Discounted Cash Flow, in Plain English
The theory every multiple is a shortcut for: a business is worth the cash it will produce, discounted for the fact that future cash is worth less than cash today.
MadStockAlerts Research · Updated August 28, 2026
What to take away
- Value equals future cash flows discounted back to the present.
- The discount rate reflects the risk and the time value of money.
- The terminal value usually dominates the answer, which is where the fragility lives.
- A discounted cash flow is most useful for showing what a market price already assumes.
- Two analysts with slightly different assumptions can differ by fifty percent.
MAD Academy Training Video · 0:46
Precision Without Accuracy
A DCF produces a number to the cent from assumptions that are guesses, and its real value is showing what has to be true.
This lesson is part of a Stock Alerts + Tools plan.
The idea
A dollar next year is worth less than a dollar today, because today's dollar can be invested and because next year's might not arrive. Discounting applies a rate to convert future amounts into present ones, and the value of a business is the sum of all its future cash converted that way.
present value = CF1/(1+r) + CF2/(1+r)^2 + ... + terminal value/(1+r)^n
- CF: free cash flow in each forecast year
- r: the discount rate, usually a weighted average cost of capital
- terminal value: everything beyond the explicit forecast
Every multiple in this pillar is a compressed version of this calculation. A P/E ratio is the answer this model would give if growth, risk and reinvestment took particular values, which is why the multiple changes when any of those changes.
The three inputs
- 1Forecast cash flowsUsually five to ten years, built from revenue growth, margins and reinvestment.
- 2Choose a discount rateHigher for riskier or more leveraged businesses. A percentage point here moves the answer a lot.
- 3Estimate a terminal valueEither a perpetual growth rate or an exit multiple applied to the final year.
The discount rate is where the model connects to the rest of the market. It is built from the risk-free bond yield plus a premium for equity risk, which is why every valuation in the market moves when Treasury yields move.
Why the terminal value dominates
In a typical ten-year model, sixty to eighty percent of the calculated value sits in the terminal value, which is the least knowable part. Small changes in the perpetual growth rate move the total dramatically.
This is why a discounted cash flow can be made to produce almost any answer. Two analysts choosing a discount rate one point apart and a terminal growth rate half a point apart can differ by fifty percent on the same company, with both models internally consistent.
There is also a discipline check worth applying: a perpetual growth rate above the long-run growth rate of the economy implies the company eventually becomes the entire economy. Models routinely contain that assumption without anyone noticing.
Scroll the chart sideways to see all of it.
The reverse DCF
The more robust use is to run it backwards. Take the market price as given and solve for the growth and margin path that justifies it.
The output is a statement about expectations rather than a target price, and it is far easier to judge whether an implied growth rate is plausible than to assert what a company is worth. A price that requires twenty-five percent growth for a decade is a testable claim; a fair value estimate is not.
Why the output is a range
A discounted cash flow produces a single number, and that number carries a false precision that is the source of most of the criticism directed at the method. The inputs are estimates with wide uncertainty, and multiplying several of them together produces an output with wider uncertainty than any of them.
The conventional response is a sensitivity table: the value computed across a grid of discount rates and terminal growth rates, so that the range is visible rather than implied. A model whose value ranges from forty to a hundred and ten across reasonable assumptions has said something, and what it has said is that the answer is not knowable to the precision the point estimate suggests.
The most useful output of the exercise is usually not the value. It is the discovery of which input the answer actually depends on, because that identifies what has to be researched and what can be left approximate.
This is also the argument for the reverse construction. Rather than forecasting inputs to produce a value, take the market price as given and solve for the growth rate that justifies it. The result is a testable statement about what the market is assuming, which is a more answerable question than what the company is worth.