How to Audit a DCF: Where It Usually Breaks

A DCF is short compared to an LBO, but almost all of its value sits in a few cells. The terminal value alone is often more than half the enterprise value. That makes DCF mistakes expensive: one wrong reference in the wrong place can move the share price by a third.

This guide walks through where DCFs break, using SheetRank's Institutional 3-Statement + DCF (Project Catalyst). It's a subscription business with $500 million of Year 0 revenue, a 7.6% WACC, a 10x exit multiple, $100 million of net debt and 100 million shares. Built correctly, it shows an enterprise value of $1.26 billion and an implied share price of $11.59.

What each mistake does to the share price

Every one of these still produces a share price that looks like a share price. Here's how to find each one.

Try it

Work out enterprise value and EV/EBITDA

Equity value plus debt, minus cash, is enterprise value. Divide by EBITDA for the multiple.

Equity value: fill B6 (1 of 3)

	A	B
1	Share price	$40
2	Shares outstanding	50,000,000
3	Debt	$600,000,000
4	Cash	$150,000,000
5	EBITDA	$245,000,000
6	Equity value
7	Enterprise value
8	EV / EBITDA

fxB6=tap cells and keys

789()456×÷123+−0.,:^

fx$

Check

Tap cells to put them in the formula. On a laptop you can type the numbers and operators.

1. The terminal value

Start here, because it's the biggest number in the model.

Is it discounted? The terminal value is a Year 5 value, so it needs the Year 5 discount factor. Adding it to the sum of discounted cash flows without discounting it inflates enterprise value from $1.26 billion to $1.69 billion. That's the single most common DCF mistake I see in student models, and it's a 37% overstatement.

Which year's EBITDA? The exit multiple applies to final-year EBITDA (here, about $142 million in Year 5). Point it at Year 1 instead and the terminal value drops from $1.42 billion to $1.08 billion, and the share price to $9.19.

Does the growth rate make sense? If the model uses a perpetuity growth method, the growth rate has to be below the discount rate and should be in line with long-run nominal GDP. A 6% perpetual growth rate against a 7.6% WACC isn't a valuation, it's a typo.

2. The discounting

Count the periods. Year 1 cash flow is discounted one period, Year 5 is discounted five. Copying a discount formula across a row is where off-by-one errors come from, so click the last year and check the exponent.

Match the cash flow to the factor. A discount factor pulled from one column and a cash flow from the next is a common wrong-year mistake. The result is close enough to look fine.

Mid-year convention. If the model uses it, every period shifts by half a year, including the terminal value under the perpetuity method. Mixing conventions inside one model is a mistake even if each piece is right on its own.

3. Free cash flow

Unlevered free cash flow is NOPAT, plus D&A, minus capex, minus the increase in net working capital. Each piece has a classic failure:

• D&A not added back. In this model that understates free cash flow by about $21 million in Year 1 and takes the share price from $11.59 to $10.66.
• Working capital sign flipped. A growing business ties up cash in working capital, so the increase is a use of cash. Flip the sign and the share price rises to $11.80. Small, which is why it survives.
• Interest in the cash flow. Unlevered free cash flow excludes interest. If the model starts from net income instead of EBIT, interest has leaked in and is being double counted against the WACC.

4. The bridge from enterprise value to share price

Equity value is enterprise value minus debt plus cash. Then divide by diluted shares.

• Net debt sign. Adding net debt instead of subtracting it moves this share price from $11.59 to $13.59. It reads as a reasonable number, so it's easy to accept.
• Which balance sheet date? Use the debt and cash as of the valuation date, not a projected year.
• Diluted shares. Options and convertibles that are in the money belong in the share count.

5. WACC

Check the inputs, not just the formula. The cost of equity here is 4.0% risk-free plus 1.2 beta times a 5.5% equity risk premium, which is 10.6%. Then make sure the weights add to 100%, the cost of debt is after tax, and the tax rate matches the one used in the cash flows.

A ten-minute DCF review

Then flex WACC and the terminal assumption by a point each way. The share price should move a lot. If it barely moves, something is disconnected.

Practice on the real model

You can audit this exact DCF in Audit Mode: the full three statements and valuation, built, with five mistakes hidden in it. Every number gets graded when you check, so the round only passes when the whole model is right again. For the interview side of DCFs, see WACC and CAPM explained.