How to Use a Discounted Free Cash Flow Model

Understanding DCF theory is one thing. Building one and using it to make an actual investment decision is another. This article walks you through the practical steps of building a working DCF model on a real business — the decision points, the information you need, and the mistakes to avoid.

Step 1: Gather Historical Data

Before you forecast the future, you must understand the past. Pull the last 5-10 years of financial statements. You need: revenue (annual growth rates and trends), free cash flow (historical generation and how it has trended relative to earnings — a large divergence signals a data quality question), capital expenditures (what has the company been spending on assets?), working capital changes (how much cash is tied up in inventory and receivables?), and the debt and cash position (leverage, interest rates, liquidity). If the company's capital intensity or margins have been stable, historical patterns are good guides to the future. If there is significant volatility, you need to understand why.

Step 2: Forecast Free Cash Flow

This is the core of DCF. Estimate what free cash the business will generate in years 1-5. Start with revenue: base your growth forecast on historical rates (while recognising that growth typically slows), the competitive landscape (if the market is saturating, growth should slow), market size and market share dynamics, and management guidance (but be sceptical — managers have incentives to overstate).

Warren Buffett prefers to underestimate growth rather than overestimate. Then forecast operating margins (base on historical margins, pricing power, and scale dynamics) and capex (historical capex as a percentage of revenue is your baseline). The formula:

Revenue × Operating Margin - Capex - Working Capital Changes = Free Cash Flow

Step 3: Choose Your Discount Rate

The discount rate is the return investors require to invest in this business. For publicly traded stocks, start with the 10-year government bond yield (roughly 4-5% in 2026), add 5-7% equity risk premium for the risk of being in the stock market versus bonds, and add 0-5% company-specific risk premium depending on business stability.

A stable utility might deserve 6-8%. A growth software company might need 10-12%. A startup might need 15%+. Damodaran publishes detailed guidance on discount rates by industry. A 1-2% swing in discount rate will change your valuation by 20-30%, which is why sensitivity analysis matters.

Step 4: Calculate Terminal Value

Terminal value is typically 60-80% of the DCF output. It is where the valuation lives or dies. Using the perpetuity growth method (most common), assume the company grows at a stable rate forever — often 2-3%, roughly GDP growth, because no company can grow faster than GDP indefinitely:

Terminal Value = Year 5 FCF × (1 + Terminal Growth Rate) ÷ (Discount Rate - Terminal Growth Rate)

Example: If Year 5 FCF is £100M, terminal growth is 2.5%, and discount rate is 8%: Terminal Value = £100M × 1.025 ÷ (0.08 - 0.025) = £1,864M

Step 5: Discount Everything and Sum

Discount each year's FCF and the terminal value back to the present: PV Year N FCF = Year N FCF ÷ (1 + Discount Rate)^N. PV Terminal Value = Terminal Value ÷ (1 + Discount Rate)^5. Sum all present values to get Enterprise Value. Subtract net debt (total debt minus cash) to get equity value. Divide by shares outstanding to get intrinsic value per share.

Step 6: Compare to Market Price and Apply Margin of Safety

Compare your intrinsic value estimate to the current stock price. If the stock trades at £30 and your DCF says £50, you have a 40% gap — a potentially attractive opportunity if your assumptions are reasonable. If the stock trades at £48 and your DCF says £50, the gap is thin. You have little margin of safety — small errors in your analysis will cost you. The gap between your estimate and the market price is your margin of safety. Benjamin Graham's principle: only buy when the discount is large enough to absorb errors in your own analysis.

Step 7: Sensitivity Analysis

Your DCF output is only as good as your assumptions. Create a sensitivity table varying your discount rate (±1-2%), terminal growth rate (±0.5-1%), revenue growth assumptions (±2-3%), and margin assumptions (±2-3%). For each variation, recalculate intrinsic value. If your conclusion changes materially, you don't have robust conviction. If valuation remains in the same ballpark across reasonable assumption ranges, you have more confidence. Example sensitivity table (discount rate vs. terminal growth rate): at 7% discount and 2-3% terminal growth, values range £58-75; at 8%, £48-62; at 9%, £41-52. The output is deliberately sensitive — the point is to understand it and not pretend your DCF is more precise than it actually is.

Worked Example: A Consumer Goods Company

Historical data: 5% average revenue growth, 18% operating margin, 3% capex/revenue, current revenue £2,000M. Forecast assumptions: 4% revenue growth (slower than historical due to market maturity), stable 18% margin, 3% capex, 8% discount rate, 2.5% terminal growth. FCF forecasts: Year 1 £304M, Year 2 £316M, Year 3 £328M, Year 4 £342M, Year 5 £355M. Terminal Value = £355M × 1.025 ÷ 0.055 = £6,618M. Discounted PVs: Year 1 £281M, Year 2 £271M, Year 3 £260M, Year 4 £251M, Year 5 £242M, Terminal PV £4,506M. Enterprise Value = £5,811M. With 500M shares outstanding and no net debt, intrinsic value per share = £11.62. Stock at £9 looks attractive (29% discount). Stock at £12 is fairly valued. Stock at £15 is expensive.

Common Mistakes to Avoid

Over-precision — your £11.62 per share output is not £11.62; it's "somewhere in the range £10-13 depending on assumptions." Unsupported growth assumptions — do not assume 15% growth for a mature business because you feel good about it; base assumptions on evidence. The terminal value trap — terminal value is 60-80% of your output; a small change in terminal growth rate swings the valuation enormously, so be conservative. Ignoring cyclicality — if you are valuing at peak earnings, your FCF forecast will be too high; use normalised earnings. Ignoring margin compression — do not assume a company maintains historical margins if competitive pressures suggest they will compress.

When DCF Works and When It Doesn't

DCF works beautifully for stable, cash-generative businesses where you can reasonably forecast 5+ years: mature consumer companies, utilities, pharmaceuticals with clear drug pipelines. DCF struggles for high-growth companies (too much uncertainty in cash flow forecasts), cyclical businesses (peak earnings distort forecasts), turnarounds and distressed situations (future too uncertain), and technology disruption scenarios (forecasts become guesswork). Use DCF as one tool among several. If DCF, P/E analysis, and peer comparison all point to the same valuation, you have high confidence. If they diverge, you have more thinking to do. The goal is not to be precisely right — it's to be approximately right, with a margin of safety, so that small errors in your assumptions do not destroy your returns.

The content on Gingernomics is for educational and informational purposes only and does not constitute financial advice. Always do your own research and consult a licensed financial advisor before making any investment decisions. Past performance is not indicative of future results.