HR for the first 10, 50, and 200 employees: a staged founder's guide
What HR actually looks like at 10, 50, and 200 employees — the systems, hires, policies, and budgets that work at each stage, and the ones that backfire if…
On this page▾
- 0–10: founders are HR. No HRBP, no HRIS, no policies. A Google Doc handbook, a payroll provider, and offer letters from a template. The risk is hiring a 'Head of People' too early — almost always a mistake before 30 employees.
- 10–50: hire your first People Ops generalist (not a CHRO) around employee 25–35. Add an HRIS (BambooHR / Rippling / a local equivalent), structured hiring, a real handbook, and equity refreshes. This is where culture either compounds or fragments.
- 50–200: split HR into specialties — recruiting, HRBP, people ops, comp. Add performance reviews that actually work, calibration, leveling, an L&D budget, and your first DEI metrics. This is where most startups under-invest and pay for it at 300.
- Three rules across all stages: write things down before you need to, hire for the next stage not the current one, and don't copy Google's playbook — copy a company that was your size 18 months ago.
The most common HR mistake at every startup stage is the same: building the next stage's systems too early, or the previous stage's systems too late. This guide is the staged answer to 'what HR do we need right now?' for founders going from 0 to 200 employees — written for tech startups but applicable to any high-growth company.
At every stage, the right HR investment is one stage ahead of where you are — not two. Build the systems you'll need in 6–9 months, not in 24. Two-stages-ahead investment is the most expensive form of premature optimization in any startup, and HR is where it shows up first.
The staging rule
HR maturity scales with people, not revenue. A 12-person company with USD 20M in revenue has the same HR needs as a 12-person company with USD 200k in revenue. What you build is dictated by the number of human relationships you're now managing — which scales quadratically. At 10 people there are 45 pairwise relationships. At 50 there are 1,225. At 200 there are 19,900. Your systems must absorb that complexity; founders' calendars cannot.
Stage 1: 0–10 employees
What HR looks like
There is no HR function. The CEO does hiring, the CFO (or whoever does finance) runs payroll, and someone keeps the offer-letter template up to date. This is correct and you should not change it. The goal at this stage is not to build HR systems — it is to delay needing them by hiring slowly, deliberately, and only for roles you can describe in two sentences.
What to build
- A 1-page hiring rubric per role: what 'great' looks like, what 'no' looks like, and the 2–3 questions you'll ask every candidate.
- A simple offer letter template (Deel, LawDepot, or a lawyer-reviewed Google Doc). Include base, equity grant %, vesting, cliff, role, start date, at-will/notice clause.
- An equity plan (4 years, 1-year cliff, monthly thereafter) — set this up before grant #1, not after grant #5. Carta and Pulley are the standards; in Nepal/India use a SAR or phantom-equity scheme until you have a proper option pool.
- A payroll provider — Gusto (US), GreytHR (India), local accountant + spreadsheet (Nepal). Don't build payroll in-house ever.
- A 5-page Google Doc 'how we work' covering: working hours, communication norms, expense limits, time-off policy, dress code (if remote: none), security basics.
- An onboarding checklist — laptop, accounts, intro meetings, 30-day goals. Owned by the hiring manager, not 'HR.'
- Founder office hours (30 min/week per direct hire). This is your performance management, your career conversation, and your culture broadcast all in one.
Who does what
| Function | Who owns it at 0–10 |
|---|---|
| Recruiting | CEO (or hiring manager) |
| Onboarding | Hiring manager |
| Payroll | CFO / Head of Finance / outsourced |
| Equity | CEO + lawyer |
| Performance feedback | Manager 1:1s |
| Conflict resolution | CEO |
| Employee relations | CEO + lawyer if it escalates |
| Compliance | Lawyer on retainer (~$500/mo) |
| Benefits / leave | CEO sets policy; nothing administered |
The most common 0–10 mistakes
- Hiring a 'Head of People' or 'CHRO' as employee #8. They will have nothing to do for 18 months, get bored, and quit — or worse, build systems for a 200-person company and slow you down.
- Buying an HRIS. A BambooHR subscription for 9 people is $50/month for features you don't need. A Google Sheet works until 25.
- Writing a 40-page handbook. Nobody reads it; it dates instantly; it gives you false security on culture.
- Skipping the offer letter or doing it on a verbal handshake. Equity disputes from this era are the single most common reason early-stage cap tables get sued.
- Not running payroll properly because 'we're small.' One missed tax filing in Nepal/India/US triggers penalties that compound monthly.
What success looks like at 10
- You can recite every employee's title, manager, and start date from memory.
- Offers go out within 48 hours of decision.
- Onboarding is 'great laptop, real meetings, clear 30-day goals' — not 60 slides.
- Equity grants are documented, vesting is tracked, and everyone has a copy of their grant agreement.
- Payroll runs on time every cycle without anyone thinking about it.
Stage 2: 10–50 employees
What HR looks like
This is the most important HR stage and the one founders consistently get wrong. Somewhere between 20 and 30 employees, three things break simultaneously: the CEO can no longer be in every interview, the Google Doc handbook contradicts itself, and the first 'we have a problem with X' employee-relations issue arrives. The answer is your first People Ops hire — and the answer is almost never a CHRO.
Your first HR hire
At 25–35 employees, hire a strong People Operations generalist (5–10 years experience) who has done this work hands-on. They will run hiring ops, build the handbook, own the HRIS, handle the first employee-relations issues, and own onboarding. Title: 'People Ops Lead' or 'Head of People' (singular). Comp: USD 80–140k in NA, USD 18–32k in Nepal, INR 18–35L in India. Do not hire a CHRO until 150+ employees — at this stage they're over-qualified and under-utilized, and they'll build strategy decks instead of fixing actual problems.
What to build
- HRIS: BambooHR or Rippling (US), HiBob or Personio (EU), GreytHR or Keka (India), Deel for multi-country contractor / EOR. Pick on day 1 of the People Ops hire and migrate within 60 days.
- Structured hiring: scorecards per role, calibrated interviewers, debrief discipline, and a hiring committee or 'bar raiser' for senior roles. This is the highest-ROI process you'll ever install.
- A real handbook (not 40 pages — 15–20). Cover: how we work, comp philosophy, leave (parental, sick, vacation), expense policy, code of conduct, equipment, security, IP, anti-harassment, grievance.
- Comp bands — 3–5 levels per function, P50 anchored to a benchmark (Levels.fyi, Carta, Mercer, Aon, AmbitionBox in India). Bands are USD-denominated for global teams; local currency for single-country teams.
- Equity refresh program — start at year 3 to combat the 'cliff retention dip.' Budget 0.05–0.15% per refresh for senior staff.
- Performance lite — quarterly check-ins, written, with 3 prompts: what went well, what to improve, what support is needed. NO numerical ratings yet.
- Manager training (3–4 sessions): how to run a 1:1, how to give feedback, how to set goals, how to handle the hard conversation. Even basic manager training cuts attrition meaningfully.
- Anti-harassment policy + reporting channel — non-negotiable from employee 15 onward. India: PoSH committee is legally required at 10+ employees. Nepal: required under the Sexual Harassment at Workplace Act 2015 for any establishment with female workers.
- Benefits decisions: health insurance, retirement match (US 401(k), India EPF, Nepal SSF + CIT). The decision is whether to match, not whether to offer — most countries have mandatory schemes.
- Engagement signal: a short pulse survey every quarter (eNPS + 3 open questions). Not Glint, not Culture Amp yet — a Typeform works.
Comp at 10–50
| Element | 10–25 employees | 25–50 employees |
|---|---|---|
| Bands | Informal, in CEO's head | Documented, 3–5 levels per track |
| Benchmarking | Founder gut + 2–3 reference points | Annual against Mercer/Levels/Carta |
| Equity policy | Per-hire negotiation | Standard grant table by level |
| Refreshes | None | Start at year 3 |
| Pay transparency (internal) | None | Bands visible to managers |
| Pay transparency (external) | None | JD-level ranges (recommended) |
| Annual increase budget | Ad hoc | 3–6% merit + 1–2% promo pool |
Who does what
| Function | Who owns it at 10–50 |
|---|---|
| Recruiting | People Ops + hiring manager (recruiter hire at ~40) |
| Onboarding | People Ops + hiring manager |
| Payroll | People Ops + outsourced provider |
| HRIS | People Ops |
| Equity | CEO + CFO + People Ops |
| Performance | Managers, supported by People Ops |
| Employee relations | People Ops + outside counsel on escalation |
| Compliance | People Ops + employment lawyer on retainer |
| L&D | Manager + ad-hoc budget per employee |
The most common 10–50 mistakes
- Hiring HR too late (employee 45 instead of 25). Founders absorb work that's no longer leveraged for their time, and the first employee-relations issue is handled by someone with no training.
- Hiring a strategic CHRO instead of an operational generalist. The strategic CHRO will write a 'culture deck' and a 5-year people plan; you needed someone to onboard 12 hires this quarter.
- Adopting a heavy performance system (formal ratings, calibration, 360s) before you have 50 people. The overhead is enormous and the data is too sparse to be meaningful.
- Skipping comp bands until the first 'why does X earn more than me' moment. By then, every comp conversation is a negotiation, not a process.
- Writing values without rituals. Values that aren't repeated in onboarding, performance reviews, and promotion decisions don't shape behavior.
- Building a careers page but not a candidate funnel. You announce roles, get applicants, and have no system to triage them — leading to candidate-experience disasters and missed hires.
What success looks like at 50
- Hiring runs to a written process; debriefs are calibrated; offers are extended within 5 business days of a 'yes.'
- Every employee has a manager, a written role, a comp band, and a 1:1 cadence.
- Quarterly check-ins happen for 90%+ of employees without prompting.
- Attrition is tracked and you know your regrettable vs non-regrettable rate.
- There is a documented grievance and harassment policy with an active reporting channel.
- You can answer 'what does someone at level X earn?' within 30 seconds.
Stage 3: 50–200 employees
What HR looks like
HR becomes a function with specialties. Your People Ops generalist either grows into a Head of People role (rare but excellent when it happens) or hands off pieces to specialists. By 100 employees you'll have 3–5 people on the team; by 200, 6–10. The big shift is from operational ('did onboarding happen?') to programmatic ('does our hiring system produce better hires than last year?').
The team you build
| Role | When to hire | Comp NA / EU / South Asia (USD) |
|---|---|---|
| Head of People (former People Ops Lead, promoted) | ~50 | $140–220k / $110–180k / $30–55k |
| Recruiter (in-house) | ~50–60 | $80–120k / $65–100k / $14–28k |
| People Ops Manager / Generalist (specialty: ops + HRIS) | ~75 | $70–110k / $55–90k / $12–22k |
| Senior Recruiter / Recruiting Lead | ~100 | $110–160k / $85–130k / $20–35k |
| HRBP (one per 60–80 employees) | ~120 | $100–150k / $80–125k / $18–32k |
| Comp & Benefits analyst (or fractional) | ~150 | $90–140k / $70–115k / $15–28k |
| L&D / Manager Development specialist | ~150–200 | $80–130k / $65–105k / $14–25k |
| DEI / People Analytics (1 person, sometimes combined) | ~150–200 | $100–160k / $80–125k / $18–32k |
What to build
- Formal performance reviews — semi-annual, with light ratings (3–5 levels: significantly exceeds → does not meet). Calibration sessions led by HRBP, decisions visible to managers, not employees.
- Career ladders for every track — IC and management dual ladders. Each level has expectations, scope, impact, behaviors. Publish internally; selectively externally.
- Promotion process — quarterly windows, written promotion case, manager + skip-level + HRBP sign-off. Cap promotion velocity at 15–20% of eligible per cycle to maintain calibration.
- Comp cycle — annual, calibrated, with merit pool, promo pool, equity refresh pool, and market adjustment pool separated and reviewed by the leadership team.
- Manager development program — 8–12 hours over the first 6 months in role. Topics: 1:1s, feedback, performance, comp conversations, conflict, hiring, your culture. Internal or via FranklinCovey / LifeLabs / a local equivalent.
- L&D budget per employee ($500–2,000 USD or local equivalent). Self-directed with manager approval; report quarterly on uptake.
- Engagement: quarterly Culture Amp / Glint / Lattice survey with action commitments per team. Without action commitments the survey is theater.
- DEI metrics: representation by level, hiring funnel pass-through by gender (and ethnicity where legally collectible), pay-equity audit annually. Required reading: your jurisdiction's discrimination law — varies widely.
- People analytics: regretted attrition by team and tenure, hiring funnel conversion, time-to-fill, time-to-productivity, engagement score, comp ratio. One dashboard, monthly to the leadership team.
- Total rewards review — by 150 employees you have benefits, equity, base, bonus, perks. Document the philosophy or it becomes a series of one-off concessions.
- Offboarding & RIF playbook — by 100 employees you will have run at least one performance termination and possibly your first small layoff. A written process, severance grid, and exit-interview protocol prevent the worst outcomes.
Performance reviews at 50–200
Every startup at 80–150 employees debates 'should we have ratings?' Forced rankings are unambiguously bad (Adobe, Microsoft, Deloitte all moved away). But the absence of any rating means every comp and promo decision is a re-litigation. The middle path: 3–5 rating bands, no forced distribution, calibrated by HRBPs, used as input into comp and promo discussions — not communicated as a number to the employee.
The most common 50–200 mistakes
- Importing Google or Meta's full HR stack. Their problems are not your problems. A 120-person company does not need OKR software, a learning platform, a survey platform, and a recognition platform — pick two.
- Hiring HR specialists before you have generalist coverage. A DEI Manager at 80 employees with no HRBP is a hire that fails — the role lacks the operational platform to be effective.
- Skipping calibration. Promotion and comp decisions made manager-by-manager produce a level-creep that is irreversible inside 18 months. Calibrate from cycle one.
- Not building a comp philosophy. By 200 you'll have 4 different ad-hoc comp logics. Write the philosophy at 60.
- Ignoring middle-management strength. The single highest predictor of culture survival from 50→200 is the quality of your middle managers. Invest there or culture dilutes regardless of any other investment.
What success looks like at 200
- There is a published career ladder for every role family.
- Comp cycles run on schedule; calibration produces decisions managers feel are fair.
- Performance reviews happen on time for 95%+ of employees.
- Regrettable attrition is below 10% annually (US/EU benchmark) or below 14% (South Asia tech benchmark).
- You can articulate your culture in 3 sentences and point to 5 practices that enforce it.
- HR runs without the CEO being involved in the day-to-day, freeing the CEO for strategy, hiring at the top, and customer time.
HR headcount by function
| Employees | Total HR FTEs | Ratio HR:Employees | Composition |
|---|---|---|---|
| 0–10 | 0 | — | Founders do HR |
| 10–25 | 0–1 | 1:25 | Maybe 1 People Ops generalist (late stage) |
| 25–50 | 1–2 | 1:35 | People Ops Lead + part-time recruiter |
| 50–100 | 2–4 | 1:30 | Head of People + Recruiter + People Ops |
| 100–200 | 4–8 | 1:25 | Add HRBPs, Comp, L&D, possibly DEI |
| 200–500 | 8–18 | 1:30 | Specialists per function, often regional HRBPs |
HR budgets, stage by stage
HR spend (excluding salaries paid to non-HR employees) typically runs 1–2% of total compensation budget. Below is a stage-by-stage breakdown for a representative SaaS / tech startup, in USD.
| Stage | Tooling | Outsourced (legal, advisory, EOR) | Programs (L&D, engagement, perks) | Total annual HR opex (ex-team salaries) |
|---|---|---|---|---|
| 0–10 | $2k | $8k | $5k | ~$15k |
| 10–50 | $15k | $25k | $25k | ~$65k |
| 50–100 | $40k | $45k | $80k | ~$165k |
| 100–200 | $80k | $60k | $200k | ~$340k |
What NOT to build at each stage
| Stage | Do NOT build |
|---|---|
| 0–10 | HRIS, 40-page handbook, CHRO hire, formal performance system, DEI program, engagement platform, OKR tool, equity refresh program, anonymous reporting platform. |
| 10–50 | Forced rankings, 360 reviews, dedicated DEI hire, sabbatical policy, complex bonus plan, multiple HRBPs, learning management system, in-house recruiting team beyond 1 recruiter. |
| 50–200 | Workday, a global mobility program if you're single-country, a leadership academy run in-house, separate functions for engagement / learning / DEI (combine until 200+), regional HRBPs unless you're truly multi-country. |
The transition moments
Three moments require deliberate change. Founders consistently miss the signals; HR teams consistently spot them 2–3 quarters too late.
- 1~25 employees: founder can't run hiring + cultureSignal: hiring slowdown, candidate experience dropping, founders complaining about being in interviews all day. Move: hire People Ops Lead, install structured hiring.
- 2~75 employees: managers are managing managersSignal: the same questions reach the CEO three layers down, comp decisions feel inconsistent, employees say 'I don't know what's expected of me.' Move: invest in manager development, install performance lite, document career expectations.
- 3~150 employees: culture starts to fragment by teamSignal: one team is 'great to work at,' another is leaking. Engagement scores diverge by manager. Move: HRBPs per business unit, calibrated performance, formal leadership program for senior managers.
FAQ
Frequently asked questions
When should we hire our first HR person?
Between 25 and 35 employees in most tech startups. Earlier if you're in a heavily regulated industry (healthcare, fintech) or operating across multiple countries. Later only if a founder is genuinely operating as a part-time People lead and enjoys it — which is rare beyond 35.
Should our first HR hire be a generalist or a recruiter?
Generalist, almost always. Recruiting can be filled by agencies, embedded contractors, or a part-time hire until ~40 employees. The generalist owns the operating system — handbook, comp, HRIS, employee relations — that everything else sits on.
We're a Nepal/India-headquartered startup. Does this guide apply?
Mostly yes — the stages, hires, and systems are universal. Adjust comp ranges to local benchmarks (×0.3–0.5 of NA in Nepal; ×0.35–0.55 in India for tech roles), prioritize local compliance (PoSH, EPF, SSF, gratuity) earlier, and consider that HR talent in these markets often has narrower exposure to modern People Ops practices — interview specifically for that experience or invest in upskilling.
What if we're remote-first?
The stages are the same; the tooling and rituals change. Add: async-first documentation, time-zone overlap rules, a virtual onboarding program, regular off-sites (twice a year), explicit comp geography policy. Subtract: office perks, expensive workplace tooling. Most remote-first companies need their first People Ops hire slightly earlier (~20 employees) because culture doesn't form by osmosis.
Do we need an HR consultant or a fractional CHRO instead of a full-time hire?
A fractional CHRO (1–2 days/week) is a great option from ~15 to ~40 employees if you don't want to commit to a senior full-time hire yet. Past 40 it usually breaks down — the operational load needs an in-house owner. Pure consulting (no operational ownership) is less useful: HR problems require someone embedded in the team.
How do we know if our People Ops Lead is the right person to grow into Head of People?
Three signals: (1) they push back on bad ideas from the CEO with data, (2) they recruit better than you (their hires last and grow), (3) they design systems that work without their daily involvement. If any of these are missing at 60–80 employees, hire above them before 120.
What's the single biggest predictor of HR success across all three stages?
Founders who treat People as a function with rigor — same operational expectations as Engineering or Finance — not as a soft 'culture' afterthought. The HR systems that work are designed and shipped with the same discipline as any other product. The ones that fail are negotiated, postponed, or delegated to someone without authority.
- First Round Review — The People Operations Playbook — First Round
- Lattice — People Strategy by Stage — Lattice
- Carta — State of Startup Compensation — Carta
- Levels.fyi — Levels.fyi
- Mercer Total Remuneration Survey — Mercer
- Nepal Labour Act 2074 (2017) — Nepal Law Commission
- India Ministry of Labour — PoSH Act — Ministry of Labour, India
Read next
All playbooksSide-by-side HRIS comparison for distributed teams. Pricing, payroll coverage, EOR, compliance, and which platform wins for Nepal, India, US, and EU hiring in…
Statutory severance minimums in 18 countries — Nepal, India, US, UK, Germany, France, UAE, Singapore, and more. Notice periods, formulas, and the practical…
A complete cash + equity band table for a 42-person Series A engineering org in Kathmandu — what we pay, why, and what we'd change knowing what we know now.