In partnership with
/
I have built hundreds of discounted cash flow models over the years. Here is what I wish someone had told me before I trusted my first one.
Ask any analyst how to value a company and you will hear the same three letters within a minute. DCF.
It is the model on every finance syllabus, every interview, every pitch deck. It is also, in my experience, the model people get wrong more often than any other, precisely because it looks so rigorous.
A DCF gives you a single, confident number. Intrinsic value, to the dollar. That precision is the trap. Change one assumption most people never stress-test, and that same model will happily hand you a number that is double, or half.
The moving parts of a DCF. Every one of them is an assumption.
The 60-Second Version
A DCF values a company as the present value of its future cash flows, discounted for time and risk.
Done right, it is the most honest way to value a business, because it forces you to state every assumption out loud.
Done wrong, it is false precision. The terminal value alone often drives the majority of the answer.
The fix is not a better formula. It is a sensitivity table and the discipline to use it.
Speak naturally. Send without fixing.
Wispr Flow turns your voice into clean, professional text you can send the moment you stop talking. Not rough transcription you have to clean up. Actual polished text — ready for email, Slack, or any app.
Speak the way you think. Go on tangents. Change your mind mid-sentence. Flow strips the filler, fixes the grammar, and gives you text that reads like you spent five minutes writing it.
89% of messages sent with zero edits. Millions of professionals use Flow daily, including teams at OpenAI, Vercel, and Clay. Works on Mac, Windows, and iPhone.
What a DCF actually is
Strip away the jargon and a DCF answers one question: how much is a stream of future cash worth in today's money? A rupee, or a dollar, five years out is worth less than one today, so you discount each future cash flow back to the present and add them up.
On WallStreetMojo we teach it as seven steps, and the picture below is the one I always start with.
The seven steps, from projecting free cash flow to the final sensitivity check.
The engine is free cash flow to the firm. You forecast it out five to ten years, discount everything at the weighted average cost of capital, WACC, and then add a terminal value for every year beyond your forecast. That last piece is where the trouble starts.
Where it goes wrong
Problem one: the terminal value tail wags the dog. In a typical five-year DCF, the terminal value, the single number standing in for the entire life of the business after year five, routinely accounts for 60 to 75 percent of the total valuation. Think about that. Most of your answer comes from the part you forecast least carefully.
Terminal value. A tiny change in the growth rate here moves the entire valuation.
Problem two: WACC is quietly subjective. The discount rate feels like math, but it is stitched together from a cost of equity, a beta, an equity risk premium, and a capital structure, each of which is an estimate. Nudge WACC by one percentage point and you can swing the valuation by 15 to 20 percent. Two honest analysts can build the same model and disagree by a third, without either making an error.
None of this means the model is broken. It means the output is a range wearing the costume of a point estimate.
The test I run before I trust a DCF
I no longer look at the headline number first. I look at the sensitivity table, the grid that shows how the valuation moves as WACC and the terminal growth rate change.
If the reasonable corners of that grid still tell a consistent story, the company looks undervalued across most of it, I trust the conclusion. If a small, defensible change flips the answer from buy to sell, the DCF has told me something more useful than a target price. It has told me this is not a DCF situation.
For fast-growing or pre-profit companies, that is often the case, and I lean on relative valuation or a reverse DCF, which flips the model around and asks what growth the current price already implies.
So should you use it?
Yes. Emphatically. Not because it gives you the right number, but because it forces you to write down what you believe about a business and then see what those beliefs are worth.
Use it for stable, cash-generative, forecastable businesses. That is its home turf.
Never present the single number without the sensitivity table beside it.
Watch the terminal value share. If it is over 75 percent, your forecast horizon is doing too little work.
Treat a DCF as a way to test a thesis, not to manufacture one.
The full step-by-step DCF guide, with the Alibaba worked example, is on the site if you want to build one properly from scratch.
Reply and tell me the valuation topic you want me to take apart next. I read every response.
Dheeraj Vaidya, CFA, FRM
Founder, WallStreetMojo
P.S. If you want professional valuation training plus everything else we teach, our WallStreetMojo Elite Membership includes the Valuation modeling along with 50+ courses and certifications covering Investment Banking, Valuation, Equity Research, AI, and more.