Comparable Company Selection: How to Pick a Peer Set
Comparable company selection screens peers by industry, size, growth, and margins. Get the five criteria, the process, and a worked peer-set example.
Updated Jul 2, 2026 / 8 min read
On this page
Comparable company selection is the process of screening public companies down to a peer group that is genuinely similar to the target on industry, size, growth, and profitability, so the multiples you spread actually mean something. Per ibinterviewquestions.com, a workable peer group runs 5 to 12 companies, drawn from an initial list of 15 to 20 candidates that gets filtered on industry, size, growth, and margins. Get the screen wrong and every multiple downstream is wrong too, since a median built from mismatched peers is not a benchmark, it is noise. This is the step analysts at Corporate Finance Institute call "probably the hardest (or most subjective)" part of a comparable company analysis.
TL;DR
- A workable peer group is roughly 5 to 12 companies, per ibinterviewquestions.com; most banks want a minimum of 5 for a credible median.
- Screen on five dimensions: industry and business model, size, growth, margins, and geography.
- A practical size filter is 0.3x to 3x the target's enterprise value or revenue, per ibinterviewquestions.com.
- Street of Walls flags that a "perfect" comparable almost never exists, so judgment trades off breadth against fit.
- The process runs wide-to-narrow: 15 to 20 candidates, filter to core criteria, then document why each name stayed or got cut.
What is comparable company selection?
Comparable company selection is the screening step that comes before you spread any multiples in a comparable company analysis. You are not looking for a single perfect match, since Street of Walls notes you will "almost never find a perfect comparable company." You are building a defensible set of public peers whose businesses, size, growth, and margins are close enough to the target that their multiples are a fair read on how the market prices similar risk and cash flow. Get this step right and the median multiple you calculate later actually means something. Get it wrong and you can compute a precise multiple off an imprecise peer set, which is the classic interview trap: a clean-looking number built on a bad screen.
What are the five criteria for selecting comparable companies?
The peer screen runs on five dimensions, and per Corporate Finance Institute the goal is always "the closer the match, the better." Industry and business model come first: companies competing in the same markets with genuinely similar products, not just a shared classification code. Size, growth, margins, and geography follow, each screening out companies that look similar on paper but carry a different risk or growth profile.
| Criterion | What it filters | Why it matters |
|---|---|---|
| Industry and business model | Product/service mix, customer base, competitive set | Same demand drivers and competitive dynamics, per Street of Walls |
| Size | Revenue, enterprise value, market cap | Bigger companies trade at different multiples than smaller ones in the same space |
| Growth | Revenue and earnings growth trajectory | Faster-growing peers command premium multiples |
| Margins | Gross margin, EBITDA margin, capital intensity | Higher-margin businesses are usually valued richer |
| Geography | Primary markets, regulatory environment | Growth rates, regulation, and currency exposure differ by region |
How do you screen peers by size?
Size screening starts from the target's enterprise value or revenue and filters candidates to a range around it, since a company ten times the target's size usually trades on a different multiple even in the same industry. Per ibinterviewquestions.com, a practical guideline is peers within roughly 0.3x to 3x the target's enterprise value or revenue. That means a target with a 5 billion dollar enterprise value would look at peers roughly in the 1.5 billion to 15 billion dollar range. The guide adds that this cutoff is not absolute: if a niche industry has few public comparables, you widen the size band rather than shrink the peer group below a usable count. The same source frames size as one of five filters, alongside industry, growth, margins, and geography, applied in sequence rather than all at once.
What does the peer-selection process look like end to end?
The process moves from a wide candidate list to a documented final set, filtering on the criteria above in order rather than screening on everything simultaneously. Per ibinterviewquestions.com, the flow is:
- Cast a wide net. Start with 15 to 20 candidates in the target's broad industry, sourced from competitor filings, equity research, and databases like CapIQ or Bloomberg (Street of Walls notes the SEC filings "Competitors" section and Bloomberg's Supply Chain page as two concrete sources).
- Apply primary filters. Cut on industry classification and the size range from the section above.
- Evaluate growth and profitability. Compare revenue growth trajectories and margin profiles, not just the current-year snapshot.
- Research company-specific factors. Read each candidate's business description to catch names that pass the numeric screen but do not actually compete in the same markets.
- Finalize and document. Land on 5 to 12 core peers, with a written rationale for each inclusion and exclusion. ibinterviewquestions.com also flags keeping 3 to 5 "secondary comparables" on the side for reference, shown separately rather than folded into the core median.
Worked example: screening a peer set
Say you are valuing a mid-size enterprise software company with 2 billion dollars in enterprise value, 400 million dollars in revenue, and 30 percent EBITDA margins. You start with 18 candidates pulled from the target's SEC "Competitors" disclosure and analyst coverage. Applying the industry filter drops 4 names that classify as enterprise software but actually sell hardware, leaving 14. Applying the size filter of roughly 0.3x to 3x enterprise value, meaning about 600 million to 6 billion dollars, cuts another 5 whose enterprise values run well outside that band, leaving 9. A growth-and-margin pass removes 2 more: one growing revenue at 3 percent versus the target's 18 percent, and one running 12 percent EBITDA margins versus the target's 30 percent. That leaves 7 core peers, inside the 5 to 12 range, with the 2 borderline names you cut on growth kept as secondary comparables for reference rather than discarded outright. That 7-name set, not the original 18, is what you spread EV/EBITDA and EV/Revenue multiples across.
How does peer selection affect the multiple you calculate?
Peer selection determines the median before you ever calculate it, since a loose screen pulls in outliers that drag the median away from a true read on the target's fair value. Per Wall Street Prep, the median is the standard benchmark precisely because it "removes outlier distortion," but that only works if the underlying peer set is genuinely comparable. A median calculated from 8 tightly screened peers is a meaningful signal. A median calculated from 8 loosely related companies, where 2 are twice the target's size and 1 operates in an adjacent industry, is precise-looking noise. This is why the screening step matters more than the multiple math: the arithmetic behind spreading and averaging multiples is mechanical, but deciding who belongs in the set is the judgment call that determines whether the output means anything. See comparable company analysis for how the multiple is applied once the peer set is locked, and enterprise value vs equity value for the metric base the size screen uses.
Frequently Asked Questions
How many comparable companies should be in a peer group?
Roughly 5 to 12, per ibinterviewquestions.com. Most banks require a minimum of 5 for the median to carry statistical weight, and pushing past 15 loosely related names starts diluting the set with noise rather than adding signal. Street of Walls notes that with fewer than 5 peers and no clear outliers, some analysts use the mean instead of the median.
What is the size range used to screen comparable companies?
A common guideline is roughly 0.3x to 3x the target's enterprise value or revenue, per ibinterviewquestions.com. A target with a 5 billion dollar enterprise value would screen peers in roughly the 1.5 billion to 15 billion dollar range. The cutoff is not rigid: in a niche industry with few public comparables, analysts widen the band rather than shrink the peer group below a usable size.
Does industry classification alone determine a good comparable?
No. Per ibinterviewquestions.com, the critical question is whether companies "compete in the same markets with similar products," not whether they share a GICS or SIC code. Two companies can carry the same industry classification and still differ enough on business model, customer base, or margin structure to be poor comparables.
What is a "pure-play" comparable and how often do you find one?
A pure-play comparable is a company whose business is almost entirely focused on the same product or market as the target. Per Street of Walls, you will "almost never find a perfect comparable company," so analysts accept imperfect matches and adjust their read of the multiples for the differences that remain, rather than waiting for an exact match that does not exist.
Where do analysts source the initial list of comparable-company candidates?
Common sources include the target's own SEC filings, specifically the "Competitors" section, sell-side equity research reports, and financial databases like Bloomberg and CapIQ. Street of Walls also flags Bloomberg's Supply Chain function as a way to surface a target's customers, suppliers, and competitors directly from its ticker.
Should you keep companies that fail the size or growth screen?
Not in the core peer set, but you do not have to discard them entirely. ibinterviewquestions.com recommends keeping 3 to 5 borderline names as secondary comparables, shown separately from the core 5 to 12 group, so they remain a reference point without diluting the primary median.
Sources
- ibinterviewquestions.com, "Selecting the Peer Group: Criteria That Actually Matter": https://ibinterviewquestions.com/guides/valuation-investment-banking/selecting-the-peer-group-criteria-that-matter (checked July 2026)
- Street of Walls, "Comparable Company Analysis": https://www.streetofwalls.com/finance-training-courses/investment-banking-technical-training/comparable-company-analysis/ (checked July 2026)
- Corporate Finance Institute, "Comparable Company Analysis": https://corporatefinanceinstitute.com/resources/valuation/comparable-company-analysis/ (checked July 2026)
- Wall Street Prep, "Comparable Company Analysis (Trading Comps)": https://www.wallstreetprep.com/knowledge/comparable-company-analysis-comps/ (checked July 2026)