The VP who takes engineering from 40 to 150 is rarely the VP who took it from 10 to 40, and almost never the VP who'll take it from 150 to 500.

‍

Boards hire engineering leaders like they're hiring a partner for the next decade. The candidate profile is written broadly, the reference checks focus on general capability, and the implicit assumption is that the right VP will grow with the company through whatever comes next.

That assumption breaks under scrutiny. The VP of Engineering job is not one job. It's at least four, and the person who does one well is often the wrong person to do the next.

The companies that get engineering leadership right treat each stage transition as its own hire. The companies that don't spend six to twelve months every few years wondering why the leader who was working suddenly isn't.

‍

The four stages, and the four different jobs

Each stage requires a different profile. Not different in emphasis. Different in kind.

Stage 1: The Builder (10 to 40 engineers). The team is small enough that the VP still writes code, still knows every engineer personally, and still has direct visibility into every project. The job is high-velocity execution. Ship product, hire aggressively, keep the founder-CEO happy, and maintain enough architectural discipline that the codebase doesn't collapse. The Builder VP is often a technically strong senior engineer with two to three years of leadership experience. They win by moving fast.

Stage 2: The Scaler (40 to 150 engineers). The team has crossed the threshold where the VP can no longer know everyone or touch every project. The job becomes organizational. Layers of managers get added. A real hiring machine gets built. Performance management becomes a system rather than a series of conversations. The architecture gets rewritten, often twice, to survive the growth. The Scaler VP is a different person than the Builder. They win by building systems that let other people move fast.

Stage 3: The Operator (150 to 500 engineers). The team is now a real organization. Multiple product lines, distributed offices, engineering functions that need their own leadership (platform, infrastructure, security, data). The VP's job is running the org, not running engineering. Cross-functional politics, executive team dynamics, board-level communication, and long-horizon planning become most of the work. The Operator VP has often run a large engineering org before, at a public or late-stage private company. They win by building the leadership bench underneath them.

Stage 4: The Platform Executive (500+ engineers). At this scale, the VP or CTO is essentially running a small company inside a larger one. The job is capital allocation across engineering functions, technology strategy at the board level, external partnerships and acquisitions, and being the technology voice to the market. Very few candidates have run this before, and most companies at this scale have to recruit from an even smaller pool of proven Platform Executives.

The four profiles overlap in the middle. A strong Scaler can sometimes carry the front half of Stage 3. A strong Operator can sometimes carry the back half of Stage 2. But the two-stage gap almost never closes. The Builder is not the Operator. The Scaler is not the Platform Executive. Hoping otherwise is the source of most engineering leadership failures we see.

‍

Where boards get this wrong

Three patterns we see repeatedly.

Hiring for two stages ahead. A 30-person company hires a VP who ran engineering at a 400-person company. The reasoning is that they'll grow into the role and be prepared for the scale. What actually happens is that the Operator VP finds the Stage 1 work unfamiliar and slow. They try to install systems and process the company doesn't need yet. They hire senior lieutenants the company can't afford. Six months in, the CEO is frustrated that the engineering hire has slowed the team down. The Operator wasn't wrong. They were hired two stages too early.

Hiring for the current stage and ignoring the next transition. A 45-person team hires a Scaler who's perfect for the current work. Eighteen months later the team is 130 people and the Scaler is running out of runway. The board delays the conversation because the Scaler is doing well, and by the time they finally admit the stage has changed, the team has already been operating past its leader's capacity for six months. The company spends a full year in the transition.

Promoting the internal engineer who doesn't want the next job. A strong Builder gets promoted to Scaler and finds out they hate the systems work. They wanted to keep coding and running small teams. They stay in the title out of loyalty, quietly disengage, and the org drifts. The board realizes what happened only when the numbers start slipping, and now the exit conversation is harder because the person is popular and long-tenured.

‍

What the search should actually diagnose

The question is not "who is a great VP of Engineering." The question is "who is the right VP for this specific 18-to-24 month transition."

Three diagnostic questions that shape the search.

What stage is the team in right now, and what stage will it be in 18 months? If the answer is "roughly the same stage," you're hiring for consolidation and depth. If the answer is "materially different stage," you're hiring for the transition. These are different searches with different candidate profiles.

What's broken today that the current leadership can't fix? If the current VP or engineering director could fix it with more support, you don't need a new VP. You need to give them the support. If the current leadership can't fix it because the work has moved past their skill set, you need the next-stage VP, not a stronger version of the current one.

Is the current leader willing to be honest about the next stage? The best transitions we see are the ones where the incumbent engineering leader participates in scoping the next hire and, in many cases, actively wants to hand off. When that conversation happens twelve months early, the transition is orderly. When it happens twelve months late, it's a rescue.

‍

What separates the right hire from the impressive one

Four signals to look for in a candidate for any engineering leadership seat, adjusted for the stage.

They can name the transition they're best at. Strong candidates know whether they're a Builder, Scaler, Operator, or Platform Executive. They can describe the specific team-size range where they operate best. Candidates who claim to be great across all stages usually aren't great at any of them, or haven't done enough of the work to know the difference.

They have handed off before. Candidates who have led one stage and then handed off to a successor for the next stage know what a clean transition looks like. Candidates who have never handed off often conflate their tenure with their identity and won't leave when the company needs them to.

They ask about the next stage during the interview. Strong candidates want to know what the company expects in 18 to 24 months so they can decide whether the seat matches their skill set. Candidates who don't ask are usually either desperate for the job or haven't thought carefully about their own fit.

They've fired an engineering leader they liked. The scale VP job includes replacing engineering managers who worked at the last stage but can't work at the next one. Candidates who've done this before know how hard it is. Candidates who haven't often can't when the moment comes.

‍

The sharper version of the question

Stop hiring engineering leaders like you're hiring for a decade. Start hiring them for the specific stage transition the company is about to make. Get that right and the leader delivers what the company needs. Get it wrong and you'll spend twelve months explaining to the board why the impressive resume isn't producing results.

The best engineering leader for your company today is often not the best engineering leader for your company in three years. Plan for both. Hire for the first one, and structure the conversation so the transition to the second happens by design rather than by crisis.

‍

If you're scoping an engineering leadership search and the profile is written broadly, we'd be glad to pressure-test the stage question before you start.

‍