Comparable Company Analysis
Comparable company analysis, often called "comps" or "trading comps," values a business by examining how the market prices similar companies. The logic is intuitive: if five software companies with similar growth rates and margins trade at 25 times earnings, a sixth company with comparable characteristics should trade at roughly the same multiple. The method does not require projecting cash flows a decade into the future or estimating terminal growth rates. It reflects what buyers and sellers in the public market are actually willing to pay right now.
Wall Street banks use comps as a primary valuation tool in virtually every equity research report, M&A advisory engagement, and IPO pricing exercise. Morgan Stanley, JPMorgan, and every other bulge bracket firm maintain comp sheets for every sector they cover. The approach is fast, grounded in observable market data, and easy to communicate to clients. Its weakness is equally straightforward: if the market is systematically overpricing an entire sector, every company in the peer group looks "fairly valued" relative to its peers, even if all of them are expensive on an absolute basis.
Selecting the Peer Group
The quality of a comparable company analysis depends entirely on the peer group. Choose companies that are too different, and the comparison breaks down. The ideal peer group shares the following characteristics with the target company:
Industry and business model. A cloud software company should be compared to other cloud software companies, not to hardware manufacturers or IT consultants. Salesforce's natural peers include ServiceNow, Workday, and HubSpot, not Dell or Accenture.
Size. Market capitalization matters because larger companies typically command lower growth rates but higher multiples due to perceived stability and liquidity. Comparing a $5 billion mid-cap to a $500 billion mega-cap introduces distortions. If the target is a $20 billion specialty retailer, the peer group should include companies in the $10-40 billion range.
Growth profile. A company growing revenue at 30% annually should not be compared to one growing at 5%, even if both operate in the same industry. Growth is the single largest driver of valuation multiples, and mixing high-growth and low-growth peers produces meaningless averages.
Profitability. Margin structure matters. A software company with 80% gross margins and 30% operating margins operates a fundamentally different economic model than one with 50% gross margins and 10% operating margins. Similar margin profiles suggest similar cost structures, pricing power, and competitive positions.
Geographic mix. A company generating 90% of revenue in the United States faces different risks and opportunities than one with a 50/50 domestic-international split. Currency exposure, regulatory environments, and end-market dynamics all differ.
A typical peer group contains 5-15 companies. Fewer than five provides too little data for meaningful comparison. More than fifteen dilutes the analysis with increasingly dissimilar businesses.
Choosing the Right Multiples
Not all valuation multiples work for all companies. The right multiple depends on the company's sector, maturity, and capital structure.
Enterprise Value Multiples
EV/EBITDA is the most widely used multiple in comparable company analysis. Enterprise value (market cap plus net debt) divided by earnings before interest, taxes, depreciation, and amortization strips out the effects of capital structure, tax jurisdiction, and depreciation policy. This makes it ideal for comparing companies with different debt loads or in different countries. As of early 2025, the median S&P 500 company traded at approximately 14x EV/EBITDA. Mature industrials might trade at 8-10x, while high-growth software companies command 25-40x.
EV/Revenue is used for companies with negative or highly volatile earnings, which is common among early-stage growth companies. When DoorDash went public in 2020, it was unprofitable, so analysts valued it on revenue multiples rather than earnings. The drawback is that revenue multiples ignore profitability entirely, treating a company with 80% gross margins the same as one with 20% gross margins.
EV/EBIT is similar to EV/EBITDA but includes depreciation and amortization as real costs. This is more conservative and more appropriate for capital-intensive businesses where depreciation represents actual economic wear on physical assets, such as airlines, manufacturing companies, and utilities.
Equity Multiples
Price-to-Earnings (P/E) is the most commonly cited equity multiple. It divides the share price by earnings per share. Forward P/E (using next year's estimated earnings) is generally preferred over trailing P/E because investors pay for future earnings, not past ones. The S&P 500's long-term average forward P/E is approximately 16-17x, though it has frequently traded above 20x during periods of low interest rates.
Price-to-Book (P/B) compares the market value of equity to its book value. This is particularly useful for financial companies like banks and insurance companies, whose balance sheets consist largely of financial assets carried at or near fair value. JPMorgan Chase trading at 1.8x book value means the market values each dollar of the bank's equity at $1.80.
PEG Ratio divides the P/E ratio by the expected earnings growth rate. A company trading at 30x earnings with 30% expected growth has a PEG of 1.0x. Peter Lynch popularized this ratio as a way to identify growth companies trading at reasonable prices. A PEG below 1.0 suggests the market is undervaluing the company's growth; above 2.0 suggests it may be overvaluing it.
Building the Comp Table
A standard comp table includes the following for each peer company:
- Company name and ticker
- Market capitalization
- Enterprise value
- Last twelve months (LTM) and next twelve months (NTM) revenue
- LTM and NTM EBITDA
- LTM and NTM earnings per share
- Revenue growth rate
- EBITDA margin
- Key valuation multiples (EV/Revenue, EV/EBITDA, P/E)
From this data, calculate the mean, median, 25th percentile, and 75th percentile for each multiple. The median is generally more reliable than the mean because it is less influenced by outliers. If one company in a group of eight trades at 50x EBITDA due to a pending acquisition, the mean will be inflated, but the median will not.
Applying the Multiples
Once the peer group multiples are established, apply them to the target company's financial metrics to derive an implied valuation range.
For example, if the peer group median EV/EBITDA is 12x and the target company has EBITDA of $2 billion, the implied enterprise value is $24 billion. Subtract net debt of $3 billion, and the implied equity value is $21 billion. Divide by 500 million diluted shares, and the implied share price is $42.
Repeat this process with multiple valuation metrics (EV/EBITDA, EV/Revenue, P/E) and use the range of implied values to establish a valuation range rather than a single point estimate. If EV/EBITDA implies $38-46 per share, EV/Revenue implies $35-42, and P/E implies $40-48, the overlapping range of $40-42 represents the area of highest confidence.
Adjustments and Refinements
Raw comps rarely tell the full story. Several adjustments improve accuracy:
Calendarization. Companies with different fiscal year-ends need to be adjusted to a common period. If one peer has a January fiscal year-end and another has a June fiscal year-end, their trailing twelve-month figures represent different economic periods. Calendarize by prorating quarterly data to align all companies to the same twelve-month window.
Non-recurring items. Strip out one-time charges, gains on asset sales, restructuring costs, and litigation settlements. These items distort underlying earnings power. When General Electric reported billions in goodwill impairments in 2018, its reported earnings were meaningless for comp purposes. The adjusted figure, excluding the impairment, was far more representative.
Stock-based compensation. This is contentious. Some analysts add back stock-based compensation to EBITDA, arguing it is non-cash. Others include it as a real expense, arguing it dilutes shareholders. The right approach depends on the peer group convention, but consistency matters more than which method is chosen. If stock-based comp is added back for the target, it must be added back for all peers.
Minority interests and associates. If a company consolidates a subsidiary it does not fully own, enterprise value should include the minority interest. Conversely, if a company holds a significant equity stake in another business, that stake may need to be valued separately.
Interpreting the Results
A company that trades at a discount to its peer group is not automatically undervalued. There is usually a reason. It might have slower growth, lower margins, higher leverage, weaker management, pending litigation, or greater regulatory risk. The analyst's job is to determine whether the discount is justified or whether the market is mispricing the stock.
Consider Coca-Cola and PepsiCo. Both are large-cap consumer staples companies, but PepsiCo has historically traded at a slight discount on some metrics because its snack food business has lower margins than a pure beverage operation. That discount reflects a real economic difference, not a market error.
Conversely, a company trading at a premium to its peers may deserve that premium if it has superior growth, better margins, a stronger competitive moat, or a more favorable market position. Apple has traded at a premium to hardware peers for over a decade because its ecosystem, brand loyalty, and services revenue justify a higher multiple.
Strengths and Limitations
Comparable company analysis is fast, market-based, and easy to explain. It reflects the collective wisdom (or folly) of thousands of market participants. It works well for companies with a clear peer group and when the overall market is not at an extreme.
The limitations are real. Comps are inherently relative: they can tell an analyst whether a stock is cheap or expensive relative to its peers, but not whether it is cheap or expensive in absolute terms. During the dot-com bubble, internet companies traded at extraordinary multiples relative to any historical standard, but they all looked "reasonable" relative to each other. During the 2008 financial crisis, bank stocks all looked cheap on P/E and P/B, but the earnings and book values underlying those multiples were about to collapse.
Comps also struggle with unique companies that have no true peers. When Tesla was the only publicly traded pure-play electric vehicle manufacturer, analysts debated whether to compare it to legacy automakers (low multiples, high volume), tech companies (high multiples, platform model), or energy companies (commodity exposure, regulatory sensitivity). The peer group choice drove massive differences in implied valuation.
The best use of comparable company analysis is as one input in a broader valuation framework. Pair it with a DCF for intrinsic value and a precedent transaction analysis for M&A context. When all three approaches converge on a similar range, the analyst can invest with greater conviction.
Put these principles into practice. Track fundamentals, build portfolios, and analyze stocks with AI-powered insights.
Start Free on GridOasis →