Startup Valuation Methods: 8 Approaches Compared (2026 Guide)

Eight methods cover almost every situation a founder will face in 2026. Pre-revenue: Berkus and Scorecard for angel rounds, Risk Factor Summation when the deal has uneven risk. Seed: VC method paired with Comparables. Series B+: DCF as primary, Comparables as sanity-check. Cost-to-Duplicate and First Chicago round out the toolkit. Pick by stage, never by familiarity.

Valuation is where most founders lose leverage. After advising 600+ startups and supporting $3B+ raised — including $630M closed in 2025 — the pattern we see is the same one every quarter: a founder defaults to the method they remember from business school (usually DCF) for a pre-seed round where it has no business being. The right answer is almost always to triangulate two or three stage-appropriate methods, then defend the range with comp data.

Below are the eight methods that show up in real term sheets, ranked by where they fit. We've stripped out the mature-business methods (asset-based, capitalization of earnings, replacement cost, market cap) that the original 2023 version of this post leaned on — they're rarely what an angel or VC actually runs, and founders searching "startup valuation methods" expect Berkus first.

8 startup valuation methods compared at a glance

Pre-revenue founders should run Berkus and Scorecard side-by-side and average the outputs. Seed-stage founders add Risk Factor Summation when the deal has lopsided risk and use VC method against a real exit thesis. Series A onward, Comparables becomes the primary anchor with VC method or DCF as the cross-check. Stage fit beats sophistication every time.

Pre-revenue startup valuation methods

Use qualitative methods that don't require financials. Berkus assigns up to $500K per success factor across five categories (idea, prototype, team, partners, early sales) for a maximum pre-money of about $2.5M. Scorecard adjusts your regional pre-money median by a weighted 7-factor scorecard. Risk Factor Summation adds or subtracts $250K–$500K per risk dimension. Run two and average.

1. Berkus method

The Berkus method, created by angel investor Dave Berkus in the 1990s, assigns up to $500K of pre-money value to each of five qualitative success factors — sound idea, prototype, quality management team, strategic relationships, and early sales rollout. Maximum pre-money tops out near $2.5M, by design. It's the fastest, most defensible method for an angel deal where there's no revenue to anchor on.

The five Berkus factors and their max contribution to pre-money:

• Sound idea (basic value, product risk): up to $500K
• Prototype (technology risk): up to $500K
• Quality management team (execution risk): up to $500K
• Strategic relationships (market risk): up to $500K
• Product rollout / first sales (production risk): up to $500K

Worked example: A pre-revenue B2B SaaS founder with a working MVP, two warm strategic-partner LOIs, a strong founding team, and no paying customers yet might score $400K (idea) + $400K (prototype) + $450K (team) + $300K (partners) + $0 (sales) = $1.55M pre-money. The Berkus ceiling makes it impossible to talk yourself into a $10M valuation on hand-waving alone — which is the entire point.

Berkus pros and cons

• ✅ Fast — most founders can score themselves in 30 minutes
• ✅ Hard cap forces realism at the angel stage
• ✅ No financial model required
• ❗ Subjective scoring — two angels can land 30% apart on the same deal
• ❗ Capped at ~$2.5M, so it's useless for valuations above that ceiling

2. Scorecard valuation method (Bill Payne)

The Scorecard method, formalized by angel investor Bill Payne in 2011, multiplies your region's median pre-money valuation for similar-stage startups by a weighted 7-factor scorecard. You score your startup against a regional baseline (1.0 = median), then apply weights to each factor. The output is anchored to real angel comps — which is why it's the most-cited method in active angel groups.

Bill Payne's original 2011 weights (practitioners commonly adjust by ±5%):

Worked example: A US-Northeast SaaS pre-seed startup uses a regional pre-money median of $5M (figure investors should source from their local angel group, Pitchbook, or Carta's State of Private Markets — never invent it). The founder scores: management 1.25× × 30% = 0.375, opportunity 1.5× × 25% = 0.375, product 1.0× × 15% = 0.15, competition 0.75× × 10% = 0.075, sales 1.0× × 10% = 0.10, financing 1.0× × 5% = 0.05, other 1.0× × 5% = 0.05. Sum = 1.175. Pre-money = $5M × 1.175 = $5.875M.

3. Risk Factor Summation method

The Risk Factor Summation method (originated at Ohio TechAngels) starts with a regional pre-money median, then adds or subtracts $250K for each of 12 risk dimensions you score from very low risk (+$500K) to very high risk (−$500K). It's the most granular pre-revenue method and the right tool when one or two specific risks dominate the deal.

The 12 risk dimensions: management, stage of business, legislation/political risk, manufacturing risk, sales and marketing risk, funding/capital-raise risk, competition risk, technology risk, litigation risk, international risk, reputation risk, potential lucrative exit risk.

Scoring band: very low (+$500K), low (+$250K), neutral ($0), high (−$250K), very high (−$500K) per dimension.

Worked example: A medtech founder starts with a $3M regional pre-money baseline. They score management +$500K (proven team), stage +$250K (MVP shipping), legislation −$500K (FDA Class II pathway), tech +$250K (patent filed), competition $0, funding −$250K (long capital path), exit +$250K (clear strategic acquirer pool). The other 5 dimensions net to $0. Net adjustment = +$500K. Pre-money = $3.5M. Where Risk Factor earns its keep is forcing the founder to face the FDA risk on the page rather than glossing it in the deck.

4. Cost-to-Duplicate method

Cost-to-Duplicate sums what it would cost a competent third party to recreate your business today — engineering time, IP filings, prototype build, team-and-recruiting cost. It's a floor, not a fair-value number. Use it as a defensive anchor when an investor lowballs you on a SAFE, or in any asset deal where intangibles aren't the main driver of value.

Worked example: A pre-revenue computer-vision startup tallies 18 months of two-engineer time at $180K/yr fully loaded ($540K), one provisional patent ($25K), GPU compute spend ($60K), and design/UX contractor cost ($40K) = $665K floor. That's the number you anchor to in a defensive negotiation — not what you ask for, but the floor below which the deal is structurally bad. Don't use Cost-to-Duplicate as your primary number unless you're selling a workshop or asset business; for venture, it ignores the most valuable thing you've built (network effects, brand, IP optionality).

Investor-driven valuation methods (seed and beyond)

From seed onward, the VC method dominates because it mirrors how investors price a deal: target an exit value, work backward via target IRR, divide by the post-money. Comparable Company Analysis is the second-most common — anchor your multiple to a clean comp set. First Chicago shows up in deeptech and biotech where outcomes are step-change. Discounted Cash Flow stays in the toolbox but only becomes primary at Series B+.

5. Venture Capital method

The VC method, formalized by Harvard Business School professor Bill Sahlman in 1987, runs the deal backward. Start with a projected exit value (revenue × industry multiple in year-of-exit). Discount it to today using a target IRR (commonly 30–60% per year for early-stage). Divide by post-money you need today, accounting for future dilution. The result is the post-money valuation an investor can underwrite to.

The two-step formula:

• Step 1 — Terminal (exit) value: Year-of-exit revenue × exit multiple (e.g., 5–8× revenue for SaaS, source from Bessemer State of the Cloud or Damodaran NYU multiples data). Net of net debt.
• Step 2 — Post-money today: Terminal value ÷ (1 + target IRR)^years-to-exit ÷ retention ratio (1 − future dilution from later rounds)

Worked example (anonymized Waveup deeptech case): A Series A deeptech client we supported targeted year-7 revenue of $80M, exit multiple of 5× (anchored to recent strategic acquirers in the sector — never invented), giving terminal value of $400M. Target IRR for the lead investor was 35%/year. Future dilution to exit was modeled at 50%. Post-money today = $400M ÷ (1.35)^7 ÷ 0.5 = $400M ÷ 8.17 ÷ 0.5 ≈ $98M post-money. Pre-money = $98M − new $15M check = $83M pre-money. Note this is one method among three we ran for that client; alone it's vulnerable to exit-multiple drift, which is why we triangulated against Comparables.

6. Comparable Company Analysis (CCV)

CCV (also called "comps") finds 5–10 similar companies, pulls their EV-to-revenue or EV-to-EBITDA multiples from public filings or transaction data, and applies a median or trimmed-mean multiple to your own metric. The output is market-grounded, which is why bankers and Series A+ investors default to it. The hard part is comp selection — get the comp set wrong and the whole exercise is a marketing deck.

Five-step CCV workflow:

1. Identify comparable companies — same industry, similar business model, similar growth and margin profile
2. Gather financial data — revenue, ARR, EBITDA, growth rate; pull EV from public filings or transaction databases
3. Calculate valuation multiples — EV / revenue, EV / EBITDA, EV / ARR for SaaS
4. Determine the relevant multiple — median or trimmed mean of the comp set; adjust for growth/margin gaps
5. Apply the multiple to your own metric — output is a valuation range, not a point estimate

Worked example (anonymized Waveup fintech case): A profitable B2B fintech we supported through Series B used a 10-comp public-and-private set with median EV / forward-revenue of 6.5×. Forward revenue: $22M. Mid-point pre-money: 6.5 × $22M = $143M. The actual round closed in that range after a single round of negotiation — exactly what a clean comp set buys you when an investor knows the market as well as you do.

7. Discounted Cash Flow (DCF) method

DCF is the right primary method at Series B and later, when you have 3+ years of revenue history and a defensible margin path. Pre-Series-B, treat it as a triangulation tool only — never as the primary anchor — because the assumption sensitivity (discount rate, terminal growth, terminal multiple) makes any pre-revenue DCF a multiplier of guesses. Aswath Damodaran (NYU) covers this nuance better than anyone.

DCF projects 5–10 years of free cash flows, discounts each year to today using a weighted average cost of capital (WACC), adds a terminal value (commonly Gordon-growth or exit-multiple), and sums. The output is enterprise value; deduct net debt for equity value.

Why DCF fails pre-revenue: the discount rate alone moves valuation 50%+ on small input changes. Add unknown terminal growth and an unproven margin path, and you're stacking guesses. Damodaran's valuation data consistently shows wider valuation bands at early stage — which is the empirical reason every angel runs Berkus or Scorecard instead.

Use DCF as a sanity-check at Series A — if your VC-method output is 4× your DCF output, one of your two assumption sets is broken — and as a primary tool at Series B+ when you have history to anchor on.

8. First Chicago method

First Chicago, named after the bank that codified it, runs three discounted-cash-flow scenarios (success, sideways, failure) and weights each by probability. Output is a probability-weighted valuation that forces honest discussion of failure cases. Most useful for deeptech, biotech, and climate startups where outcomes are step-change rather than continuous.

Worked example: A climate-tech founder builds three scenarios. Success ($200M exit at 7 years, 20% probability) = $40M weighted contribution. Sideways ($40M exit at 7 years, 50% probability) = $20M. Failure ($0 outcome, 30% probability) = $0. Probability-weighted enterprise value = $60M. Apply the same target-IRR discount and dilution math from the VC method to get to today's post-money. The probabilities are still guesses — but they're explicit guesses you can defend, instead of one optimistic forecast pretending to be base case.

How to triangulate — the Waveup approach

Run two stage-appropriate methods, compare the outputs, then use a third as a sanity check. If the two outputs are within 25%, you have a defensible range. If they're more than 50% apart, one of your assumption sets is broken — usually exit multiple or comp selection. Anchor the final number on the method most directly tied to recent comp deals in your sector.

Founders we work with on real raises run this same loop:

1. Pick two stage-appropriate methods — pre-revenue: Berkus + Scorecard. Seed: VC method + Risk Factor. Series A: VC method + Comparables. Series B+: DCF + Comparables.
2. Compute both outputs with conservative inputs — never the optimistic forecast
3. Compare ranges — within 25% is a defensible band; 25–50% means revisit assumptions; >50% means one method is broken
4. Anchor on the comp-grounded method when there's tension (Comparables at Series A+, Scorecard at pre-seed)
5. Document the assumption set — every input source, every multiple, every probability — so it survives investor questions

Methods that don't fit startups

Skip Capitalization of Earnings, pure Asset-based valuation, Replacement Cost, and Market Cap when valuing a startup. The first three assume stable historical earnings or a tangible asset base — neither holds for a high-growth startup. Market Cap only applies to public companies. They're useful frameworks for a small business sale or M&A audit, not for a venture round.

Capitalization of Earnings divides recent earnings by a capitalization rate to get value — useful for stable, mature small businesses (think a 10-year-old restaurant) but meaningless for a startup whose value is in growth, not current earnings.

Asset-based valuation sums tangible plus intangible assets minus liabilities. It floors the value of asset-heavy businesses (manufacturing, real estate) but undervalues any startup whose moat is software, brand, or network effects.

Replacement Cost is the cost to rebuild assets at today's prices. We've folded the startup-relevant version of this into Cost-to-Duplicate above — keep that, skip the rest.

Market Cap is share price × shares outstanding. Public companies only. If you're private, useful as a comparison anchor when running CCV; otherwise, skip.

Get help with your startup valuation

Triangulate two stage-appropriate methods, defend the range with comp data, and have a second pair of eyes pressure-test the assumptions before you take the deck to investors. The fastest founders we work with run the math first, then call us to red-team it — not the other way around. The number you walk in with sets the ceiling for the round.

Valuation isn't a one-time spreadsheet — it's the foundation of every conversation with every investor for the next 18 months. Get it wrong and you spend the round defending a number that doesn't survive diligence. Get it right and you walk into term-sheet conversations with leverage.

Waveup has supported $3B+ raised across 600+ startups, including $630M closed in 2025 alone — and 70% faster close is the median outcome when founders pair our financial modeling and fundraising work with their own raise. If your raise hinges on getting the valuation right, get a second opinion before you send the deck.

FAQs about startup valuation methods

These are the questions we hear most often during real fundraises across 600+ startups: which method to use at which stage, when DCF stops being a toy, how to triangulate, where the data comes from, and whether the same method works for pre- and post-money. Each answer below reflects what we recommend in actual term-sheet conversations.