What is comparable company analysis and how do analysts build a comps table?

Absolute vs relative valuation: why the distinction matters

There are two broad approaches to valuing a company. Absolute valuation — like a discounted cash flow (DCF) — asks: what are this company’s future cash flows worth today? The answer depends on your assumptions about growth, margins, and the discount rate. Change those assumptions by a few percentage points and the output swings dramatically.

Relative valuation asks a different question: what are investors currently paying for similar businesses? It anchors the answer to market reality. If every listed software company with 20% revenue growth trades between 8x and 12x EBITDA, a buyer paying 15x for an equivalent business is paying a significant premium — and needs a strong reason to justify it.

Neither approach is superior on its own. Bankers almost always run both and present a “football field” — a chart showing the valuation range from each method side by side. The overlap, if any, is where a deal gets done.

How to build a peer group

The quality of a comps analysis lives or dies by peer selection. The market prices each company based on its specific characteristics — growth rate, margins, competitive position, capital intensity. A peer group of genuinely similar businesses produces a meaningful valuation range. A peer group built from vague industry matches produces noise.

Analysts typically screen on four dimensions:

• Industry and sub-industry: Same sector is not enough. A diversified industrial conglomerate is not a good peer for a pure-play precision manufacturer, even if both sit under “industrials.”
• Size: Revenue, EBITDA, and market cap. A $500m company rarely trades at the same multiple as a $20bn company in the same sector — scale commands a premium.
• Business model: Product mix, revenue type (recurring vs project-based), margin structure, and customer concentration all affect how the market prices a business.
• Geography: US-listed companies often trade at different multiples than European equivalents due to investor base, liquidity, and home-market growth rates.

The core valuation multiples

Once you have your peer group, you calculate valuation multiples for each company. A multiple divides a measure of value (the numerator) by a measure of performance (the denominator). The most important ones in practice:

EV/EBITDA — enterprise value divided by earnings before interest, tax, depreciation and amortisation. The workhorse of M&A. Capital-structure neutral, which means you can compare companies with different debt levels without distortion. Typically the headline multiple in any deal.

EV/Revenue — enterprise value divided by revenue. Used when EBITDA is negative or very small — common in early-stage, high-growth, or turnaround situations where profitability lags revenue growth. A software company losing money on the path to scale might be valued at 5–8x revenue.

P/E (Price-to-Earnings) — share price divided by earnings per share. The equity multiple. Affected by capital structure (interest payments reduce earnings), tax rate, and accounting choices. Most useful for mature, stable, lightly-leveraged businesses.

EV/EBIT — enterprise value divided by earnings before interest and tax. Like EV/EBITDA, but includes depreciation. Better for capital-intensive businesses where depreciation is a real economic cost, not just an accounting entry.

Reading a comps table: a worked example

A comps table puts all the data in one place. Analysts calculate the enterprise value, key financial metrics, and multiples for each peer, then identify the median and range.

Applying the median EV/Revenue of 5.2x to the target’s $7.2bn revenue gives an implied EV of $37.4bn. Applying the median EV/EBITDA of 19.0x to $2.1bn EBITDA gives $39.9bn. The two methods bracket a range of roughly $37–40bn — and that range informs where a deal might get priced.

“Comparables are not a truth machine — they are a market-calibration tool. They tell you what investors are currently willing to pay, not what a business is intrinsically worth.”

The three main limitations

Markets can be wrong — and comps will follow them. If the whole sector is overvalued because of a bubble, a comps analysis anchored to those valuations will look fine right up until the bubble bursts. Comps are inherently backward-looking and sentiment-dependent.

No two companies are truly identical. A faster-growing peer deserves a higher multiple; a peer with a stronger balance sheet or better management also earns a premium. The median multiple obscures these differences. Experienced analysts apply a premium or discount to the median based on the target’s relative profile — but that judgment call reintroduces subjectivity.

Public market data is not always available for private companies. Comps work best when both the target and its peers are publicly listed. For private company transactions, you often have limited financial data on the peers — which is one reason precedent transactions analysis (which uses deal-level data) is used alongside or instead of trading comps.

What this means for you

If you are studying for a career in investment banking, equity research, or private equity, comparable company analysis is the first valuation skill you will be tested on. Interviewers routinely ask: “walk me through a comps analysis” — and they expect you to cover peer selection, multiple choice, how you calculate EV, and the limitations, in around two minutes.

If you are following a company as an investor, watching where it trades relative to its peer group on EV/EBITDA is a useful reality check. A company trading at a persistent discount to peers either deserves it (lower growth, weaker margins) or is an opportunity — and understanding which requires digging deeper than the multiple itself.