A Simple DCF Build Walkthrough
Theory matters, but execution matters more. This article walks through building a simple DCF model on a real business, showing how to move from initial assumptions to a final valuation. The goal is to demystify the process and show that a useful DCF does not require sophisticated modeling.
Quick Definition
A walkthrough DCF is a practical, step-by-step valuation model that converts business assumptions into a present-value estimate. It includes revenue projections, margin assumptions, capital expenditure, working capital, a discount rate, and terminal value—all in a simple, transparent structure that anyone can build and defend.
Key Takeaways
- Start with revenue: historicals, growth assumptions, and drivers.
- Layer in profitability: EBITDA margin, tax rate, and interest expense.
- Estimate reinvestment: capex, depreciation, and working capital changes.
- Calculate free cash flow: EBIT × (1-tax rate) + D&A − capex − change in working capital.
- Apply a discount rate (WACC) and terminal value to get enterprise value.
- From enterprise value, derive implied share price and compare to current price.
- Stress-test the key assumptions and articulate your margin of safety.
The Target Company: GronCo Inc.
Let's build a DCF on GronCo, a hypothetical software-plus-services company. Here are the starting facts:
- Current revenue (Year 0): $200 million.
- Business model: 60% software subscription (high-margin, recurring), 40% professional services (lower-margin, project-based).
- Current profitability: 25% EBITDA margin (driven by software; services are 10% margin).
- Outstanding shares: 50 million.
- Current stock price: $40 per share.
- Management guidance: Revenue growth 15% annually for 3 years, then decelerating to 8% long-term.
We will build a 5-year explicit forecast, then apply terminal value.
DCF Model Flow
Step 1: Project Revenues
Start with historical revenue and build a forward forecast based on assumptions about growth.
| Year | 0 (Current) | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|---|
| Revenue ($M) | 200 | 230 | 265 | 305 | 329 | 355 |
| Growth % | — | 15% | 15% | 15% | 8% | 8% |
Assumption: 15% growth years 1–3 (management guidance + market tailwinds), then deceleration to 8% (market saturation, competition) in years 4–5.
This is conservative: we are moving from a 15% growth story to an 8% story, not assuming perpetual 15%. Most mistakes come from assuming growth does not slow.
Step 2: Project Operating Margins and EBITDA
Next, forecast operating profitability. We will assume software margins expand (from 60% to 65% of revenue) as the company scales and reduces support costs per dollar of software revenue. Services margins remain at 10%.
| Year | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| Software revenue (60% of total) | 138 | 159 | 183 | 197 | 213 |
| Software margin (%) | 62% | 63% | 64% | 65% | 65% |
| Software EBITDA | 86 | 100 | 117 | 128 | 138 |
| Services revenue (40% of total) | 92 | 106 | 122 | 132 | 142 |
| Services margin (%) | 10% | 10% | 10% | 10% | 10% |
| Services EBITDA | 9 | 11 | 12 | 13 | 14 |
| Total EBITDA | 95 | 111 | 129 | 141 | 152 |
| EBITDA margin (%) | 41% | 42% | 42% | 43% | 43% |
Assumption: Software is the margin engine and should expand as a percentage of revenue (it is growing faster than services). Software margins improve modestly (2–3% over 5 years) as operational leverage kicks in. Services margins are stable because the business model is project-based.
Sanity check: A 41–43% EBITDA margin is high but realistic for a software company. It is below the best-in-class SaaS players (50%+) but ahead of traditional services firms (15–25%). This is reasonable.
Step 3: Calculate Taxes and Operating Profit (NOPAT)
Now add taxes. We assume an effective tax rate of 21% (US federal corporate rate), giving us net operating profit after tax (NOPAT).
| Year | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| EBITDA | 95 | 111 | 129 | 141 | 152 |
| D&A | 20 | 23 | 26 | 28 | 30 |
| EBIT | 75 | 88 | 103 | 113 | 122 |
| Tax rate (%) | 21% | 21% | 21% | 21% | 21% |
| Taxes | 16 | 19 | 22 | 24 | 26 |
| NOPAT | 59 | 69 | 81 | 89 | 96 |
Assumption: D&A is roughly 10% of revenue. Most software companies have modest D&A (servers, equipment) and limited capex once they are mature.
Step 4: Estimate Capex and Working Capital Changes
Free cash flow is NOPAT plus D&A minus capex minus changes in working capital.
| Year | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| NOPAT | 59 | 69 | 81 | 89 | 96 |
| D&A | 20 | 23 | 26 | 28 | 30 |
| Capex (% of revenue) | 5% | 5% | 5% | 4% | 4% |
| Capex (absolute) | 11 | 13 | 15 | 13 | 14 |
| Change in NWC | 5 | 5 | 6 | 2 | 3 |
| Free Cash Flow | 63 | 74 | 86 | 102 | 109 |
Assumptions:
- Capex is 5% of revenue in growth years (supporting the expansion), then 4% in years 4–5 (lower reinvestment as growth moderates).
- Change in NWC (net working capital) is roughly 2.5% of incremental revenue (accounts receivable grow with sales; payables grow with costs; the net is small for a software company).
Sanity check: A software company with 43% EBITDA margins and low capex intensity (4–5%) should convert a large portion of EBITDA into free cash flow. We see FCF of ~35–72% of EBITDA, which is reasonable.
Step 5: Calculate Discount Rate (WACC)
We need a rate at which to discount these cash flows. For a software-plus-services company with moderate risk, we will calculate a simplified WACC.
Cost of Equity (using CAPM):
- Risk-free rate: 4.5% (10-year Treasury).
- Equity risk premium: 6.0% (market has historically returned 6–7% above risk-free).
- Beta: 1.2 (software company, moderately volatile relative to market).
- Cost of equity = 4.5% + (1.2 × 6.0%) = 4.5% + 7.2% = 11.7%.
Cost of Debt:
- GronCo has $50 million in debt at 4% interest rate.
- After-tax cost of debt = 4% × (1 − 0.21) = 3.16%.
WACC:
- Enterprise value of equity: ~$2 billion (rough estimate for weighting).
- Debt: $50 million.
- Total enterprise value: ~$2.05 billion.
- Weight of equity: 97.5%.
- Weight of debt: 2.5%.
- WACC = (0.975 × 11.7%) + (0.025 × 3.16%) = 11.40%.
Simplified approach: Many investors use a simpler WACC estimate. For a software company with minimal debt, WACC ≈ cost of equity ≈ 11–12%. We will use 11.5% as our discount rate.
Step 6: Discount Free Cash Flows to Present Value
Now we discount each year's FCF to present value using the discount rate.
| Year | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| FCF | 63 | 74 | 86 | 102 | 109 |
| Discount factor (11.5%) | 0.897 | 0.804 | 0.721 | 646 | 0.579 |
| PV of FCF | 57 | 59 | 62 | 66 | 63 |
Calculation: PV = FCF / (1 + discount rate)^year. For year 1: 63 / 1.115 = 57.
Running total PV of explicit forecast: 57 + 59 + 62 + 66 + 63 = 307.
Step 7: Estimate Terminal Value
Terminal value is the value of all cash flows beyond year 5. We use the perpetuity growth method:
Terminal Value = FCF(Year 5) × (1 + g) / (WACC − g)
where g is the perpetual growth rate.
Assumption: Long-term growth = 3% (in line with GDP + mild market-share gains).
Terminal Value = 109 × 1.03 / (0.115 − 0.03) = 112.3 / 0.085 = 1,321.
PV of Terminal Value: 1,321 × discount factor for year 5 (0.579) = 765.
Step 8: Calculate Enterprise Value and Implied Share Price
Enterprise Value = PV of explicit forecast + PV of terminal value = 307 + 765 = 1,072 million.
Now we adjust for net debt to get equity value:
- Enterprise value: 1,072.
- Less: Net debt (debt − cash): 50 − 10 = 40.
- Equity value: 1,072 − 40 = 1,032 million.
Implied share price = Equity value / shares outstanding = 1,032 / 50 = $20.64 per share.
Step 9: Compare to Current Price and Assess Margin of Safety
Current stock price: $40. Implied fair value (DCF): $20.64. Implied downside: 48%.
Interpretation: The stock trades at a significant premium to our DCF estimate. This divergence suggests either:
- Our DCF is too conservative. We underestimated revenue growth, margin expansion, or terminal growth.
- The market is pricing in higher growth. The market might expect 18% growth (not 15%), or 45% EBITDA margins (not 43%), or terminal growth of 4% (not 3%).
- The stock is overvalued. Market expectations are unrealistic; we should sell or avoid buying.
Let's stress-test to investigate.
Step 10: Sensitivity Analysis and Stress Testing
Let's see how sensitive the valuation is to two key assumptions: revenue growth and terminal growth.
Sensitivity Table: Terminal Growth vs. Revenue CAGR (Years 1–5)
| Terminal Growth | 12% Revenue CAGR | 15% Revenue CAGR | 18% Revenue CAGR |
|---|---|---|---|
| 2.5% | $16 | $20 | $25 |
| 3.0% | $18 | $21 | $27 |
| 3.5% | $21 | $24 | $30 |
Insights:
- At our base case (15% revenue growth, 3% terminal growth): $20.64 implied share price.
- If growth is higher (18% revenue) and terminal growth higher (3.5%): $30 implied share price.
- If growth is lower (12% revenue) and terminal growth lower (2.5%): $16 implied share price.
The stock at $40 is above even the bull case in this sensitivity table. This suggests either:
- Revenue growth will exceed 18%, or
- Margins will expand more than modeled, or
- Terminal growth will exceed 3.5%, or
- The market is wrong.
Step 11: Make a Final Investment Decision
Given the DCF output of $20.64 and the market price of $40:
Option 1: The DCF reveals an overvalued stock. If you trust your assumptions (15% growth is reasonable, 3% terminal growth is reasonable, 43% EBITDA margin is reasonable), then the stock is not attractive at $40. You would require a 30%+ margin of safety, meaning you would buy at $14 or below.
Option 2: Stress-test your assumptions. Before concluding the stock is overvalued, ask: Could revenue growth sustain at 18%? Could margins reach 50%? Could terminal growth justify 4%? If the answer is yes to any, rebuild the DCF with those assumptions. But be honest about whether you are adjusting for new information or for overconfidence bias.
Option 3: Check multiples for a reality test. If GronCo trades at $40 per share and earns $1 per share in year 1 (from our NOPAT of 59 / 50 shares), the P/E is 40x. If cloud-software comparables trade at 30–35x P/E, a 40x multiple suggests the stock is pricier than peers. This aligns with the DCF conclusion.
Conclusion: Based on this DCF walkthrough, the stock appears overvalued. If you had high conviction in a more aggressive forecast, you might justify a higher valuation. But based on reasonable, conservative assumptions, $40 offers insufficient margin of safety. Pass or wait for a lower price.
Common Mistakes in This Walkthrough (and How to Avoid Them)
1. Assuming management's guidance is achievable
GronCo management guided 15% growth. We used it. But what if management has a track record of missing guidance? Or what if the competitive landscape is changing? Always sense-check management guidance against industry trends and competitive dynamics.
2. Assuming margins expand when they might compress
We modeled EBITDA margin expansion from 41% to 43%. What if new competition enters and forces price cuts? Build a scenario where margins stay flat or compress by 100 basis points. See if the valuation still works.
3. Using a single perpetual growth rate when scenarios would be better
We used 3% terminal growth for all scenarios. A more honest approach: model a 30% probability of 2% terminal growth (competitive pressure), 50% probability of 3% (base case), and 20% probability of 4% (sustained moat). Probability-weight the terminal values.
4. Not stress-testing the discount rate
We used 11.5%. What if rates rise and WACC increases to 12.5%? Or decline to 10.5%? A 100-basis-point change in discount rate can swing valuation by 10–15%. Always show sensitivity to WACC.
5. Forgetting the margin of safety
Even if the DCF says fair value is $21, you should not buy at $20. Use a 20–30% margin of safety. In this case, you would buy at $15–$17.
Real-World Examples of DCF Walkthroughs
Simple DCF on Coca-Cola: A mature beverage company with stable 2–3% volume growth, 3–5% price increases, high margins (35%+ EBITDA), and dividend payouts. A simple 5-year explicit forecast, terminal growth of 2.5%, and WACC of 8% would likely justify the current valuation. The DCF would be straightforward and reliable.
Complex DCF on Tesla: A high-growth auto-plus-energy company with uncertain margins, ramping production, and multiple business lines. A simple 5-year forecast is insufficient; you need scenarios. Bull case (50% gross margins on cars, energy revenues boom) vs. bear case (20% gross margins, energy stalls). The DCF requires explicit scenario weighting because the outcome is binary.
DCF on a private company acquisition: You are considering acquiring a private software company with $10 million in revenue, 20% growth, and 15% EBIT margin. A simple 5-year explicit forecast (growing to $25 million revenue, 25% EBIT margin) plus 3% terminal growth, discounted at 12% WACC, might justify a 5–6x revenue multiple ($50–60 million enterprise value). This is a practical, fast DCF that helps with acquisition pricing.
FAQ
Q: Should I use historical growth or management guidance?
A: Start with management guidance, but adjust down by 50% for conservatism or if the company has a track record of missing. Better yet: use historical growth as a floor (the company can at least repeat the past), and guidance-minus-a-haircut as the ceiling.
Q: How do I estimate capex?
A: Look at historical capex as a percentage of revenue. For a software company, it is typically 3–6% of revenue. For a capital-intensive business (manufacturing, utilities), it is 8–15%. If the company is ramping, capex might be higher; if it is mature, lower.
Q: Should I forecast D&A, or just use historical numbers?
A: Forecast D&A as roughly 5–10% of revenue for software companies, 10–15% for industrial companies. Or, simply project it as a percentage of capex (since depreciation is largely the cash effect of historical capex).
Q: What if the company has no debt or has net cash?
A: Adjust the final equity value. If net cash is $100 million, add it to equity value. If net debt is $50 million, subtract it. The DCF is built on enterprise value (unlevered); adjusting for debt/cash gets you to equity value.
Q: Can I use a shorter explicit forecast period (3 years instead of 5)?
A: Yes, if it simplifies the model and the error is small. A 3-year forecast followed by terminal value is quicker and sometimes clearer. But be aware that a shorter forecast period makes terminal value even more dominant in the valuation. For stability, 5 years is a good default.
Q: What if I think the company will be acquired in year 3?
A: Instead of using perpetual terminal value, use an exit multiple. For year 3, assume the company trades at 8x EBITDA (typical for the industry). That becomes your terminal value assumption. This is more honest than modeling perpetuity if you expect M&A.
Related Concepts
- Free cash flow (FCF) — Cash flow available to all investors (debt and equity) after capital expenditure and working capital changes.
- Net operating profit after tax (NOPAT) — Operating profit after adjusting for taxes; the starting point for FCF.
- Weighted average cost of capital (WACC) — The blended cost of equity and debt; the discount rate used to present-value future cash flows.
- Enterprise value — The total value of a company to all investors (equity + debt), before adjusting for net debt.
- Terminal value — The value of all cash flows beyond the explicit forecast period; usually 60–80% of DCF value.
- Margin of safety — A discount to fair value required to protect against forecast error and valuation uncertainty.
Summary
Building a practical DCF is not difficult if you follow a step-by-step process: forecast revenue, project operating margins, calculate free cash flow, apply a discount rate, and estimate terminal value. The example of GronCo shows that a useful DCF can be built in a spreadsheet without sophisticated modeling.
The hard part is not the math; it is the judgment. Estimating revenue growth, margin expansion, capex, and terminal growth requires business insight. Picking a discount rate requires understanding the company's risk. The DCF forces you to make these judgments explicit, which is where the value lies.
Use the walkthrough as a template. Build your own models. Stress-test the assumptions. Compare to multiples. Maintain a margin of safety. And remember: the output is a starting point for thought, not the end of analysis.