, meaning conflict, motivation, alignment and trust; and final accountability for outcomes.
The governance layer should own decision memory, retaining what was decided, why, and what was deprioritized; commitment tracking, monitoring who committed to what and whether it happened; cadence enforcement, so execution rhythms persist without policing; drift detection, catching priorities fading and scope creeping; cross-functional visibility across sales, delivery, finance and people; and pattern recognition on recurring failures.
That second list is not leadership work. It is cognitive labor, and it exceeds what even exceptional people can sustain continuously without becoming the bottleneck themselves. The finance analogy holds: no serious firm asks its CFO to reconcile ledgers by hand, and in Era 3 no serious firm should ask its operations leader to govern execution by memory.
The test of whether it is working is specific. If the founder remains the execution backstop, the role has failed, whatever the title.
With ownership assigned, the org design question becomes narrow. Two rules cover most of it.
Organize on one axis. Geography, industry or function. Stay away from a matrix. A three-spoke matrix works at 505,000 employees and is absurd at fifty. The reason to care extends past daily clarity: a simple model is easier for an acquirer to absorb, and buyers notice. One large consulting firm chose a small regional boutique over larger targets specifically because it rolled cleanly into one partner's geography and one industry practice, making the integration painless. The investment bank was surprised; their fee was tied to deal size.
Make the structure reflect the niche and the business model, and make it obvious where the synergies would come from. If your organizational model does not describe how you actually make money, it is decoration.
On the pyramid question, the up-or-out model of finders, minders and grinders has real advantages: a talent pipeline, clear career paths, high productivity against established standards, and easy arithmetic, since a firm growing 30 percent needs 30 percent more people. Its disadvantages matter more now. Revenue growth and people growth stay linear. It assumes impatient young employees will sit in a role for years, and many will not. It requires most new hires to enter at the bottom, limiting senior lateral talent. And it makes pushing equity into the ranks difficult.
The judgment is that the model is outdated for a firm trying to scale, and the alternative is not a different chart but a different production model.
This is the point most org design conversations miss. Structure follows the production model, and if the production model is linear the structure will be too.
Three routes exist to break the link. Make services tech-enabled, so the work is produced differently rather than merely supported. Offshore, where market leaders run about 40 percent of work offshore and boutique firms under 5 percent, which is a large arbitrage sitting unused. And use gig networks and labor marketplaces to flex capacity up and down against demand rather than carrying it.
The framing worth adopting: scale does not mean the number of employees, it means the amount of cash flow. The best boutique firm would have no employees and many clients. Labor is your biggest cost, so organize to reduce labor-related expense.
In Era 3 the same logic reaches the production of the work itself. AI systems deliver significant portions of the work while humans supervise judgment, quality and accountability, output per employee increases materially, and revenue can grow without proportional increases in labor. Firms that adopt tools without redesigning the firm see limited benefit. Structure is part of the redesign, not a separate exercise.
Two external tests are worth applying, because they are the ones that eventually get applied to you.
The first is the founder time test. When execution ownership works, the founder works on strategy rather than execution, growth initiatives progress without constant intervention, decisions stick, and the firm moves forward without founder bottlenecks. If none of that is true, the structure is not working regardless of how sensible the chart looks.
The second is the transferability test. Execution ownership has to survive a leadership transition: knowledge preserved beyond individuals, successors able to step in, founder dependency materially reduced. At exit this is not a soft consideration. Buyers see either continuity or risk, and they price accordingly.
If you are under roughly ten people, most of this is premature. Three boxes, marketing and sales, service delivery and service development, is genuinely sufficient at that size, and installing governance machinery early creates overhead without relieving anything. The one thing worth doing early is writing decisions down, because decision memory is cheap to start and expensive to retrofit.
If you are deliberately running a lifestyle firm, the transferability test does not apply and you should not organize for it. Founder-centric structures are efficient when nobody intends to sell. The cost is optionality, which is a legitimate thing to trade away knowingly.
And if you are in the middle of a sale process, do not reorganize. Buyers underwrite demonstrated performance, and a structure with two quarters of history reads as instability rather than improvement. Fix execution ownership before you go to market or after you close, not during.
Assign execution ownership before you redraw anything, because firms stall for lack of it rather than for lack of strategy, and it is a role rather than a personality trait: if nothing explicitly owns converting decisions into action, enforcing priorities, holding decision memory and maintaining cadence, no amount of talent compensates. Split it cleanly, with humans owning interpretation, judgment, escalation, people leadership and final accountability, and the governance layer owning decision memory, commitment tracking, cadence, drift detection, cross-functional visibility and pattern recognition, because that second list is cognitive labor no one can sustain continuously. The test is whether the founder is still the execution backstop. Then keep the structure boring: organize on one axis of geography, industry or function, avoid the matrix, and make the model reflect how you actually make money, which also makes you easier to acquire. Treat the up-or-out pyramid as outdated for a scaling firm, and design so revenue can outgrow headcount through tech-enabled delivery, offshoring where market leaders run 40 percent against your likely 5, and flexible capacity. Scale means cash flow, not employees.
Because nobody owns execution. Decisions get made but not enforced, plans are articulated but not operationalized, priorities are declared but not protected, initiatives start and rarely finish, and accountability diffuses across leaders who already have full-time jobs. The founder becomes the escalation point for everything. This is failure by drift rather than collapse, and none of it appears on a strategy slide. The critical point is that execution ownership is a role rather than a personality trait: if no role carries explicit responsibility for converting decisions into action and maintaining cadence, talent does not compensate.
Because the role breaks down differently at each size. Small firms need it acutely and cannot afford a capable one. Mid-sized firms can afford it but confuse senior project management with execution ownership, operational support with operational leadership, and process optimization with decision enforcement, producing overqualified hires who outgrow the role, underpowered hires who cannot enforce decisions and costly churn. Larger firms have an operations leader hired to keep the firm running rather than to institutionalize execution, so at exit the work still flows through the founder and the deal stalls or discounts.
Organize on one axis, geography, industry or function, and stay away from a matrix. Beyond daily clarity, a simple model is easier for an acquirer to absorb: one large consulting firm chose a small regional boutique over larger targets precisely because it rolled cleanly into the geography of one partner and into a single industry practice. The up-or-out pyramid of finders, minders and grinders has real advantages but keeps revenue growth and people growth linear, assumes young employees will stay in a role for years, forces entry at the bottom and makes pushing equity into the ranks hard. It is outdated for a firm trying to scale.
By changing the production model, since structure follows it. Three routes break the link: making services tech-enabled so work is produced differently rather than merely supported; offshoring, where market leaders run about 40 percent of work offshore against under 5 percent for boutique firms, a large unused arbitrage; and gig networks that flex capacity against demand rather than carrying it. In Era 3 the same logic reaches production itself, with AI delivering significant portions of the work while humans supervise judgment and quality. Scale means cash flow, not employees.
Sources: Greg Alexander, The AI-Native Boutique Firm (Advantage Books, January 2027), specifically The AI Operations Manager for the finding that firms fail for lack of execution ownership rather than lack of strategy, for the symptoms of unowned execution including decisions that resurface, meetings that produce agreement without momentum and founders who become the escalation point, for execution ownership as a role rather than a personality trait, for the three-stage failure of the COO model across small, mid-sized and larger firms, for the mismatch between COO assumptions and boutique conditions, for the division between human leadership covering strategic interpretation, judgment under ambiguity, escalation, people leadership and final accountability and governance covering decision memory, commitment tracking, cadence enforcement, drift detection, cross-functional visibility and pattern recognition, for the finance analogy that no serious firm asks its CFO to reconcile ledgers manually, for Founder Load Protection and the test that the role has failed if the founder remains the execution backstop, and for Leadership Transferability as the capability that determines whether the firm can exit. Greg Alexander, The Boutique: How to Start, Scale, and Sell a Professional Services Firm (Advantage, 2020), chapter 21 for the three-box start-up structure, the up-or-out pyramid of finders, minders and grinders popularized by David Maister with its advantages and disadvantages, the judgment that the model is outdated for firms trying to scale, the three routes to decoupling revenue growth from employee growth through tech-enabled services, offshoring where market leaders offshore about 40 percent against under 5 percent for boutique firms, and gig networks, and for the position that scale refers to cash flow rather than employee count; chapter 38 for organizational design as an exit consideration, the instruction to organize around geography, industry or function and avoid the matrix, and the account of a large consulting firm choosing a small regional boutique because it integrated cleanly into one partner's geography and one industry practice. Greg Alexander, The Era Framework, for Era 3 as a change in the production model in which AI delivers significant portions of the work while humans supervise judgment, output per employee increases materially and revenue can grow without proportional increases in labor, and for the finding that firms adopting AI tools without redesigning the firm see limited benefit.
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.