Build a Startup Financial Model

Why Your Financial Model Matters

A startup financial model serves two purposes that founders often conflate: it’s a management tool (to understand your business and make decisions) and a fundraising document (to communicate your trajectory to investors).

Most early-stage models are built primarily for fundraising and fail at the management purpose. The best models do both — they’re built on operational assumptions that founders actually believe and update regularly, so the projections remain a living picture of the business, not a one-time presentation deck.

This guide walks you through building a model that works for both.

---

Step 1: Define Your Revenue Model

Before building any spreadsheet, get clear on how your company generates revenue. The structure of your model follows directly from this.

Common startup revenue models

Model	Revenue driver	Key metrics
SaaS subscription	MRR/ARR	ARPU, churn, NRR
Usage-based	Consumption volume	Revenue per unit, usage growth
Marketplace	GMV × take rate	GMV, attach rate, order frequency
Transactional	Transactions × fee	Transaction volume, fee per transaction
Services/consulting	Hours × rate	Utilization rate, headcount
Hybrid	Mix of above	Combination

Most real businesses are hybrids. A SaaS company might have subscription revenue plus professional services. A marketplace might have take rate revenue plus a SaaS subscription layer. Model each stream separately.

Establish unit economics first

Before projecting total revenue, nail down the per-unit economics:

• What is your average deal size (ACV) or ARPU?
• What is your gross margin per customer or transaction?
• How long does a customer stay (LTV)?
• What does it cost to acquire one customer (CAC)?

These are the inputs your revenue forecast is built from. If you don’t know them from data, estimate conservatively and flag them as assumptions.

---

Step 2: Build the Revenue Forecast (Bottom-Up)

The cardinal rule: build revenue from the bottom up, never top-down.

Top-down: “The TAM is $10B. We’ll capture 2% = $200M.” This is meaningless to an experienced investor.

Bottom-up: “We’ll have 3 sales reps by Q2, each closing 2 deals/month at $25K ACV = $150K new ARR/month by Q3.”

Bottom-up revenue model example (SaaS)

Month 1:
  Starting ARR: $200,000 (existing)
  New ARR: 3 AEs × 2 deals × $15,000 ACV = $90,000
  Churned ARR: $200,000 × 1.5% monthly churn = -$3,000
  Ending ARR: $287,000

Month 2:
  Starting ARR: $287,000
  New ARR: 3 AEs × 2.2 deals (improving) × $15,000 = $99,000
  Churned ARR: -$4,305
  Ending ARR: $381,695

Build this month by month for 36 months. The key inputs are:

• Number of AEs/sales reps by month (from your hiring plan)
• Ramp time (new AEs typically reach quota in months 3–6)
• Quota attainment assumption (typically 70–80% of quota for planning purposes)
• Average ACV
• Monthly churn rate

For growth-stage companies, also model expansion revenue separately — upsells and seat expansions from existing customers.

---

Step 3: Map Headcount and Payroll

Payroll is typically 60–80% of total burn for most startups. Getting this right is critical.

Build a headcount plan by role

Role	Q1	Q2	Q3	Q4
Engineering	5	7	9	11
Sales (AE)	2	4	6	8
Sales (SDR)	1	2	3	4
Marketing	1	2	2	3
Customer Success	2	3	4	5
G&A	2	2	3	3
Total	13	20	27	34

For each role, estimate:

• Base salary: Use market data (levels.fyi, Glassdoor, Carta comp data)
• Fully-loaded cost multiplier: 1.2–1.3× of base salary to include employer taxes, benefits, equipment, and workspace
• Hiring lag: Build in 1–2 months between “hire decided” and “person starts”

Example:

• Engineering hire at $150K base = ~$187,500 fully-loaded annual = ~$15,600/month

---

Step 4: List Operating Expenses

Beyond payroll, startups have four main opex categories:

Infrastructure and tools

• Cloud hosting (AWS, GCP, Azure): typically 5–15% of revenue for SaaS
• SaaS tools: Salesforce, HubSpot, Notion, GitHub, Slack, etc.
• APIs and third-party services

Sales and marketing

• Paid acquisition (Google, LinkedIn, Meta ads)
• Events and conferences
• Content production
• PR and outbound tooling

Office and equipment

• Rent (or co-working credits)
• Laptops and hardware (typically $2,000–$3,500/employee, upfront)
• Office supplies and perks

G&A (General and Administrative)

• Accounting and audit: $30,000–$100,000/year depending on complexity
• Legal: $20,000–$60,000/year (more if actively fundraising or in contract-heavy negotiations)
• Insurance (D&O, E&O, general liability)
• Finance software (QuickBooks, Pilot, Brex)

---

Step 5: Build Cash Flow and Runway

With revenue and expenses modeled, you can produce the core outputs:

Monthly P&L structure

Revenue:
  Subscription/MRR × 12
  Services revenue

Cost of Revenue:
  Infrastructure
  Customer Success (if considered COGS)
  Third-party APIs

Gross Profit = Revenue - COGS
Gross Margin % = Gross Profit ÷ Revenue

Operating Expenses:
  R&D (Engineering + Product)
  Sales & Marketing
  G&A

EBITDA = Gross Profit - OpEx
Net Burn = Cash out - Cash in (monthly)

Ending Cash = Previous Cash - Net Burn
Runway (months) = Ending Cash ÷ Average Monthly Burn

Scenario modeling

Always build at minimum two scenarios:

Scenario	Revenue assumption	Hiring pace	Key signal
Base	80% of targets hit	As planned	Expected case
Downside	50–60% of targets	Delayed hiring	Stress test

Show investors both. A downside scenario that still shows 12+ months runway signals that you’ve thought through risk, not that you’re being pessimistic.

---

Step 6: Create Investor-Ready Summary Outputs

The investor-facing financial model summary should fit on one page and include:

• Monthly ARR (or MRR for earlier stage)
• Monthly burn rate (net cash consumed)
• Cumulative cash balance
• Headcount growth
• Key assumptions tab: one row per major assumption, current estimate vs. range

Red flags investors look for

• Revenue growing while burn stays flat (implies impossible sales efficiency)
• Gross margins that don’t reflect the actual business model
• No churn or churning at exactly the industry average (looks like a guess)
• Projections that break from recent trend without explanation
• Headcount that doesn’t support the revenue plan (too few salespeople to hit revenue targets)

---

Key Takeaway

A financial model’s value is not in the precision of its outputs — it’s in the clarity of its assumptions. Build bottom-up. Know your unit economics cold. Update the model monthly with actuals so the projections drift less from reality. And when investors stress-test your assumptions, be able to explain the logic behind every number. A founder who can do that has already passed the most important test the model is used for.