Startup valuation is the unglamorous part of fundraising: a structured estimate of what a private company is worth, used to set per-share price for the round. Unlike public markets, there is no daily quote — the number is negotiated between founder and investor and depends heavily on which valuation method anchors the conversation. Pre-revenue companies rely on qualitative scoring; growth-stage businesses can support DCF or cap table modeling tied to detailed financials. The skill is matching method to stage, then triangulating with a second method to defend the range.
For exit-related valuation work — modeling per-share proceeds across share classes — see waterfall analysis. This article focuses on entry valuation: the number the lead investor and founder agree on at the term sheet.
Why Startup Valuation Is Hard
Three structural problems make this messier than public-company analysis. No financial history — pre-revenue companies can’t support traditional discounting. Extreme outcome variance — most early-stage startups fail; even within winners there’s a 100x gap between modest exits and breakouts. Intangible-heavy balance sheets — a SaaS company’s value is team, code, customers, and brand, none of which sit at fair value on the books.
The practical consequence: valuations are ranges, not points. Anyone who tells you a seed-stage startup is “worth $7.42M” is overselling precision.
Match Method to Stage
| Stage | Revenue | Recommended methods | Typical multiple | Discount rate |
|---|---|---|---|---|
| Pre-seed | $0 | Scorecard, Risk factor summation | n/a | 50–60% |
| Seed | $0–$500K ARR | VC method, Scorecard | 15–25x revenue | 40–50% |
| Series A | $1–5M ARR | VC, Comparables, light DCF | 10–20x | 30–40% |
| Series B | $5–20M ARR | DCF, Comparables, Precedent | 8–15x | 25–35% |
| Series C+ | $20M+ ARR | DCF, Comparables, Precedent | 5–12x | 20–30% |
Most professional valuators run two or three methods and triangulate. A 30%+ gap between methods isn’t an error — it’s a signal that one method’s assumptions don’t fit the company.
Asset-Based Methods
Sum tangible and intangible assets, subtract liabilities. Book value uses recorded cost; adjusted NAV restates lines to fair market value with intangibles added in. For software companies, both almost always understate value because operating IP and team aren’t on the balance sheet. Useful only as a liquidation floor or as a sanity check for hardware-heavy businesses. Skip for pure-software startups.
Market-Based Methods
Comparable company analysis
Apply valuation multiples from similar public companies to the startup’s metrics. The hard part is “comparable” — same industry, similar size, similar growth profile, similar business model. Apply a 20–40% illiquidity discount because private shares can’t be sold freely.
The common multiples are EV/Revenue (3–15x for pre-profit SaaS and tech), EV/EBITDA (8–20x for profitable businesses), EV/Gross Profit (2–8x for high-growth), and Price/Earnings (15–40x for mature profitable companies).
Worked example. Series B SaaS with $5M ARR growing 120% YoY. Public comparables in 2024 (Snowflake, Datadog, Cloudflare, MongoDB) traded at 12–18x forward revenue. Median ~15x. Apply 30% illiquidity discount: $5M × 15 × 0.7 = $52.5M. That’s a starting point — Datadog’s growth profile is a poor match for a Series B startup, so the multiple needs further adjustment.
Precedent transactions
Same idea using actual M&A multiples instead of trading multiples. Real transactions reflect strategic premiums (20–40%) and competitive auction premiums (15–30%). Sources: PitchBook, Crunchbase, CB Insights, SEC filings. Two-year-old comparables may need a 10–20% downward revision in a softer market.
Income-Based Methods
Discounted cash flow
DCF projects free cash flow over 5–10 years, discounts to present value, and adds a terminal value. Three inputs drive the answer:
- FCF forecast. Year 1–3 should tie to the operating model; later years are speculative.
- Discount rate. 30–50% for startups reflects failure risk and illiquidity. Public companies use 8–12%.
- Terminal value. Often 60–80% of total DCF. Calculated as perpetuity growth (FCF × (1+g) / (WACC − g)) or a forward EBITDA multiple.
A 5-point change in discount rate moves DCF by 30%; a 2x change in exit multiple moves terminal value by 50%. DCF only makes sense once revenue history constrains year-1 projections.
VC method
The VC method works backward from a target exit. It’s how most early-stage investors actually price rounds.
Step 1: Exit value Year-N revenue × exit multiple
Step 2: Target return 10x seed, 5x Series A, 3x Series B
Step 3: Required ownership (Investment × Multiple) / Exit Value
Step 4: Adjust for dilution Required / (1 − Future Dilution)
Step 5: Post-money valuation Investment / Required OwnershipSeries A example: $25M ARR in 5 years × 8x multiple = $200M exit. 5x target. After 30% future dilution: required ownership = 20% / 0.70 = 28.6%. Adjusted post-money ≈ $140M. The VC method makes the bet visible: at $140M post on $8M, the investor needs a $200M exit — a more honest conversation than a DCF result.
Risk-Based Methods for Pre-Revenue
Scorecard valuation
Standard angel-investor framework. Take the regional average pre-money for comparable-stage deals, evaluate the target across weighted factors, multiply. The standard weights: management team 30%, size of opportunity 25%, product/technology 15%, competitive environment 10%, marketing/sales 10%, capital efficiency 5%, other 5%.
Each factor scores 50–150% versus the regional average. A team scoring 130% on a 30% weight contributes +9 percentage points. A startup scoring 121% overall against a $4M regional average gets a $4.84M valuation. The 30% management weight reflects what most early-stage angels report: founders who can hire and execute beat better-positioned products with weaker teams.
Risk factor summation
Twelve standard risk factors (management, market, technology, regulation, competition, etc.) each scored from -2 to +2, with each step worth ±$250K against a stage-appropriate baseline. Adjustments typically span ±$3M total.
Both qualitative methods share one vulnerability: optimistic founders self-score generously, then can’t grow into the valuation. Have an advisor score independently — gaps surface fast.
Industry Patterns
SaaS valued on ARR, growth, and net revenue retention. Rule of 40 (growth + margin) drives premium multiples — 100%+ growth justifies 15–20x revenue, sub-30% caps at 4–8x. Biotech uses probability-weighted scenarios tied to clinical trials, with binary jumps at FDA decisions. Marketplaces trade on GMV and take rate; two-sided dynamics make scaling capital-intensive. Hardware carries 30–50% gross margins versus 70–90% for software, compressing multiples by half or more.
Limitations
No method survives reality unmodified. Sensitivity is brutal: 5% change in discount rate moves DCF ±30%; 10% change in revenue growth moves it ±40%; 2x change in terminal multiple moves terminal value ±50%. Comparables assume you can find truly comparable businesses, rarely true for differentiated startups. Public multiples compressed 40–60% during the 2022 SaaS correction.
Practical Workflow
- Pick two or three methods appropriate for the stage.
- Build each independently — don’t reverse-engineer one to match another.
- Triangulate. If results spread by more than 30%, investigate which assumption set is doing the work.
- Express the answer as a range, not a point.
- Update after every material event: financing, product launch, customer concentration shift, 409A valuation refresh.
For founders preparing a round, the deliverable isn’t a single number — it’s a defensible range with assumptions in plain sight. Once the round closes, the post-money valuation feeds the cap table, interacts with any SAFE conversions on the books, and ultimately drives the waterfall analysis at exit.
FAQ
What’s the most accurate method? None individually. Triangulate. Pre-revenue: scorecard + risk factor summation. Series A+: VC method + comparables + light DCF.
How do you value a startup with no revenue? Scorecard, risk factor summation, or recent comparable rounds. Heavy weighting on team and market size. Typical range $1–5M; more for repeat founders or large markets.
What discount rate should I use? 50–60% pre-revenue, 40–50% seed, 30–40% Series A, 25–35% Series B, 20–30% later.
How often should valuations be updated? Annually for 409A compliance, plus after any material event — funding round, major customer change, convertible preferred stock modification.
Why do different methods produce different numbers? Each weights different inputs. DCF prices future cash flow; comparables price market sentiment; scorecard prices team and market. A 30%+ gap means one method’s assumptions don’t fit.