When to hire a VP of engineering

Most startups hire a VP of engineering when the founder or first technical lead can no longer both ship the roadmap and grow the team at the same time. In practice that crossover lands somewhere around 12 to 25 engineers, when you have two or three teams and the cost of bad coordination starts to exceed the cost of a senior leader's salary.

That number is a symptom, not the cause. The real trigger is a set of signals you can watch for: the founder is in too many meetings to write the code that matters, planning has gone from a Slack thread to a fight, and good engineers are leaving for reasons you only hear about in the exit conversation. This post is about reading those signals early, knowing what the role actually does, and avoiding the two expensive mistakes: hiring a VP eng before the team is large enough to need one, and waiting until the team is already breaking.

What does a VP of engineering actually do?

A VP of engineering owns delivery and the organization that produces it. They are not the person with the deepest architecture opinions in the building. That is usually a CTO or a staff engineer. The VP eng is the person accountable for whether the team ships predictably, whether it is staffed correctly, and whether your best people stay.

Concretely, the job splits into four areas:

• Execution. Turning a roadmap into shipped software on a believable timeline, and giving you honest dates instead of hopeful ones.
• People. Hiring, leveling, performance, and retention. They own the org chart and the quality bar for who joins it.
• Process. Planning cadence, on-call, code review norms, incident response. Enough structure to scale, not so much that it slows a 20-person team down.
• Translation. Sitting between you and the engineers, turning business priorities into technical work and technical risk into language the rest of the leadership team can act on.

This is different from a CTO, and conflating the two is a common early mistake. A CTO sets technical direction and is often a co-founder. A VP eng runs the machine that executes against it. At a 15-person team you may want both functions, but you rarely want them in one new external hire. For more on the surrounding mistakes, see common startup hiring mistakes.

What signals say you need one now?

You are ready for a VP eng when most of these are true at the same time, not just one:

1. 01The founder is the bottleneck for delivery. Decisions wait on one person. The most technical founder spends more time unblocking others than building, and the roadmap slips because of it.
2. 02You have crossed two or three teams. One team can be run by a strong lead. Three teams need someone whose full-time job is coordination across them.
3. 03Planning has become contentious. Estimates are missed, priorities reshuffle weekly, and nobody owns the calendar that holds it together.
4. 04You are hiring fast and leveling is guesswork. You are about to add five or more engineers and have no consistent way to interview, level, or onboard them.
5. 05Senior people are quietly leaving. Good engineers leave for managers, not companies. If retention is slipping with no clear technical reason, the gap is leadership.

The cleanest single tell: you are spending more on the cost of disorganization (missed dates, duplicated work, attrition) than a leader's fully loaded salary would cost. At that point the hire pays for itself even if the leader is only average.

How many engineers before a VP of engineering?

There is no exact threshold, but the ranges below hold across most venture-backed teams. The point is the shape of the team, not the raw count.

If you are below roughly 12 engineers, your better first move is usually a single experienced engineering manager, not a VP. The manager costs less, slots in faster, and buys the founder back the hours needed to keep shipping. We cover the earlier version of this decision in first engineering hires after a seed round.

What does hiring too early cost?

Hiring a VP eng before you have a team to run is one of the more expensive unforced errors at seed and Series A. Here is the math, with round numbers.

A VP of engineering in San Francisco or New York costs roughly $250,000 to $350,000 in base salary, plus a meaningful equity grant, often 0.5 percent to 1.5 percent at Series A (illustrative, not advice). On a small team, that is one of your largest line items spent on someone whose core skills (scaling an org, building a management layer) have nothing to do yet.

The worse cost is not the salary. It is what an underused VP does to stay busy. With no org to build, they add process a 10-person team does not need, insert a layer between the founder and the engineers who were doing fine without one, and slow the very thing that made the early team fast. When it does not work out, you lose the search time, the ramp time, and the months spent unwinding the structure they installed. For a venture-backed company, that can quietly burn a quarter or two of runway. If you are weighing senior hires against your stage generally, hiring senior engineers after a raise walks through the sequencing.

What does hiring too late cost?

Waiting too long is the more common failure, and it is harder to see because nothing breaks on a single day. The team just gets slowly worse at shipping. The signals: dates that are always optimistic, the same two senior engineers burning out as informal managers, onboarding that takes months because no one owns it, and attrition that creeps up while the founder is too deep in delivery to notice the pattern.

By the time it is obvious, you are hiring a VP eng into a fire instead of ahead of one. They spend their first six months on damage control: rebuilding trust, replacing people who left, and untangling a planning process that has calcified. You pay for the leader and you pay again for the months of compounded disorganization before they arrived.

Hire the VP eng to prevent the fire, not to put one out. The first is a strategic hire. The second is an emergency, and emergencies hire worse candidates.

How do you actually hire one?

Once the signals line up, run a deliberate search. A VP eng is a force multiplier or a force divider, and the wrong one is hard to remove without resetting the team. A workable sequence:

1. 01Write the mandate first. Decide what this person owns in the first year: scale from 15 to 35, install a planning cadence, build a hiring machine. Vague mandates attract generalists who interview well and execute poorly.
2. 02Match the scale to their experience. Someone who ran a 200-person org may not enjoy or be good at building from 15. Hire for the next two years of your size, not the logo on their resume.
3. 03Test for execution, not philosophy. Ask for a specific story: a team they took from X to Y, what broke, what they changed. Probe the numbers. Real operators have details. See how to do engineering reference checks.
4. 04Have your best engineers interview them. The people who will report up should pressure-test whether this leader would make them better or just add overhead.
5. 05Check references hard. For a leadership hire, backchannel references matter more than the formal ones. Talk to engineers who reported to them, not just peers.

Most founders making this hire for the first time underestimate how different it is from hiring an engineer. The skills you can evaluate yourself (does this person write good code) are not the skills that matter (can this person build and retain a team). This is a place where a recruiter who has placed engineering leaders earns the fee, because the evaluation is genuinely outside most founders' experience. If you are deciding whether to run it yourself, in-house recruiter vs agency for startups lays out the tradeoff.