What percentage of burn should be headcount at Series A?

People costs typically represent 65–80% of gross burn at 20–100 person SaaS companies. If headcount exceeds 85% without corresponding ARR growth, efficiency metrics (burn multiple, ARR per employee) usually degrade.

How far ahead should startups plan headcount?

Rolling 12-month plan with quarterly true-ups. Board-approved annual plan, revised each quarter based on ARR vs plan and runway. Open reqs beyond 90 days without pipeline should be challenged.

Who should approve new headcount?

Standard at Series A: manager requests, finance models impact, CEO or COO approves against plan. Offers above comp band or outside plan require explicit exec sign-off before Greenhouse offer letter.

All articles

Playbooks12 min readJuly 3, 2026

Headcount Planning for Startups: The Model COOs Use at 20–100 Employees

Hiring is your biggest burn lever and your least synchronized system. Here is the headcount model Series A COOs use to align recruiting, finance, and the board before reqs go live.

Headcount planningHiringComp bandsRunway

By AethelLayer Editorial · Executive Layer Insights

Headcount planning model connecting open roles to runway and budget

Recruiting opened six roles in Q2. Finance modeled three. Two offers went out above comp band because the hiring manager copied last year's spreadsheet. Runway in the board deck was wrong by seven weeks. Nobody lied — the systems never shared a model.

Headcount plan spreadsheet alternative with live Greenhouse and finance sync — Headcount planning works when reqs, comp bands, and runway share one source of truth.

The quarterly headcount model structure

Column	Purpose	Owner
Role / level	Title, department, priority (P0–P2)	Hiring manager
Target start	Month req opens → offer → start	Recruiting
Fully loaded cost	Base + benefits + payroll tax + equipment	Finance
Runway impact	Cumulative monthly burn delta	Finance / COO
Status	Approved / open / offer / started / frozen	Recruiting
Comp band check	Within band Y/N, exception approver	People + Finance

Approval workflow that prevents offer-stage surprises

Quarterly: leadership sets headcount budget tied to ARR plan and runway floor (e.g. never below 12 months)
Req open: manager submits role; finance auto-checks against remaining budget
Pipeline: Greenhouse stages sync; time-to-fill tracked against plan
Pre-offer: comp-band validation against finance model — block or flag exceptions
Post-start: actual loaded cost feeds next month's burn and runway recalc

Median conflict rate

Teams without pre-offer finance checks report budget conflicts on 18–25% of senior offers at Series A — usually discovered after verbal accept, when renegotiation is expensive.

Scenario rows every plan needs

Base plan: approved hires only
+3 engineering if product milestone hits by Sept 30
Freeze plan: hiring pause if runway <10 months or ARR misses by >15%
Acceleration plan: +2 AEs if pipeline coverage >3x for two consecutive months

Spreadsheet vs connected ops layer

Spreadsheets work until ~25 employees. After that, Greenhouse, Rippling or Deel, and Xero diverge weekly. Hiring agents that read live pipeline and finance data catch conflicts at req approval — not at board review. AethelLayer ties Greenhouse stages to comp bands and runway impact in one workspace.

FAQ

What percentage of burn should be headcount at Series A?: People costs typically represent 65–80% of gross burn at 20–100 person SaaS companies. If headcount exceeds 85% without corresponding ARR growth, efficiency metrics (burn multiple, ARR per employee) usually degrade.
How far ahead should startups plan headcount?: Rolling 12-month plan with quarterly true-ups. Board-approved annual plan, revised each quarter based on ARR vs plan and runway. Open reqs beyond 90 days without pipeline should be challenged.
Who should approve new headcount?: Standard at Series A: manager requests, finance models impact, CEO or COO approves against plan. Offers above comp band or outside plan require explicit exec sign-off before Greenhouse offer letter.

Book a demo