DCF Valuation Model

What a DCF actually does

A company is worth the cash it will hand you, discounted for the fact that cash arriving in 2036 is worth less than cash arriving today. That is the entire idea. Everything else is bookkeeping around it.

This model does the bookkeeping. You forecast ten years of revenue, margins, taxes, capital spending and working capital. It converts that into free cash flow, discounts each year at a rate you build explicitly, adds a terminal value for everything after year ten, subtracts net debt, and divides by diluted shares.

What it will not do is tell you the truth. A DCF is not a truth machine, and anyone quoting a fair value to the cent is misusing one. It is a device for making your assumptions explicit, so that you and other people can argue with them properly instead of trading adjectives.

What is in the workbook

Five tabs. Two you fill in, three that do the work.

• Start Here. Version, instructions, a legend for which cells to touch, and the disclaimer.
• Assumptions. Company financials, forecast drivers, the WACC build, and terminal assumptions. Ten years of revenue growth and margin are entered year by year, so you control the fade rather than inheriting one.
• Model. The ten-year projection: revenue, EBIT, taxes, NOPAT, D&A, capex, change in working capital, free cash flow, discount factor, present value. Entirely formulas.
• Valuation. Terminal value by both methods, enterprise value, equity value, fair value per share, upside against the current price, and a sensitivity grid across discount rate and terminal growth.
• Reverse DCF. The most useful tab, and the reason this model exists.

Yellow cells with blue text are yours to edit. Everything else is a live formula. The sensitivity grid and the reverse DCF table are built from formulas too, not macros or Excel data tables, so they work identically in Sheets, Numbers and LibreOffice.

The reverse DCF, and why it is the point

A forward DCF tells you what you think a company is worth. That is mildly interesting. A reverse DCF tells you what the market thinks, which is considerably more useful, because it converts a vague disagreement into a specific one.

The tab holds every assumption fixed except revenue growth, then solves the other way: what constant growth rate would justify today's share price? You read down the table until the difference column crosses zero. That rate is the market's implied expectation.

Then you have exactly one question to answer, and it is a good one: is that growth rate more or less likely than the one I put in myself? That is a question you can research. "Is this stock cheap" is not.

This is the same technique behind the implied-expectations tables in our research reports. In the Palantir initiation it showed that the share price required roughly 79% revenue growth in 2027 at a 30 times exit multiple. In the Texas Pacific Land report it showed the price implied about $73 oil. Neither figure is a forecast. Both are statements about what you are being asked to believe.

Read the terminal value percentage first

The Valuation tab reports terminal value as a share of enterprise value, and it sits directly above the fair value figure on purpose.

In most DCFs that share lands between 60% and 80%. This means the majority of your answer comes from a single perpetuity formula applied to year ten, not from the ten years you carefully forecast. If the share climbs above roughly 80%, the model is a terminal value with decoration attached, and small changes to the discount rate or terminal growth will swing the output enormously.

That is not a reason to avoid a DCF. It is a reason to treat the output as a range rather than a number, and to spend your effort on the discount rate rather than on year six revenue. The sensitivity grid exists so you can see the range instead of imagining it.

Where DCFs go wrong

Three failure modes account for most bad DCFs, and all three are user error rather than model error.

A flat growth rate for a decade. Almost nothing grows at a constant rate for ten years. Entering 20% in every year is the single most common way a DCF quietly becomes fiction. This model asks for a rate per year specifically so that fading it down is the path of least resistance.

Terminal growth that eats the economy. A terminal rate above roughly 3% assumes the company permanently outgrows the economy it operates in, which cannot be true indefinitely. The spreadsheet will let you type it; the sensitivity grid will show you how unstable it makes everything.

Reverse-engineering the answer. The most dangerous failure, because it feels like analysis. You decide a company is attractive, then adjust growth and discount rate until the model agrees. The defense is to write your assumptions down before you look at the output, and to check them against the reverse DCF afterward.

What this tool cannot tell you

A DCF values a going concern with reasonably predictable cash flows. It is poorly suited to pre-profit companies, to businesses whose value sits in optionality rather than operations, to banks and insurers where the capital structure is the business, and to anything whose next five years are genuinely unknowable. For those, a DCF produces a number with false precision, which is worse than no number.

It also cannot tell you whether you should own an individual company at all. Our own position on that is that most people should own an index and spend their time elsewhere. This tool is for the minority who have decided otherwise and want to do it with the arithmetic visible.

Using it

Start with the company's actual reported figures for the most recent fiscal year, taken from a filing rather than a screener. Enter a price and the date you took it, because a valuation without a price date is not reproducible. Make your forecast assumptions, read the sensitivity grid rather than the point estimate, then go to the reverse DCF and find out what you are actually disagreeing with the market about.

Found an error, or think an assumption here is wrong? Tell me. Corrections get published with a date attached, the same as they do for the research reports.