Match the economic engine
Industry labels are too broad. Compare how companies acquire customers, price the product, recognize revenue, fund growth, and retain demand. Two software firms can deserve different peer groups when one sells long contracts through enterprise teams and the other depends on self-service usage. Similar revenue does not guarantee similar economics.
Control for maturity and capital intensity
Growth rate, margin structure, reinvestment needs, cyclicality, geography, and balance-sheet risk shape the multiple investors can rationally pay. Build a core group with the closest economics and a secondary reference group that illuminates one dimension. Do not average them all simply because a data provider lists them together.
- Explain why every peer is included.
- Show the operating difference that limits comparability.
- Use medians and ranges, then test the conclusion against fundamentals.
Treat the multiple as an output
A peer multiple is evidence about market expectations, not an intrinsic answer. Ask which growth, margin, and risk profile is embedded in the subject company's valuation and whether its operating record supports that position. A clean comparison makes disagreement visible instead of hiding it inside an average.
Test the peer set with a scorecard
Imagine a software company growing 20%, earning a 15% operating margin, and generating 80% recurring revenue. A mature peer growing 8% at a 35% margin may share an industry code but is a weak anchor for growth expectations. A company with similar growth but significant hardware inventory and factory spending has different cash economics. Growth, margin, recurrence, and capital intensity reveal more than one label.
Give each candidate a simple 0–2 score, but do not turn the average into an automatic valuation. The scorecard exists to reveal why a premium or discount may be justified. Separating three to five core peers from reference peers used for one specific dimension also prevents one extreme observation from controlling the answer.
Finally, write why the subject deserves to trade above or below the median. Higher growth, a longer runway, or lower customer concentration can be tested. Without a testable reason, a premium may be popularity; a discount may hide leverage, dilution, or cyclicality rather than opportunity.
- Compare customers, billing units, contract duration, and distribution channels.
- Align growth and margins to the same period and adjustment policy.
- Prefer medians and quartile ranges to a simple mean.
- Record one inclusion reason and one limiting difference for every peer.
