Valuation is a disciplined art, not an exact science — the multiple is a conclusion, not a start.

Part Five turns from “what kind of company is this?” to the question investing actually rides on: “at this price, is it worth buying?” A great company at too high a price is a bad investment; a mediocre one cheap enough can be an opportunity. That leap — from what it is to what it’s worth — is where research becomes investment, and step one is admitting no model computes a “precise intrinsic value.” The goal is a framework, not false precision.

SaaS valuation runs on three rulers, each with its trap. EV/Revenue is the most common and the most dangerous, because it ignores the quality of that revenue — margin, growth durability, cash. It’s how people slap a software multiple on low-margin payment flow, or on revenue piled up by M&A. EV/FCF and P/E are stricter — but remember Chapter 7: FCF is flattered by SBC, so use it net of SBC as a stress test, or you’ll overpay for a cash flow propped up by dilution. And reverse DCF — the one this book favors most — flips the exercise: instead of predicting the future to compute a value, it asks what growth and margins the current price already implies, and whether that’s reasonable. That turns “I predict the future” (almost always wrong) into “I judge whether the market’s expectation is sane” (far more useful).

Notice what all three share: the multiple a company deserves is a pricing of everything in Parts One through Four — growth quality, margin (now an AI variable), retention, cash quality, moat depth, AI fate. That’s why valuation comes last. The multiple is the conclusion of the analysis, never its starting point — and “whose multiple is lower?” divorced from those fundamentals is the most naive way to invest.

One consequence is that a low multiple and a high multiple carry no information on their own. A company at a rich multiple can be cheap if it grows into and past those expectations; a company at a depressed multiple can be expensive if its fundamentals are quietly eroding. “Cheap” and “expensive” are statements about the relationship between price and future fundamentals — never about the multiple in isolation, which is exactly why the number is where the analysis ends, not where it starts.

The core tool is three-scenario valuation — bear, base, bull — the best weapon against false precision. You give up the single fake number and sketch three futures, estimating a value for each, then compare to the price. Read the result honestly: it’s a modeled range of outcomes and their risk/reward, not a true probability distribution unless you assign probabilities — and “limited downside” is a modeled claim, subject to model error, omitted scenarios, and balance-sheet risk. (This chapter deliberately gives no specific market multiples: those are market data that require a dated source, and fabricating them would break the book’s discipline.) Done right, three scenarios tells you not “what it’s worth” but the more useful thing — what risk/reward the current price offers.

This is Chapter 20 of “Compounding in the Cloud.”