How to Build a High-Performance Leadership Team for Startup Scaling

Your VP of Sales just quit. Or worse — they stayed, and you're now watching them manage a team of twelve with the same playbook that worked when there were three of them. Sales is down 18% quarter over quarter, the board wants answers, and you're realising that the leadership team you built for the seed round isn't the one you need for the Series B.

I've watched this movie from the inside, twice as a founder and three times as an advisor to startups between 15 and 200 people. Building a high-performance leadership team for startup scaling isn't about hiring impressive résumés. It's about sequencing, honest scorecards, and knowing when a role has outgrown the person holding it — including yourself.

This isn't a motivational piece about "hustle." It's a working manual based on what I've seen break and what I've seen hold.

Why your first leadership team will fail you (and why that's normal)

Nobody warns you that leadership team building is a replacement exercise as much as an addition exercise. In my own second startup, I hired a head of engineering at 12 people who was extraordinary — hands-on, fast, wrote half the codebase. At 45 people, she was drowning in coordination work she'd never signed up for, and both of us pretended not to notice for about five months. That pretence cost us two senior engineers who quit because the org felt rudderless.

The pattern repeats. A generalist founder-CEO runs things fine until roughly 30-40 people, at which point the number of decisions crossing your desk becomes physically impossible to process. A jack-of-all-trades operations lead works until you need someone who actually understands unit economics. Nobody's at fault. The job changed shape.

The headcount triggers that actually matter

Here are the thresholds I've seen hold up across a dozen companies, give or take:

• Around 15-20 people: you need a real first-line lead per function. Not a manager title — a person whose calendar is mostly other people's problems.
• Around 40-50: a dedicated finance lead, even part-time fractional. Founders doing their own runway modelling is where nasty surprises are born.
• Around 70-90: a COO or equivalent who owns cross-functional execution. This is the role most founders resist and most need.
• Around 120-150: functional VPs who manage managers. Below this, you'll over-hire and create layers nobody asked for.
• Engineering-specific: a dedicated VP Engineering once you pass 25-30 engineers. Before that, a strong tech lead plus you is often enough.

These aren't laws. They're the points where I've seen the same cracks appear. If you're past a threshold and the role doesn't exist, you're not saving money — you're borrowing against future chaos at a terrible interest rate.

Start with the job, not the person

Most founders hire leadership backwards. They meet an impressive person, get excited, then retro-fit a role to justify the hire. Six months later they wonder why the org chart looks like a game of Tetris.

Do the reverse. Write a one-page scorecard before you post anything: what outcomes does this role own in the next 12 months, what does success look like at month 3, 6, and 12, and what does failure look like? I'll admit, my first attempt at this looked like a job description with the word "ownership" sprinkled in eleven times. It took me three tries to write one that a stranger could actually evaluate against.

What the scorecard should contain

1. Three to five outcome statements, each measurable.
2. The one or two hard trade-offs the role has to make (e.g. "growth pace vs. gross margin").
3. The failure modes you're most worried about.
4. What this person will not be doing — the boundaries that prevent role creep.

That last point is underrated. In a scaling startup, every senior person drifts toward whatever the founder is anxious about that week. Boundaries in writing give you something to point at when the drift starts.

What is the 70-20-10 rule for leadership?

The 70-20-10 rule for leadership is a rough model of where leadership capability actually comes from: roughly 70% from challenging on-the-job experience, 20% from relationships and mentoring, and 10% from formal training. The percentages are a heuristic, not a precise measurement — but the proportion matters enormously for how you invest.

Here's what that means in practice. Most startups I know spend their entire leadership development budget on the 10% — a course, a coach, an offsite workshop — and then wonder why behaviour doesn't change. If 70% of growth comes from the job, then the job design is your development programme.

Designing roles that grow people

Concretely, that means deliberately putting stretch into the work:

• Give a first-time manager a project with genuine ambiguity and real stakes — not a fake "leadership opportunity."
• Pair the stretch with the 20%: a mentor who has done the thing before, a peer group of other leads, and a fortnightly conversation that isn't a status update.
• Use the 10% for targeted input — a negotiation workshop before a big renewal round, not a general "leadership fundamentals" course in month one.

I removed a 3-day leadership bootcamp from my last team's onboarding and replaced it with a 6-week internal project where each new lead owned a cross-functional deliverable. Retention of new leads past month four went from "we keep losing them" to "most of them stayed." Not a scientific experiment — but the direction was clear.

What are the 7 C's of high-performing teams?

The 7 C's of high-performing teams are a behavioural checklist you can use in any leadership offsite: Clarity, Communication, Collaboration, Commitment, Capability, Consistency, and Candour. Frameworks like this are only useful if you can audit against them, so here's how I actually use each one.

C	The question to ask your team	The symptom when it's missing
Clarity	Can each leader state the company's top three priorities without looking at a slide?	Mismatched priorities, duplicated effort
Communication	Do decisions reach the people they affect within 48 hours?	Rumours filling the vacuum
Collaboration	Do leaders solve cross-functional problems together, or escalate to you?	Founder as permanent bottleneck
Commitment	Is there one plan, or a plan plus three private ones?	Passive resistance, quiet non-execution
Capability	Could this leader do the next stage's version of their job?	Constant firefighting, missed builds
Consistency	Do people get the same answer from different leaders?	Teams playing the org chart
Candour	Does bad news travel up faster than good news?	Surprises at the board meeting

The one I'd die on a hill for is candour. If your leaders can't tell you the project is failing until it has failed, nothing else on that list matters. You can build clarity with a strategy doc, but you can't build candour with a template.

What is the 30-60-90 rule in leadership?

The 30-60-90 rule in leadership is a structured onboarding plan for a new senior hire: in the first 30 days they learn (listening, mapping stakeholders, no big changes), by day 60 they diagnose (present their read of the situation and a plan), and by day 90 they execute (deliver a first visible win against the scorecard).

Startups routinely skip this and then blame the hire. I did, once. A new VP of Marketing joined in the middle of a rebrand, I threw her into execution on day four, and by day 60 she was making decisions from panic rather than diagnosis. We agreed to part ways at month five. In hindsight, I'd given her neither the listening window nor the scorecard, so she was set up to fail.

Running a 30-60-90 properly

• Day 0: hand over the scorecard. Explicit. Written.
• Days 1-30: 15 scheduled listening meetings, including people who will be hardest to hear from. No restructuring.
• Day 30-60: the hire presents their diagnosis and plan to the leadership team. You correct in the room, not in DMs afterwards.
• Day 90: one visible deliverable — a process fixed, a metric moved, a decision made.
• Day 90 review: scorecard in hand, both of you answer "on track / not on track" honestly. No "let's see how Q2 goes."

The heavy lifting is on you as the founder. If you can't write the scorecard or hold the day-90 review, don't blame the hire when it doesn't work.

What are the 5 KPIs of a team leader?

The 5 KPIs of a team leader are team outcomes, team health, people development, decision speed, and stakeholder trust — one lagging outcome metric and four leading indicators. Every one of these can be operationalised for a startup without a spreadsheet empire.

What each KPI looks like in a real team

• Team outcomes — the thing the function exists to deliver. Revenue, delivery time, uptime, whatever it is, tied directly to the scorecard.
• Team health — voluntary attrition, and one honest anonymous signal per quarter. I've used a single question: "would you recommend working on this team?" with a free-text box. Cheap, brutal, useful.
• People development — how many people on the team have grown into a bigger role in the last 12 months. If the answer is zero for two years, the leader isn't leading.
• Decision speed — median time from "we need to decide X" to "X is decided." Founders feel this in their bones when they're being bottlenecked, but your leaders should be measured on it too.
• Stakeholder trust — how often other leaders get blindsided by this leader's area. Track it informally; it surfaces fast in offsites.

The trick is calibration fairness: don't apply the same thresholds at every stage. A first-time lead at 30 people and a VP at 150 should be measured on the same five dimensions, but with very different expectations of each.

What to do when a leader outgrows the role (or the role outgrows them)

The hardest conversation in scaling is the one where you tell a loyal, competent leader that the role has changed. I've had it four times. Every time I delayed it by roughly two quarters longer than I should have, and every delay cost more — lost team members, missed quarters, personal resentment building on both sides.

The alternative is not "keep them in place." The alternatives are:

• Redefine the role around what they're genuinely exceptional at, and hire the missing capability alongside them.
• Move them sideways into a role that matches their strengths, with a title and compensation that reflects reality.
• Coach them through the transition — but only if you can honestly name what's not working and give them a real runway. Six months is a runway. Two years of hoping is not.

Whichever path you choose, write down what "working" looks like within 30 days and hold yourself to the review. Vague hope is the most expensive leadership development strategy there is.

What actually makes the difference

The teams I've seen scale well weren't built by hiring the most impressive résumés. They were built by founders who treated leadership team design as a recurring job — reviewing the org every six months, holding the day-90 reviews they said they'd hold, and being honest about which leader the next stage actually needed.

Which means the question worth sitting with is not "who do I hire next" but "what is the next version of this company actually asking of its leaders, and who in this room can deliver it?" Ask it out loud at your next leadership offsite. If nobody in the room can answer it without hedging, you already have your answer — and it probably isn't the one you wanted.