Skip to content
StockSummarizer
Menu
Valuation & multiplesHub

How to Build a Simple DCF Without Fake Precision

A plain-English, five-input discounted cash flow (DCF) for Canadian DIY investors: free cash flow, growth, discount rate, terminal value and a sensitivity grid — with a fully worked hypothetical example and the traps that create fake precision.

By Published Updated

The neoclassical columned facade of the Bank of Montreal head office on Place d'Armes in Old Montreal
Bank of Montreal head office, Place d'Armes, Montréal, August 2022. Photo by D. Benjamin Miller, CC0, via Wikimedia Commons. Cropped and resized.

A discounted cash flow (DCF) model estimates what a business is worth today based on the cash it might generate in the future. Done badly, it produces a precise-looking number ($47.83!) built on guesses. Done well, it produces a range and tells you which assumption the price depends on.

Educational only — not financial advice. Every number below is hypothetical.

Direct answer

Use five inputs: starting free cash flow, a growth rate for 5 years, a discount rate, a terminal growth rate, and net debt + share count. Discount each year’s FCF, add a terminal value, subtract net debt, divide by shares. Then build a sensitivity grid and report a range. If most of the value comes from the terminal value (it usually does), your real bet is on the long-term assumptions — be honest about that.

The five-input model

1. Starting FCFNormalized,trailing or mid-cycle2. Grow 5 yearsOne growth rateyou can defend3. DiscountRequired return(e.g. 8–10%)4. Terminal valueModest perpetualgrowth (1–3%)5. Equity / shareSubtract net debt,divide by shares
Keep it this simple. Adding twenty line items rarely improves accuracy; it mostly hides assumptions.
InputHypothetical choiceHow to choose yours
Starting FCF$100M (year 1)Normalized FCF from FCF analysis; mid-cycle for cyclicals
Growth, years 1–55% a yearAnchor on history and guidance; be skeptical above long-run revenue growth
Discount rate9%Your required return; higher for riskier, more levered or cyclical names
Terminal growth2%At or below long-run nominal economic growth; never above the discount rate
Net debt / shares$400M / 50M sharesLatest balance sheet; diluted shares

Worked example (hypothetical)

YearFCF ($M)Discount factor at 9%Present value ($M)
1100.00.91791.7
2105.00.84288.4
3110.30.77285.1
4115.80.70882.0
5121.60.65079.0
Sum, years 1–5426.3

Terminal value at end of year 5 = FCF₅ × (1 + g) ÷ (r − g) = 121.6 × 1.02 ÷ (0.09 − 0.02) ≈ $1,771M, worth ≈ $1,151M today (× 0.650).

Step$M
PV of years 1–5426
PV of terminal value1,151
Enterprise value1,577
− Net debt−400
Equity value1,177
÷ 50M shares≈ $23.5 per share

Where the value comes from

0 $M500 $M1000 $M1500 $M2000 $M426 $MPV years1–51151 $MPV terminalvalue1577 $MEnterprisevalue
In this hypothetical, about 73% of enterprise value comes from the terminal value — the years you can't see.

The sensitivity grid — your real output

Enterprise value ($M) as the discount rate and terminal growth change (5% growth in years 1–5, hypothetical):

Discount rate ↓ / terminal growth →1%2%3%
8%1,6321,8442,142
9%1,4241,5771,782
10%1,2621,3771,526

Per share after $400M net debt and 50M shares, that’s roughly $17 to $35 — with the base case at about $23.5. A one-point change in the discount rate moves value by ~13–20%. That’s why a single-number DCF target is fake precision.

Growth in years 1–5 (r = 9%, g = 2%)2%5%8%
Enterprise value ($M)1,4291,5771,739

Near-term growth matters less than most people expect; the discount rate and terminal assumptions dominate.

Turning the range into a decision

Price vs your DCF rangeReasonable reading
Below the low endMarket is more pessimistic than your worst case — check what you’re missing, then consider it
Inside the rangeFairly priced on your assumptions; other factors decide
Above the high endMarket assumes more than your best case — need a reason you’re too conservative

You can also reverse the DCF: solve for the growth rate the current price implies, and ask whether that’s believable.

Fake-precision traps

  1. Cents in the output (report a range, rounded).
  2. Terminal growth near the discount rate — value explodes toward infinity.
  3. Peak-cycle FCF as the starting point for a producer or miner.
  4. Forgetting net debt, leases, preferreds, or dilution.
  5. Using a DCF for banks — use P/B and ROE instead.
  6. Ignoring the market: always compare to relative multiples like P/FCF and EV/EBITDA.

Put the final range into the valuation block of your stock brief.

Go deeper

A spreadsheet version of this five-input DCF is planned for the Valuation Without Fake Precision ebook. The free method above is complete.

Media credits

  • Hero: “Bank of Montreal Head Office, Aug 31 2022.jpg” by D. Benjamin Miller, CC0, via Wikimedia Commons. Cropped and resized to 1280×720. Real photograph, not AI-generated.

By Andrew · stocksummarizer.com

Guides · coming soon

Want the playbook in one place?

We are compiling free literacy hubs into paid ebook playbooks (Research, Valuation without fake precision, Dividend safety). Soft launch only for now — keep using these free guides, and browse ebooks when a guide ships.

Preview ebook shelf

Educational only — not financial advice. Do your own research. See our disclosures.