How do you answer "walk me through trading comps" in an interview?

Q: How do you answer "walk me through trading comps" in an interview?

Five steps, in order. Step 1: pick the peer set, public companies in the same business with similar size, growth, and margin. Step 2: spread their multiples, usually EV/EBITDA, EV/Revenue if profits are thin or negative, and P/E for the equity view. Step 3: take the peer median and apply it to your company's matching metric to get an implied enterprise or equity value. Step 4: bridge it home by subtracting net debt and dividing by diluted shares for an implied share price. Step 5: say where your company sits relative to the median and why.

Q: Which multiples pair with enterprise value and which pair with equity value?

Above the interest line it is enterprise value; below the interest line it must be equity value. Ask whose money it is. If it belongs to everybody, use enterprise value, which is why EV/Revenue, EV/EBIT, and EV/EBITDA work. If it belongs only to shareholders, use equity value, which is why P/E works. EV/net income breaks because net income already has debt holders paid out, and equity value/EBITDA breaks because EBITDA has not subtracted anything for debt yet.

Q: Why do two companies in the same industry trade at different multiples?

Three reasons account for most of the gap. Growth, because a dollar of earnings that is growing is worth more than a dollar standing still. Margin, because if more of every revenue dollar survives to profit, each dollar of sales is worth more. Risk, because a business with three customers has shakier earnings than one with thirty, and the shakier the earnings, the less anyone pays for a dollar of them. Leverage also matters in practice: enterprise multiples are supposed to shrug at financing, but a business drowning in debt trades at a discount anyway.

Q: How do you get from a trading comps multiple to an implied share price?

Apply the peer median to your company's matching metric first. Median EV/EBITDA times your EBITDA gives an implied enterprise value. From enterprise value, subtract net debt to get equity value, then divide by diluted shares outstanding for the implied share price. Use the fully diluted share count, not the basic count, or the equity value is wrong before you start.

Q: What errors do interviewers plant in a trading comps analysis?

Four show up more than any others. The target sitting inside its own comp set, which drags the median toward the company you are valuing. Basic share count used instead of fully diluted. Mixed time periods, where comps are pulled from different quarters or years. One-time items left in, such as a lawsuit or a one-off charge, which skew the set badly and should be stripped out because they will not repeat.

The five-step answer to "Walk me through trading comps" that works at every bank. The peer set, the median, the EV-to-share-price bridge, and the one step candidates skip.

OFFERGOBLIN·6 min read·updated August 2026

On this page

The prompt
What the question is really testing
The 60-second answer
Step 1: Pick the peer set
Step 2: Spread their multiples
Step 3: Take the peer median and apply it
Step 4: Bridge it home
Step 5: Say where you sit against the median, and why
Step five is where the answer is won
Follow-up traps
"Which of these multiples doesn't belong?"
"Would you include an apple farm in an orange farm's comp set?"
"What's wrong with this comp set?"
Execution notes

"A multiple is compressed information. The market's opinion of a business squeezed into one number."

The prompt

"Walk me through trading comps" is one of the valuation questions you get asked cold. It comes up constantly, because multiples underpin every deal conversation you will have, and a lot of interview questions run straight through them.

Valuation sounds like it should be deep math. Most of the time you are asking something simpler: what is this thing currently selling for?

Think about how a house gets priced. Nobody forecasts a house's future cash flows or breaks down the cost of its materials. They look at the neighborhood, look at what other houses sold for, and compare on a square foot basis. In M&A you look at a business the same way. It is a bundle of assets, liabilities, and equity, and the fastest route to a number is to compare it against businesses that are trading or have sold, on a unit basis that makes sense.

That is the appeal. Comps are fast, they are current, and they anchor on prices real buyers actually paid. A forecast bakes in your opinion about the future. A comp set bakes in the market's.

Comps are one of three valuation methods. Trading comps look at public companies. Precedent transactions look at businesses that have already sold. A DCF values the cash flows directly and gives you a number instead of a multiple. In real work you run all three and triangulate. See valuation for how the three fit together.

What the question is really testing

A company trades at 12x earnings. Good deal or ripoff?

Standing alone, that number tells you nothing. Four hundred dollars a square foot is the deal of a lifetime in one neighborhood and robbery in another. Same number, different neighborhoods, opposite verdict. Cheap and expensive are always relative, and the comp set is what supplies the relation.

So the interviewer is checking three things at once. Whether you can defend a peer set instead of just naming one. Whether you know which price tag pairs with which earnings line. And whether you can explain your own number once you have it.

One thing to accept before the script: there is no clean answer in comps. Peer selection is judgment, and there is no objectively correct set waiting to be found. What gets checked is whether you can defend the set you picked.

The 60-second answer

Five steps. Deliver them in order, then stop talking.

Step 1: Pick the peer set

"Public companies in the same business, with similar size, growth, and margin."

That sentence carries more than it looks. Industry, business model, size, and geography are the four dimensions you defend on, and you should be ready to name them if pushed.

Step 2: Spread their multiples

A multiple is a fraction. The numerator is a price tag, either enterprise value or equity value. The denominator is a type of earnings. What the fraction tells you is the price to purchase one dollar of that earnings.

You could invent an infinite number of multiples. Three carry most of the load.

Multiple	Price tag	Reach for it when
EV/EBITDA	Enterprise value	Default. EBITDA sits before financing and depreciation choices, so it compares two companies no matter how they are funded
EV/Revenue	Enterprise value	Profits are thin or negative and there is nothing to divide by. It prices the sales and takes the margin on faith
P/E	Equity value	You want the shareholder view: market cap over net income

The denominator does not have to be earnings at all. Data centers get valued per gigawatt, real estate per square foot. Other multiples covers the sector-specific ones. Every multiple still answers the same question: how much does one unit of the thing in the denominator cost?

Step 3: Take the peer median and apply it

Apply the median to your company's matching metric. Median EV/EBITDA times your EBITDA gives you an implied enterprise value. Median P/E times your net income gives you an implied equity value. Which one you land on depends entirely on which multiple you used.

Step 4: Bridge it home

If you are sitting at enterprise value, subtract net debt to get equity value. Then divide by diluted shares outstanding for an implied share price.

Use the diluted count. The equity value is not right without it.

Step 5: Say where you sit against the median, and why

"We trade at a premium to the peer median on growth and margin, and the discount case is customer concentration and leverage."

That is the sentence. Now here is why it matters more than the four steps in front of it.

Step five is where the answer is won

Two companies. Same industry, same size. One trades at 13x EBITDA, the other at 9x. Why?

When two multiples differ, the market's opinion of those two businesses differs. Three reasons account for most of the gap.

Growth. A dollar of earnings that is growing is worth more than a dollar standing still. You are buying the whole future stream at today's price.

Margin. If more of every revenue dollar survives to profit, each dollar of sales is worth more. Better engine, higher price per unit. See margins.

Risk. One business has three customers. The other has thirty. Losing one customer means something very different in each case. The shakier the earnings, the less anyone pays for a dollar of them.

Leverage belongs on the list too, with an asterisk. Enterprise multiples are supposed to shrug at financing. A business drowning in debt trades at a discount anyway.

Growth and margin push you above the median. Risk and leverage pull you below it. Interviewers frequently hand you two multiples with no context and ask you to hypothesize why they differ, and this framework is what lets you speak intelligently instead of guessing.

Step five is the difference between doing the math and doing your job.

Follow-up traps

"Which of these multiples doesn't belong?"

The most common way candidates fail this question is pairing the wrong price tag with the wrong line item.

Ask whose money it is. If the money belongs to everybody, it is enterprise value, which is why it pairs with the lines debt has not touched yet: revenue, EBIT, and EBITDA. If the money belongs only to shareholders, it is equity value, and it pairs with what is left after debt has been paid.

EV/net income does not work. By the time you reach net income you have already subtracted what belongs to debt holders, so net income is shareholders-only. Two identical businesses with different debt loads would print wildly different numbers.

Equity value/EBITDA fails for the mirror reason. Equity value represents shareholder ownership, and EBITDA has not yet subtracted anything for debt.

The rule in one line: above the interest line it is enterprise value, below the interest line it must be equity value. Match the price tag to the people the money belongs to.

"Would you include an apple farm in an orange farm's comp set?"

Answering yes or no is the trap. Both are wrong.

Give the consulting answer, "it depends," and then back it up, because backing it up is the entire test. It depends on whether you have enough other comps. If you do, throw the apple farm out. If you do not, an apple farm may be the closest thing to an orange farm you are going to find. Same industry, same business model, potentially comparable size, potentially the same part of the world.

Then land the point: comparables feed a judgment call, and all you can ever claim is that this is the best data available to produce the best possible estimate.

"What's wrong with this comp set?"

Some superdays hand you a finished comp analysis and ask you to find the errors. Four get planted more than any others.

The target sitting in its own comp set. It drags the median toward the company you are trying to value. The company being valued never belongs in its own peer group.
Basic share count instead of fully diluted. The equity value is wrong before you start. Use the diluted count.
Mixed time periods. Pick a quarter or a year and make every comp fall inside it.
One-time items left in. A lawsuit, a disaster, a one-off charge. Any of them skew the set badly. Strip out what will not repeat.

There is no clever shortcut for this one. Finding errors in a comp set requires knowing what a proper comp set looks like and then checking the details one at a time.

Execution notes

Memorize the five steps in order. Peer set, multiples, median, bridge, positioning.

Keep the pairing rule as one sentence you can produce under pressure: above the interest line, enterprise value; below it, equity value.

Never call a multiple cheap or expensive without naming the set it is cheap or expensive against.

Rehearse the bridge out loud. Enterprise value minus net debt, divided by diluted shares.

The DCF walkthrough is the next script worth having ready, and walk me through an LBO after that.

create your free account to keep reading

12 more sections. free forever. takes 30 seconds.

By continuing, you accept our Privacy Policy.

Frequently Asked Questions

How do you answer "walk me through trading comps" in an interview?

Which multiples pair with enterprise value and which pair with equity value?

Why do two companies in the same industry trade at different multiples?

How do you get from a trading comps multiple to an implied share price?

What errors do interviewers plant in a trading comps analysis?