Your First Ten Hires: Sequence, Range, and the Cost of Getting the Order Wrong

What order should I make my first ten hires in?

Order by the constraint that is currently costing you the most, measured in weeks of delay rather than in job titles. If customers are asking for something you cannot build, the constraint is build capacity. If you have a product that works and nobody is hearing about it, the constraint is distribution. Founders frequently hire ahead of the constraint because a plan they wrote three months ago said seat four was a designer, and the plan has more authority than the evidence in front of them.

A practical test: write down the three things that did not happen last month, and for each, name the person who would have made it happen. If the same seat appears twice, that is your next hire. If three different seats appear once each, you do not have a hiring problem yet, you have a prioritization problem, and adding a person will spread the same confusion across more payroll.

The one exception to constraint-first ordering is dependency. Some seats make the following seats possible and some do not. A senior engineer who can own architecture makes the next three engineers productive faster. A first sales rep with nobody generating pipeline is dependent on you continuing to do the part you were trying to hand off. When a seat is a dependency, it earns the right to jump the queue even if it is not the loudest pain.

Should my early hires be generalists or specialists?

Range wins early for a structural reason rather than a philosophical one: at ten people the work arriving each week is unpredictable, and a narrow specialist idles between the tasks that match their specialty. Someone who can write the onboarding docs, run the customer call, and then fix the billing edge case absorbs variance that a specialist passes back to you.

The cost of range is that generalists rarely produce a defensible depth in any one area, and there are problems where shallow is worse than absent. Security posture, payroll and employment compliance, and anything that creates a legal or financial obligation are areas where a capable generalist making reasonable-sounding decisions can create liabilities that take a year to unwind. Those are the areas to buy as a service or an advisor rather than to spread across a generalist seat.

There is also a compensation consequence founders underestimate. Someone hired for range at seat five will, by seat thirty, be doing one third of what they were hired for, because the other two thirds have been split into specialist roles. If you never discuss that trajectory at the offer stage, the conversation happens instead as resentment eighteen months later. Say out loud that the job will narrow, and that narrowing is what growth looks like rather than a demotion.

Which roles should I avoid hiring in the first ten?

Any role whose main output is coordinating other people is premature at ten, because there are not enough people to coordinate. A project manager, a chief of staff, or a manager brought in to manage two individual contributors mostly generates meetings and status artifacts. The work those roles are meant to relieve is usually founder work that has not been written down, and writing it down is cheaper than staffing around it.

Second, avoid hiring a leader for a function that has never been run at all. Hiring a head of marketing before anyone has run a single campaign means you are asking a senior person to invent the function, staff it, and prove it works, with no baseline to argue from and no evidence of what your market responds to. It also removes your own learning: a founder who has never done the function cannot tell a good plan from a confident one.

Third, be honest about roles you are hiring to avoid a task you dislike. Founders often try to hire away sales, recruiting, or customer conversations at exactly the stage where those activities are the primary source of information about the business. Delegating the task removes the discomfort and the signal at the same time.

How much of my own time will the first ten hires cost me?

Budget in blocks rather than in hours, because the load is lumpy. A single filled seat at this stage typically consumes several weeks of concentrated founder attention spread across writing the role, personal outreach, screening, running most of the interviews, references, and the closing conversation. Running two searches at once does not double the cost, it more than doubles it, because context switching between two candidate pools degrades your judgment on both.

The largest hidden cost is not interviewing, it is sourcing. If you are not paying for a search partner and you have no inbound brand yet, someone has to write the outreach, and at ten people that someone is you or a co-founder. Founders who refuse to accept that number end up running searches that stay open for months, which is more expensive in aggregate than an intense four weeks would have been.

Set a rule for how many seats can be open simultaneously and hold it. Two open searches for a founding team of two or three is usually the ceiling. If the plan requires four seats filled this quarter, the honest options are to extend the timeline, bring in outside recruiting help, or accept that you will run a worse process and hire from a thinner pool.

How do I set the hiring bar when I have no calibration?

With no internal reference points, most founders substitute a proxy: a recognizable employer on the resume, an impressive school, a smooth interview. Each of those correlates with something, but none of them tells you whether the person can operate at ten people, which is a different skill from operating well inside an organization that already had process, budget, and colleagues.

Build calibration by interviewing more people than you need to before you decide. Talking to eight candidates for the first seat feels wasteful and is the cheapest calibration you will ever buy, because the eighth conversation tells you what the first one actually meant. Write your assessment before the debrief, not after, and keep the notes so that hire three can be compared to hire one on the same terms.

Use work rather than conversation wherever the role allows it. A paid short project, a live problem-solving session using a real problem from your backlog, or a review of something they built and shipped will discriminate better than any behavioral question. Say clearly what good looks like before the exercise starts, so you are evaluating the output against a stated standard rather than against your mood on the day.

What do the first ten get that hire fifty will not?

The first ten join before the answers exist, so they take on risk that later hires do not, and they influence outcomes in ways later hires cannot. That asymmetry usually shows up in equity, in title latitude, and in access to information. Being explicit about which of these you are offering, and which you are not, prevents a slow drift where early employees assume permanence in a role that the company will outgrow.

Equity arrangements, vesting schedules, and the tax treatment of any grant differ substantially by jurisdiction and by company structure, so the mechanics belong with a lawyer and an accountant who work in the country where you are incorporated and where the employee will be resident. What is portable across jurisdictions is the principle: write down what the person receives, when it becomes theirs, and what happens on the events they are most likely to worry about.

Information access is the underrated one. Early employees who see the actual numbers make better independent decisions, which is the entire reason you hired for range. The trade-off is that bad news travels through a ten-person company in an afternoon, so opening the books means committing to a level of steadiness in how you present setbacks.

How do I recover when one of the first ten is wrong?

The damage from an early misfit is concentrated rather than diffuse. On a team of ten, one person carries roughly a tenth of the total output and considerably more than a tenth of the atmosphere, so the situation does not stay contained while you decide. Colleagues route around the person, work quietly moves back to whoever it came from, and the standard everyone else is holding starts to look optional.

Separate two questions before acting, because they have different remedies. Is this a capability gap, where the person cannot yet do the work but could with structure, or is it a fit gap, where the work they are good at is not the work the company needs? Capability gaps respond to clearer expectations, a narrower scope, and a defined check-in date. Fit gaps do not improve with coaching, and a founder who applies coaching to a fit problem buys several months of everyone's time for nothing.

Whatever you decide, the process for ending an employment relationship is governed by local law and by the contract you signed, including notice, documentation, and final pay, and those rules vary widely between countries and sometimes between states or regions. Talk to an employment lawyer in the relevant jurisdiction before you act rather than after. Handle the exit with more care than the size of the company suggests is necessary, because the other nine are watching how a person is treated on the way out and drawing conclusions about their own downside.

What to take away

• ▸Order the first ten by the constraint costing you the most weeks, not by the plan you wrote last quarter.
• ▸Favor range early, buy depth as a service for anything that creates legal or financial obligation.
• ▸Do not hire coordinators, or leaders for functions nobody has ever run, before you reach ten people.
• ▸Cap simultaneous open searches at two per founder actively running them, or accept a worse process.
• ▸Interview more people than the seat requires so that hire three can be compared to hire one.
• ▸Distinguish a capability gap from a fit gap before you spend months coaching the wrong problem.