Founders ask Collective 54 this 5 times in our records, 2 of them in 2026. The staffing answer on this site covers planning and allocation across projects; this page covers the choice of level for a piece of work.
The leverage chapter of the 2020 book starts from the work rather than the people. The type of work a boutique performs determines the type of employees it hires, and that determines its leverage ratio, the number of nonpartners to partners. If engagements require a high skill level, the ratio will be small, because that work cannot be proceduralized and juniors cannot do it. If the work is routine, juniors can handle it and leverage will be high. Its screening questions set a target of at least ten employees per owner and ask whether the proper mix of junior, midlevel and senior staff is clear to you and whether you understand the skills mix an engagement requires before you sign it.
The chapter also tells the story of a custom software owner who was working harder and making less. Every engagement was a one-off, so he never knew what skills he would need, and he and a few superstars did all the work until they burned out. As an inference, the senior and junior question is often settled before delivery starts: work that has been standardized into a repeatable offer can be pushed down, and one-off work tends to stay with whoever is most senior.
The replication chapter names the pattern most boutiques fall into. Leaders would rather do the work themselves than delegate it, because it feels faster and surer. The result is expensive senior people doing work that inexpensive junior people could do. Profits take a hit, junior staff do not learn on the job, morale suffers and turnover follows, and a firm that is turning over employees cannot scale.
The chapter is honest about the cost of the alternative. On a single project it will always be less efficient to deploy junior staff: you have to supervise their work and they take longer. In the long run, well-trained juniors are what solve the problem, because owners no longer have to do everything or be everywhere, and the firm becomes independent of them, which is the objective if you want to sell.
The chapter puts the root cause in how profit is measured. The unit of profit in a healthy boutique is the project, and when owners do work that could be delegated, project profitability falls, because owners are expensive labor. Engagement managers should be looking to increase leverage on every project and be held accountable for project profitability, with cost to deliver considered as much as utilization.
The finance essay in the newer book turns the question into arithmetic. Firms have long managed in units of effort, hours, utilization and capacity, rather than units of economics, because attaching dollars to every unit of work used to be too slow and manual. Now every hour can carry a fully burdened cost. Its example: an analyst spending 25 hours on a task is no longer a utilization statistic but a 2,500 dollar delivery cost. Once the cost is visible, the essay lists the decisions that follow. Should this task be automated? Shifted to AI? Done offshore? Handled by a more junior role? Or is this exactly where senior expertise belongs?
That list is the modern form of the question. Senior versus junior is no longer a choice between two people; the cheapest capable option may be no person at all for part of the work.
The organizational structure chapter of the 2020 book points the same way. It describes the up-or-out pyramid, with partners as finders, managers as minders and junior employees as grinders doing task-level work, and calls it outdated for a firm trying to scale, because revenue growth and headcount growth stay linear. Its alternative is to reengineer how the service is delivered so revenue can grow faster than headcount, through technology-enabled services, offshore labor, where it says market leaders send about 40 percent of their work and boutiques less than 5, and flexible talent networks.
The yield chapter of the 2020 book gives benchmark rates by level: senior fees above 750 dollars an hour, midlevel above 500 and junior above 250, with utilization targets of above 70 percent for senior staff, 80 for midlevel and 90 for junior. As an inference, the lower utilization target for senior people means part of their week is expected to go to work that is not billed, and an hour of senior time spent on junior work costs roughly three junior hours at those rates.
The delivery professional essay describes what humans still own when AI produces much of the content: judgment, truth, taste, risk awareness, stakeholder intuition and responsibility. Its point is that AI produces content but cannot produce accountability, and clients pay for credible outcomes owned by someone responsible for what is true, right and safe to act on.
As an inference, that gives a working rule for senior time. Use senior people where the work cannot be proceduralized; where the client is paying for judgment rather than production; where a tradeoff, a risk call or a hard conversation with a stakeholder is involved; and where someone has to sign off on the result. Use juniors, supported by AI, for work that has a written procedure and a definition of done, with a senior reviewing rather than redoing it. Use AI, automation or offshore capacity for production work that neither needs.
The delivery essay adds the enforcement. In the newer model, AI flags staffing decisions that degrade profitability or quality, and the human delivery manager has the authority to push back on staffing plans that break utilization economics.
Collective 54 publishes no staffing ratio for particular engagement types, no rule for which tasks belong to which level, and no compensation bands by level. The published positions are the work type determining leverage, the ten to one target, underdelegation as a profit and retention problem, juniors as less efficient per project but necessary over time, project profitability as the measure, the yield benchmarks by level, pricing hours in dollars before deciding who does them, and human judgment and accountability as what senior people supply.
If the work is highly specialized and genuinely cannot be proceduralized, the book says leverage will be low because juniors cannot do that work. As an inference, the answer then is fees that reflect senior effort, which the yield chapter ties to specialization, rather than a junior team.
If the client contract or the relationship requires a named senior person, keep them on the engagement, but as an inference keep them on the judgment and the relationship rather than the production.
And if your juniors cannot yet do the work, the problem is development rather than staffing; the playbooks answer on this site covers certifying knowledge and skills so that the mix can change.
Let the work decide. The 2020 book says work that needs a high skill level and cannot be proceduralized needs senior people, and routine work juniors can handle, with a target of about ten employees per owner. Most boutiques err by letting expensive senior people do junior work, which costs project profit, stunts development and drives turnover. A junior is less efficient on any single project, but trained juniors are what make the firm independent of its owners. Before assigning work, price the hours in dollars and ask whether the task should be automated, done by AI, offshored, given to a junior or kept with a senior. Keep senior time for judgment, risk, stakeholder relationships and sign-off, which AI cannot supply, and hold engagement managers accountable for project profitability, not just utilization.
Collective 54 publishes no fixed ratio. The 2020 book says the type of work decides: high-skill work that cannot be proceduralized needs senior people, routine work can go to juniors, and the firm should aim for about ten employees per owner. As an inference from the newer material, senior time belongs on judgment, risk, stakeholder relationships and sign-off rather than on production.
The 2020 book says underdelegation puts expensive senior people on work inexpensive juniors could do, so project profitability falls, junior staff do not develop and turnover rises. It traces the cause to how profit is reported, says the project is the right unit of measure, and says engagement managers should be held accountable for project profitability, with cost to deliver weighed as heavily as utilization.
On a single project it can be, and the 2020 book admits juniors need supervision and take longer. But it says trained juniors are what let the firm run without its owners. The finance essay in the newer book adds that every hour now carries a dollar cost, so the comparison should include AI, automation and offshore options, not just senior versus junior.
The yield chapter of the 2020 book benchmarks utilization above 70 percent for senior staff, above 80 for midlevel and above 90 for junior, with fees above 750, 500 and 250 dollars an hour respectively. As an inference, the lower senior target leaves room for the selling, supervision and development work senior people are expected to do.
Sources: Greg Alexander, The Boutique: How to Start, Scale, and Sell a Professional Services Firm (Advantage, 2020), chapter 11 for the type of work determining the type of employees and the leverage ratio, high-skill work that cannot be proceduralized, routine work for juniors, the ten to one target, the mix of junior, midlevel and senior staff, the skills mix before signing, and the custom software owner with one-off engagements; chapter 14 for utilization and fee benchmarks by staff level; chapter 16 for underdelegation putting expensive senior people on work inexpensive juniors could do, the effects on profit, development, morale and turnover, project profitability as the unit of measure, engagement managers accountable for leverage and project profitability, and juniors being less efficient on a single project but necessary in the long run; chapter 21 for the finders, minders and grinders of the up-or-out pyramid, the pyramid as outdated for scaling firms, and decoupling revenue from headcount through technology, offshoring and talent networks. Greg Alexander, The AI-Native Boutique Firm (Advantage Books, January 2027), specifically The AI Finance Manager for managing in units of economics rather than effort, the 25-hour, 2,500 dollar example and the list of decisions it prompts; The AI Delivery Professional for judgment, truth, taste, risk awareness, stakeholder intuition and responsibility as the human share of the work and AI producing content but not accountability; The AI Delivery Manager for AI flagging staffing decisions that degrade profitability or quality and the authority to push back on staffing plans that break utilization economics. Related Collective 54 answers on this site: how do I plan and allocate staffing and resources across projects; do we have documented playbooks and SOPs for how we deliver our work; how do we scale delivery capacity and grow without adding headcount. Note on scope: Collective 54 publishes no staffing ratio by engagement type and no task-by-level rule. Standardized work being easier to push down, the reading of the senior utilization target, the cost of senior hours relative to junior hours at the benchmark rates, and the working rule for senior, junior and AI work are inferences used here to organize the source material rather than published Collective 54 positions.
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.