Walk me through a DCF

A discounted cash flow analysis, or DCF, is the most-tested valuation method in finance interviews. It is also the one that scares beginners most.

It should not. The whole thing rests on one idea you already understand: money you get later is worth less than money you get now.

Everything else is bookkeeping around that idea.

Interviewers ask this because a DCF touches almost everything — accounting, risk, growth, and judgment. If you can walk through one clearly, you have shown you understand most of the toolkit.

Aim for about two minutes. Not thirty seconds, and not five.

The one idea underneath everything

Imagine someone offers you a choice. $100 today, or $100 in five years.

You take it today. Obviously.

Why? Two reasons. You could invest that money and have more than $100 in five years. And there is always a chance you never get paid at all.

So future money is worth less than present money. To compare them fairly, you have to shrink the future amount down to what it is worth today.

That shrinking is called discounting. And a DCF is just that idea, applied to every dollar a business will ever make.

What a DCF actually says

A DCF says: a business is worth all the cash it will produce in the future, converted into today's money.

Not what the stock market thinks it is worth. Not what someone paid for a similar company. Just the cash.

That is why it is called an intrinsic valuation. It looks at the thing itself.

The five steps

Step 1: Project the cash flows

You start by estimating how much cash the business will produce each year for the next five to ten years.

The specific measure you want is called unlevered free cash flow. That sounds intimidating. It means: the cash the business generates, before paying any interest on its debt.

"Unlevered" means before interest. That matters because this cash belongs to everyone who funded the business — lenders and shareholders both.

Here is how you build it, line by line:

Start with operating profit (EBIT). This is profit from running the business, before interest and tax.

Subtract tax. The company has to pay tax on its profits.

Add back depreciation and amortisation. These are costs that appear on the accounts but do not involve any money leaving the company. Depreciation spreads the cost of something like a machine over the years you use it. The cash went out when you bought the machine, not now. So you add it back.

Subtract capital expenditure (capex). This is real money spent on things like buildings, machines and equipment. It is cash going out, so it comes off.

Subtract the increase in working capital. Working capital is the money tied up in the day-to-day running of the business — stock sitting in a warehouse, invoices customers have not yet paid. When a business grows, more money gets tied up like this, and that is cash you cannot use.

What you are left with is the real cash the business produces.

Step 2: Choose a discount rate

Now you need the rate to shrink those future cash flows by.

For unlevered cash flow, that rate is WACC — the weighted average cost of capital.

In plain terms: WACC is the return the business must earn to keep everyone who funded it satisfied. Lenders want their interest. Shareholders want a return for taking the risk. WACC blends the two together, weighted by how much money each group provided.

A riskier business needs a higher return to attract funding, so it has a higher WACC. A higher WACC shrinks future cash flows harder, which produces a lower value. That is the maths agreeing with common sense: riskier businesses are worth less, all else equal.

There is one rule you must not break. Unlevered cash flow goes with WACC. If you use cash flow that is already after interest, you must use the cost of equity instead. Mixing them counts the benefit of debt twice, and it is the fastest way to show you have never built one of these.

Step 3: Add a terminal value

Here is a problem. Businesses do not stop after year five. But nobody can forecast year forty-seven.

So you forecast in detail for five or ten years, then add one single number covering everything after that. That number is the terminal value.

There are two ways to work it out.

The perpetuity growth method. You assume the business grows slowly forever at a steady rate. That rate must be small — usually 1 to 3 percent. It can never be higher than the growth of the whole economy, because a business growing faster than the economy forever would eventually become the economy.

The exit multiple method. You assume the business is sold at the end of your forecast, at a multiple similar to what comparable companies trade at today. This is more common in banking because it is grounded in real market data.

One thing surprises people: the terminal value is usually 60 to 80 percent of the total answer. Most of the value sits in the number you had to estimate, not the years you carefully forecast. Saying this out loud in an interview shows you have seen a real model.

And do not forget to discount the terminal value back too. It sits at the end of your final forecast year, so it gets shrunk down like everything else.

Step 4: Add it all up

Discount each year's cash flow back to today. Discount the terminal value back to today. Add them all together.

What you get is the enterprise value — the value of the whole operating business.

Step 5: Bridge to equity value

Usually you want to know what the shares are worth, not the whole business. So you do one final step.

Equity value = enterprise value − debt + cash

Subtract debt, because lenders have to be paid before shareholders see anything. Add cash, because that money belongs to the owners too.

Divide by the number of shares and you have a value per share, which you can compare to where the stock actually trades.

Why you never give one number

A DCF is built on estimates. Small changes in the discount rate or the growth rate cause large changes in the answer.

Because of that, nobody presents a DCF as a single number. You present a table showing what the value would be at different discount rates and different growth rates. That table is called a sensitivity analysis.

Mentioning it unprompted is a good signal. It says you understand that the output is a range built on judgment, not a fact.

Saying it out loud

Something like this, in about two minutes:

"A DCF values a business based on the cash it will generate in the future, converted into today's money.

You start by projecting unlevered free cash flow for five to ten years. You get there from operating profit, minus tax, plus depreciation, minus capex, minus the increase in working capital.

Then you discount those cash flows back at WACC, which is the blended cost of the company's debt and equity.

Because you cannot forecast forever, you add a terminal value covering everything past the final year. You can do that with a small perpetuity growth rate or with an exit multiple. That terminal value is usually most of the total, and you discount it back too.

Add everything up and you get enterprise value. Subtract debt and add cash to get to equity value, then divide by the share count.

And because the answer moves a lot with the assumptions, you would show it as a sensitivity table rather than one number."

What the interviewer is really checking

Do you know the steps in order?

Cash flows, then discount rate, then terminal value, then the bridge to equity. Someone who has built a DCF knows this order without thinking. Someone who has read about one tends to jump around.

Do the cash flow and the rate match?

This is the big one. Unlevered cash flow goes with WACC. Levered cash flow goes with the cost of equity. Getting this wrong is the clearest possible sign that you have not built one.

Do you know where the value comes from?

The terminal value is usually 60 to 80 percent of the answer. Volunteering that fact tells the interviewer you have seen a real output page rather than a textbook diagram.

Can you explain the pieces in plain words?

Expect to be asked what WACC is, or why you add depreciation back, or why the growth rate has to be small. These are not traps. They just check whether you understand what you are saying.

Do you know its weaknesses?

Strong candidates volunteer that a DCF is highly sensitive to assumptions, and that a small change in the discount rate moves the value a lot. Saying so is not admitting a flaw. It is showing judgment.

Common mistakes

Forgetting to discount the terminal value.

It sits at the end of your forecast, so it has to be brought back to today like everything else. Forgetting it makes your answer far too high, and it happens surprisingly often under pressure.

Using a growth rate that is too high.

Anything above about 3 percent needs a reason. Above 5 percent is very hard to defend, because it implies the business outgrows the entire economy forever.

Setting growth too close to WACC.

In the perpetuity formula you divide by WACC minus growth. As those two numbers get close together, that gap gets tiny and the value explodes toward infinity. If your model produces a wild number, this is almost always why.

Mixing up the cash flow and the discount rate.

Discounting cash flow that already accounts for interest at WACC double-counts the benefit of debt. This is the classic error.

Skipping the bridge to equity value.

Enterprise value is not what the shares are worth. If you are asked for a share price, you have to subtract debt and add cash.

Giving a single number.

Real DCF outputs are ranges shown in a sensitivity table. One confident number suggests a level of precision that does not exist.

Not knowing what your own terms mean.

If you say WACC, be ready to explain it. If you say unlevered free cash flow, be ready to say why "unlevered" matters. Using words you cannot unpack is worse than using simpler ones.