Valuation

Trading comparables

4 min read · updated June 29, 2026

Trading comps answer a market question, not a fundamental one: what is the market paying for similar companies right now, and what does that imply this one is worth? Where a DCF values a business on its own cash flows, comps value it by analogy — you find a peer set, see what multiple they trade at, and apply it. It's fast, it's grounded in real prices, and it's the first thing an analyst pulls together.

The logic is simple. If five companies in the same business trade at roughly 10x EBITDA, and your target does $200 of EBITDA, the market is implicitly telling you it's worth about **$ 2,000** of enterprise value. The whole craft is in choosing a defensible peer set and the right multiple.

Step 1 — Pick the set

A comp is only as good as its comparability. You want public companies that look like the target on the dimensions that drive value:

Industry / business model — same sector, similar products and end-markets.
Size — revenue and market cap in the same ballpark; a $50bn leader doesn't trade like a$ 500m challenger.
Growth and margins — the market pays up for faster growth and fatter margins.
Geography — where they earn the money, since that drives risk and tax.

Common mistake

Throwing every name in the sector into the set to make it bigger. A loose peer set produces a wide, meaningless multiple range. Five genuinely comparable companies beat fifteen loose ones — and you should be ready to defend why each name is in there, because a good interviewer will ask you to justify the set or kick one out.

Step 2 — Pick the multiple

A multiple is just value divided by a financial metric, and the two halves have to match. An enterprise-value multiple goes over a pre-interest metric (because EV belongs to all capital providers); an equity-value multiple goes over an after-interest metric (because equity is what's left for shareholders).

EV / EBITDA — the workhorse. Enterprise value over operating cash earnings.
EV / Revenue — for early-stage or unprofitable companies with no meaningful EBITDA yet.
P / E — price (equity value) over net income; equity-side, so capital-structure dependent.

Key insight

EV/EBITDA is the default because it's capital-structure and (largely) accounting-policy neutral. EBITDA sits above interest, so it isn't distorted by how a company is financed, and adding back D&A strips out depreciation-policy differences. That lets you compare a debt-heavy company to a debt-free one on an apples-to-apples basis — which is exactly what P/E can't do, because net income is hit by both interest expense and tax structure.

Step 3 — Apply it

You don't pick one number — you take the range. Line up the peers, look at where they cluster (the median and the interquartile range are more robust than the mean, which one outlier can drag around), and apply that range to the target's metric.

Say comps trade at a median of 9x to 11x forward EBITDA and your target does $200** of EBITDA. That implies an enterprise value of roughly **$ 1,800 to $2,200. Bridge by net debt and you've got an equity value range. Notice you get a range, not a point estimate — comps frame the market's view, they don't pretend to precision.

Trading vs. transaction multiples

One thing to keep straight: comps use current trading prices, so they reflect what minority shareholders pay for small stakes day to day — no control changing hands. That's why comps typically come in below precedent transactions, which embed a control premium. More on that in the precedent transactions article.

Interview tip

Be able to say the sequence cleanly: pick a tight peer set → choose the right multiple → apply the range to the target's metric → bridge to equity. And always pair EV multiples with pre-interest metrics and equity multiples with after-interest metrics — mismatching them (P/EBITDA, say) is an instant tell that you don't know which value the multiple describes.

Make it stick

Test yourself on this

→