How professionals estimate what a company is actually worth — discounted cash flow, comparable analysis, and dividend discount models. The tools that separate informed investing from guessing.
1. Explain the concept of intrinsic value and why it differs from market price.
2. Build a simplified discounted cash flow (DCF) model step by step.
3. Use comparable company analysis ("comps") to triangulate valuation.
4. Apply the dividend discount model to income-generating stocks.
5. Recognize the limitations of every valuation model and where each breaks down.
6. Combine multiple methods to arrive at a valuation range, not a single number.
The stock market gives you a price — $84 per share for TechCorp. But price and value are not the same thing. Intrinsic value is an estimate of what a company is actually worth based on its ability to generate cash over time. If intrinsic value is $100 and the stock trades at $84, the stock is undervalued — you're buying a dollar for 84 cents. If intrinsic value is $70, the stock is overvalued.
The gap between intrinsic value and market price is the margin of safety. Benjamin Graham, the father of value investing, argued that investors should only buy when the market price is significantly below intrinsic value — the larger the margin of safety, the less likely you are to lose money even if your analysis is partially wrong.
Every valuation model is an estimate. They require assumptions about the future that are inherently uncertain. The goal is not to calculate a precise intrinsic value to two decimal places — it's to determine whether a stock is approximately cheap, fairly valued, or expensive. If three different methods all suggest a stock is worth $90–$110 and it trades at $65, you don't need more precision. If they suggest $80–$120, the signal is weaker and you should demand a larger margin of safety.
The DCF model is the gold standard of fundamental valuation. Its logic is simple: a company is worth the total of all the cash it will generate in the future, discounted back to today's dollars. A dollar earned ten years from now is worth less than a dollar today because of the time value of money and the risk of not receiving it.
Using TechCorp's current FCF of $1,950M (Module 3.3), project how it will grow over the next five years. This requires assumptions about revenue growth, margin trends, and capital expenditure needs. For TechCorp, assume 10% FCF growth for years 1–3 (established growth), slowing to 7% in years 4–5 (maturation):
| Year | Year 1 | Year 2 | Year 3 | Year 4 | Year 5 |
|---|---|---|---|---|---|
| Growth Rate | 10% | 10% | 10% | 7% | 7% |
| Projected FCF ($M) | $2,145 | $2,360 | $2,595 | $2,777 | $2,971 |
| Discount Factor (10%) | 0.909 | 0.826 | 0.751 | 0.683 | 0.621 |
| Present Value ($M) | $1,950 | $1,949 | $1,949 | $1,897 | $1,843 |
Sum of 5-year present values: $9,588M
The terminal value captures all cash flows beyond year 5, in perpetuity. The most common method is the Gordon Growth Model:
TechCorp: $2,971M × (1 + 2.5%) / (10% − 2.5%) = $3,045M / 7.5% = $40,603M
Discounted to present: $40,603M × 0.621 = $25,214M
Enterprise value = sum of present values + present value of terminal value = $9,588M + $25,214M = $34,802M
To convert to equity value: subtract net debt ($4,300M debt − $3,200M cash = $1,100M net debt).
Equity value = $34,802M − $1,100M = $33,702M
Per-share value = $33,702M ÷ 500.5M shares = $67.34
Our DCF suggests TechCorp is worth $67 per share. The stock trades at $84. Does this mean it's overvalued? Maybe — but notice how sensitive the result is to assumptions. Changing the discount rate from 10% to 9% increases the value to ~$80. Changing the perpetual growth rate from 2.5% to 3% pushes it to ~$75. Increasing the growth projections by 2% per year gets you to ~$85. This is the fundamental weakness of DCF: small changes in assumptions produce large changes in output. Never treat a DCF as producing a precise answer — it produces a range that depends on your assumptions.
Always run your DCF at multiple discount rates and growth assumptions to see how the output changes. This is called a sensitivity analysis:
| Discount Rate → Growth Rate ↓ | 8% | 9% | 10% | 11% | 12% |
|---|---|---|---|---|---|
| 1.5% | $72 | $63 | $56 | $50 | $46 |
| 2.0% | $80 | $69 | $61 | $54 | $49 |
| 2.5% | $90 | $77 | $67 | $59 | $53 |
| 3.0% | $104 | $87 | $75 | $65 | $58 |
| 3.5% | $124 | $101 | $85 | $73 | $64 |
The valuation ranges from $46 to $124 depending on assumptions. This isn't a bug — it's the honest reality of forward-looking valuation. The table tells you that at $84, TechCorp is fairly valued if growth and risk assumptions are moderate, cheap if growth is strong, and expensive if the discount rate is high (i.e., the market demands a high return for the risk involved).
Comparable company analysis values a company by applying the valuation multiples (Module 3.4) of similar public companies. The logic: if TechCorp's competitors trade at 15x EV/EBITDA on average, and TechCorp's EBITDA is $3.35B, then TechCorp should be worth roughly 15 × $3.35B = $50.2B in enterprise value.
Choose 4–8 companies in the same industry, of roughly similar size, with similar business models. For TechCorp, you might select companies in the same GICS sub-industry with $5B–$50B in revenue.
Pull P/E, EV/EBITDA, P/S, and EV/FCF for each peer. Note the median and the range.
If the peer median EV/EBITDA is 14x: TechCorp implied EV = 14 × $3,348M = $46,872M. Minus net debt ($1,100M) = $45,772M equity value ÷ 500.5M shares = $91.45 per share.
If the peer median P/E is 22x: TechCorp implied price = 22 × $4.16 EPS = $91.52.
Strengths: Reflects what the market is actually paying for similar companies right now. Simple, fast, and anchored in market reality. Doesn't require forecasting future cash flows.
Weaknesses: If the entire sector is overvalued, comps will tell you the stock is "fairly valued" relative to overvalued peers. "Cheap relative to an expensive group" isn't the same as "cheap." Also, no two companies are truly comparable — differences in growth rates, margins, and risk make every comparison imperfect. Comps tell you what the market is paying; DCF tells you what you should pay.
The DDM values a stock as the present value of all future dividends. It applies only to companies that pay dividends and is most useful for mature, stable dividend payers (utilities, consumer staples, banks).
TechCorp paid $420M in dividends on 500.5M shares = $0.84 per share. Assuming 8% dividend growth and a 10% required return:
Value = ($0.84 × 1.08) / (10% − 8%) = $0.907 / 2% = $45.36
This value is much lower than our other methods because TechCorp pays a relatively small dividend (20% payout ratio). DDM works best for companies that distribute most of their earnings as dividends. For growth companies that reinvest earnings, DCF or comps are more appropriate.
Notice the formula's denominator: r − g. If g approaches r, the value approaches infinity. If g exceeds r, the formula produces a negative number (meaningless). DDM requires that the discount rate exceed the growth rate — it's designed for stable, moderate-growth dividend payers, not high-growth companies.
No single method gives you "the answer." Professional analysts use multiple methods and look for convergence:
| Method | TechCorp Value/Share | Confidence |
|---|---|---|
| DCF (base case) | $67 | Moderate — sensitive to assumptions |
| DCF (optimistic) | $85–$104 | If growth exceeds expectations |
| Comps (EV/EBITDA) | $91 | High — reflects market pricing |
| Comps (P/E) | $92 | High — peer-validated |
| DDM | $45 | Low — company reinvests most earnings |
The DCF base case ($67) is the most conservative. The comps ($91–$92) reflect what the market pays for similar companies. The DDM ($45) is less relevant because TechCorp retains most of its earnings. A reasonable valuation range is $67–$92, with the current price of $84 sitting in the middle. TechCorp looks approximately fairly valued — not a screaming buy, not obviously expensive.
If the stock trades well below the low end of your range: Strong buy signal (large margin of safety).
If it trades within the range: Fairly valued — hold if you own it; wait for a pullback to buy.
If it trades well above the high end: Potentially overvalued — consider trimming or avoiding.
The single biggest misuse of valuation models is confirmation bias — starting with the conclusion you want ("this stock is cheap") and adjusting assumptions until the model agrees with you. If you want a DCF to show $100, you can get there by assuming 15% growth for ten years and a 3.5% terminal growth rate. The model will dutifully produce $100. But you haven't discovered value — you've manufactured it.
The antidote: Always run the model before forming an opinion about the stock's value. Use conservative assumptions for your base case. Let the sensitivity analysis show you the range of outcomes. If you need heroic assumptions to justify the current price, the stock is probably not a good investment — even if you love the company.
During the dot-com bubble (1998–2000), Wall Street analysts created increasingly creative valuation methods to justify astronomical stock prices. Companies with no revenue were valued on "eyeballs" (unique visitors), "mind share," and "total addressable market" — metrics disconnected from any actual cash generation.
The poster child was Pets.com, which sold pet supplies online at prices below cost, generating negative gross margins. Its stock peaked at $14 after its IPO in February 2000. By November 2000, it was bankrupt. The company's "valuation" had been based entirely on growth projections that assumed it would eventually become profitable — an assumption supported by nothing but optimism.
The lesson: every valuation model requires inputs grounded in financial reality. Revenue, margins, cash flow, return on capital — these are the building blocks. When an analyst tells you to ignore current financials and value a company on "the vision," you're not investing; you're speculating. Speculation has its place (Module 13), but it should never be confused with fundamental valuation.
Complete this exercise to build muscle memory with DCF analysis:
Step 1: Choose a company from your watchlist (Module 2.4) that you've already analyzed in Module 3's practical exercises.
Step 2: Pull the company's current free cash flow from the cash flow statement (Module 3.3), share count, and net debt from the balance sheet (Module 3.2).
Step 3: In a spreadsheet, project FCF for 5 years using growth rates based on historical trends and analyst estimates (available on Finviz or your broker's research page).
Step 4: Discount at 10% (a reasonable starting point for most stocks). Calculate terminal value using the Gordon Growth Model with 2.5% perpetual growth.
Step 5: Run a sensitivity analysis varying the discount rate (8–12%) and growth rate (1.5–3.5%). Does the current stock price fall within, above, or below your valuation range?
Step 6: Compare your DCF result to a comps analysis using the peer group's median P/E and EV/EBITDA. Do the methods agree?
The financial data for this exercise comes from Modules 3.1–3.4 (financial statements and ratios). The peers for comps analysis come from Module 4.3 (sector/industry identification). Module 4.6 (Building a Thesis) will tie your valuation into a complete investment thesis with entry criteria and risk factors.
1. What does intrinsic value represent?
2. In a DCF model, what is the biggest driver of the final valuation?
3. A company's peer group trades at a median P/E of 18x. The company earns $3.50 EPS. What is the comps-implied price?
4. What is the primary weakness of comparable company analysis?
5. When is the Dividend Discount Model (DDM) most appropriate?
6. An analyst builds a DCF that requires 25% annual growth for 10 years to justify the current stock price. What should you conclude?