What is a DCF model?

A DCF model values a company by projecting the cash it will generate in the future and discounting that cash back to what it's worth today. Forecast the future free cash flows, shrink each to present value, add them up, and the total is the company's estimated intrinsic value — what the business is worth on its own cash generation, regardless of what the market is currently paying.

The core formula

Value = CF₁/(1+r)¹ + CF₂/(1+r)² + … + CFₙ/(1+r)ⁿ + Terminal Value/(1+r)ⁿ

CF is the cash flow in each period, r is the discount rate, and n is the number of forecast years. The further out the cash, the more it's reduced — a $100 cash flow discounted at 10% is worth ~$91 next year, ~$83 the year after, and so on.

What cash flow gets discounted

DCF models discount free cash flow — cash left after operating costs and the reinvestment needed to keep running. Which flavor you use changes the whole build:

• Unlevered free cash flow (FCFF) — cash to all capital providers, before financing. Discounted at WACC, it yields enterprise value. This is the standard approach.
• Levered free cash flow (FCFE) — cash to equity holders after debt service. Discounted at the cost of equity, it yields equity value directly.

The unlevered approach dominates: most of the time "a DCF" means FCFF discounted at WACC to enterprise value, then bridged to equity. Match the cash flow to the rate — discounting FCFF at the cost of equity (or FCFE at WACC) is a common, material error.

Terminal value: the part that dominates

You can't forecast forever, so a DCF projects cash flows explicitly for 5–10 years and captures everything after in a single terminal value, computed two standard ways:

Perpetuity growth: TV = (Final-year FCF × (1 + g)) ÷ (r − g) · Exit multiple: TV = Final-year EBITDA × exit multiple

g is the perpetual growth rate — typically 2–3%, at or below long-run GDP, because no mature company outgrows the economy forever. Terminal value often makes up the majority of total DCF value — frequently ~75% — so the answer is acutely sensitive to two assumptions (growth/exit multiple and the discount rate) about cash flows you can't see in detail. Always compute it both ways and cross-check: if the Gordon-Growth TV implies a 40x EBITDA exit, the growth rate is too high.

The workflow: from data to value

A DCF isn't just the arithmetic — the defensible version runs an end-to-end process, and most of the judgment happens before any cash flow is discounted: collecting and normalizing the data, deciding which DCF you're building, researching where the market is heading, and building the discount rate. Only then do you build the model itself.

Here's the explicit forecast for Costco (COST) — revenue fading from 8% growth to a 2.5% terminal rate over ten years, each year's FCFF built from EBIT, taxes, D&A, capex, and the change in working capital.

DCF vs. the market-based methods

DCF is one of three main approaches, alongside comparable company analysis (trading multiples of similar public companies) and precedent transactions (prices paid in similar deals). The philosophies differ: comps and precedents are market-based — what others are paying — while DCF is intrinsic — what the cash flows are worth regardless of sentiment. Experienced analysts rarely rely on a DCF alone; they run it beside comps as a cross-check, since each catches what the others miss. (Multiples-based work leans on a normalized earnings base — see what an EBITDA bridge is.) When the DCF and the multiples disagree sharply, that gap is the analysis — not an error to smooth over.

A note on the dividend discount model

The dividend discount model (DDM) is a specialized DCF: instead of free cash flow it discounts expected dividends at the cost of equity to value equity directly. It suits stable dividend-payers (mature financials, utilities) where dividends proxy distributable cash; the single-stage version is just Gordon Growth applied to next year's dividend. For companies without steady dividends, a free-cash-flow DCF is the better tool — and valuing a bond follows the same present-value logic on its contractual coupons and principal.

Edge cases and common errors

• Cross-check the terminal value both ways. Compute it by perpetuity growth and back out the implied exit multiple (and vice versa). This single check catches most overvaluations.
• Keep g disciplined. The perpetual growth rate must be below WACC (or the formula breaks) and at or below long-run nominal GDP. Typically 2–3%.
• Use the mid-year convention where cash flows arrive through the year, or you systematically understate value by discounting a full period.
• Normalize a cyclical or negative base year. Don't build a perpetuity off a trough or a one-off spike; forecast to a mid-cycle steady state first.
• Growth has to be funded. Revenue growth without the capex and working capital to support it produces a structurally overstated value.
• Match the cash flow to the rate. FCFF ↔ WACC; FCFE ↔ cost of equity. Mixing them is material.
• Bridge enterprise value to equity properly — add non-operating assets (excess cash, investments, NOLs), subtract debt and other claims (preferred, minority interest, unfunded pension).
• Not for every company. Pre-revenue or highly unpredictable businesses are better valued with market-based methods; a DCF there is false precision.

Limitations

The strength of a DCF — that it's built from fundamentals — is also its weakness: it's only as good as the assumptions you feed it, and small changes swing the answer hard. Accuracy is driven far more by the realism of the cash-flow forecast than by discount-rate precision; the terminal-value concentration means most of the value rests on a future you can't model in detail; and because the inputs are judgment, a DCF is easy to nudge — intentionally or not — toward a number you already wanted. Used well it's a disciplined way to think about worth; used carelessly it's a spreadsheet that confirms your priors.

Where each input comes from

MacrosLM's DCF Builder constructs the model from a company's financials, projects the cash flows, builds WACC and terminal value two ways, and bridges to equity value with each figure traced to source. What none of it does is supply the assumptions — the growth trajectory, the margin path, the discount rate, the terminal method. Those stay with the analyst who owns the result.

Bottom line

A DCF values a company on its own cash generation. Unlevered cash flows at WACC give enterprise value; terminal value captures the rest and usually dominates. The arithmetic is straightforward; the valuation lives or dies on the assumptions — which is exactly where the analyst's judgment belongs, and why a DCF is always read next to the market-based methods.

This article is for general information and is not investment or financial advice. A DCF depends on assumptions that vary by company and analyst and should be prepared or reviewed by a qualified professional.