How to Answer “How do you pick comparable companies for a trading comps analysis?” in Investment Banking Interviews

Start by describing the target the way public markets would: primary products/services, where value is created in the value chain, customer/end-market mix, and the revenue model (recurring vs transactional, usage-based, subscription, etc.). In trading comps analysis, comparability is mainly about similar drivers of revenue, EBITDA, and risk, not just sharing a broad sector label.

Then name 3–5 “anchor” drivers you’ll use to judge peers: subsector and value chain position, growth profile, margin structure/operating leverage, capital intensity and working capital needs, cyclicality/commodity exposure, and regulatory intensity. If the target is multi-segment, state upfront you’ll focus on peers where the relevant segment is the majority, or you’ll use segment disclosures to approximate a pure-play lens.

This framing shows you’re building investment banking comparable companies from first principles, not from memorised tickers.

Next, explain how you create a long list: screen public companies in the right subsector/keywords and then apply hard filters that make the multiples comparable and usable. Typical filters include geography/primary listing (similar investor base and macro), revenue or EBITDA scale (avoid tiny microcaps if the target is mid/large-cap), and liquidity/research coverage (so trading multiples aren’t stale or distorted).

Add practical analyst constraints: prefer companies with clean filings, consistent fiscal periods, and sufficient disclosure to calculate consistent LTM and (if available) NTM metrics. Be cautious with names undergoing major M&A, restructurings, or accounting changes unless you can normalise to a run-rate.

End with a tangible output: a long list of roughly 10–20 candidates to refine. That “funnel” is what interviewers expect in investment banking interview prep.

Now do the “economics pass” to select the real peers. Compare how candidates make money and what explains their valuation levels:

• Revenue drivers: end markets, customer type, pricing power, contract length, retention/churn, and mix (product vs services).
• Profitability and cost structure: gross margin profile, EBITDA margins, fixed vs variable cost base, operating leverage.
• Growth and reinvestment: historical and expected growth, capex and R&D intensity, working capital swings.
• Risk: cyclicality, customer concentration, FX exposure, regulatory constraints, and balance-sheet flexibility.

State your selection philosophy: prioritise “pure-play” peers even if they’re not perfect on size, but keep a few larger diversified players as context comps if investors genuinely benchmark the space that way. This is a core best practice for trading comps analysis and shows judgement beyond a simple screen.

Before finalising, do a quick multiple sanity-check (EV/Revenue, EV/EBITDA, and sector KPIs) to see whether differences are explainable by growth, margins, or risk. If one company trades far above/below the pack, don’t reflexively remove it—explain the driver: different business mix, accounting (e.g., capitalised costs), leverage tolerance, cyclicality, or a one-off event.

Then describe basic data hygiene: ensure consistent definitions (LTM vs NTM), consistent EBITDA adjustments where possible, and consistent treatment of leases, SBC, or exceptional items per your team’s convention. Use EV-based multiples for operating comparability and explicitly note when capital structure is a meaningful differentiator.

This “clean + interpret” step signals you can do trading comps analysis step by step, not just pick names.

Close with how you would deliver the output: a tight set of 6–10 core Comparable Companies, plus 2–4 context/adjacent names to triangulate the range. For each core peer, provide a one-line rationale tied to the anchor drivers from Step 1 (subsector, end markets, margins/capital intensity, growth, geography).

Also mention exclusions explicitly—e.g., “excluded due to predominantly emerging-market exposure,” “excluded due to materially different revenue model,” or “excluded because post-merger numbers aren’t clean.” That documentation is what makes the selection defensible in an Investment Banking setting.

If asked to go further, you can add that you’d weight interpretation toward the most comparable names rather than taking a simple average across all peers.