Startup Valuation Methods: Deal Heuristics vs. Fair Value Reporting

"How do you value a pre-revenue startup" is really two separate questions with two separate audiences.

• Pricing a deal. An angel investor or early-stage VC needs a negotiating anchor before any money changes hands. There is no revenue, sometimes no product, so the tools are qualitative heuristics dressed up with dollar figures.
• Reporting fair value. A fund that has already invested has to report what that stake is worth at each subsequent reporting date, for its own financial statements, under fair value accounting. This is a formal, standard-governed exercise, not a negotiation.

Most online explainers list both sets of methods together as though they were answering the same question. They are not.

• Berkus Method. Assigns a dollar value, up to roughly half a million each, to five qualitative factors: sound idea, working prototype, quality of the management team, strategic relationships, and early sales rollout. Pre-money value tops out by design, since it was built for early angel rounds.
• Scorecard Method. Starts from the average pre-money valuation of recently funded startups in the same region and sector, then adjusts it up or down using a weighted scorecard across factors like team, market size, and competition.
• Risk Factor Summation. Same starting point, a regional and sectoral average, adjusted against roughly twelve named risk categories instead of a scorecard.

These are genuinely useful for what they are built for: giving two people in a negotiation a number to start from when there is no financial history to anchor to. None of them is a recognised valuation standard, and none is what a registered valuer or a fund's finance team would use to certify a number for a statutory filing or a set of audited accounts. The reasons why are specific, not just a matter of taste, and are covered directly in Are Berkus and Scorecard Acceptable to a Professional Valuer?.

Funds valuing portfolio companies for their own financial statements work from the International Private Equity and Venture Capital Valuation (IPEV) Guidelines, the recognised standard for exactly this question, sitting alongside the International Valuation Standards covered in What Valuation Standard Actually Governs an SFA Registered Valuer? IVS governs the conduct of valuation work generally. IPEV addresses specifically how private capital practitioners apply the fair value objective to their portfolios. Pulled directly from the current IPEV text, the named methods are:

• Calibration (Guideline 2.6). The dominant technique in practice. If the most recent funding round was an arm's-length transaction, its price is treated as fair value at that date. The valuer works backward to find what multiple or discount rate that price implies against comparable companies, then rolls that calibrated figure forward at each later reporting date based on how the comparables have moved, rather than rebuilding a valuation from first principles every time.
• Scenario-based methods, including the full probability-weighted expected return method (PWERM). Multiple future outcomes are modelled, each assigned a value and a probability, and the results are weighted together.
• The option pricing method (OPM). Treats each class of equity as a call option on total enterprise value, useful when outcomes form a continuous range rather than a handful of discrete scenarios. This is the allocation problem a multi-class cap table creates, the same kind of complexity that makes pricing a CCPS conversion more involved than pricing plain equity.
• The hybrid method, literally a blend of scenario-based methods and OPM.
• The current value method (CVM), used when an exit is imminent. Allocates value across the equity classes as though the company were being sold on the measurement date itself.

IPEV's own illustration makes the mechanic concrete. A fund invests at a price implying 8x EBITDA. Comparable companies are trading at 10x at the time. That gap, a 20% discount to the comparables, is not an error, it is calibrated in: it captures everything about this specific company, illiquidity, control, growth rate, that justifies trading below the peer set.

At the next reporting date, the valuer does not start over. If the comparable set has moved from 10x to 15x, judgement is applied to how much of that move the portfolio company should track, informed by the calibrated 20% discount, not by rebuilding a fresh multiple from scratch.

IPEV is explicit that no single method is superior in all circumstances, but it does give a practical steer: early in a company's life, when the eventual exit route is still unclear, more weight tends to go to the hybrid or OPM approach, since these account for the different rights and preferences each investor class holds. As a company nears an actual exit through an IPO or acquisition, where everyone is likely to receive close to the same price per share, more weight shifts to a fully-diluted, common-stock-equivalent view.

None of this displaces the Income Approach used for statutory work under Section 62(1)(c) and described on the Services page. DCF still answers the question these methods do not: what is the underlying business actually worth, based on its own projected cash flows. What Calibration, PWERM, OPM, the hybrid method, and CVM answer is a different question layered on top: given that enterprise value, or given a real transaction price nobody wants to throw away, how should it be allocated across a cap table with multiple classes of rights, or rolled forward without a full valuation exercise every quarter. Both questions come up constantly in the same startup. They are just not the same question. The same allocation logic is what a valuer actually reaches for when a cap table's multiple classes differ in voting power rather than liquidation preference, covered in DVR Shares Are a Valuer's Job. The Standards Go Quiet on the Number..