Comparing Companies in the Same Sector

Why absolute numbers are meaningless without context

A $5B revenue company sounds large — until you realize its competitors generate $50B each. A 15% operating margin sounds impressive — until the sector average is 28%. A 20x P/E sounds expensive — until peers trade at 35x. Absolute numbers in isolation tell you almost nothing. Context from peers reveals relative strength or weakness.

This is why professional analysts always build a "comps table" — a comparison of a company against its closest peers on a standardized set of metrics. The comps table instantly reveals who is expensive, who is cheap, who has the best margins, who is growing fastest, and where the business risks lie.

Picking the right comparable companies

A bad peer set produces bad analysis. The right comps must share the same essential business characteristics: similar revenue model (subscription vs. transactional vs. product), similar end market, similar geography, and similar competitive dynamics. Comparing Microsoft to Meta because both are "tech" is lazy — their business models, customers, and competitive forces are very different.

General rules for picking comparable companies:

• Same primary revenue model (SaaS vs. hardware vs. marketplace)
• Similar size (within 0.5x-2x of revenue)
• Same geographic focus where possible
• Same regulatory environment

Industry benchmarks: what's normal for this sector?

Every industry has its own normal ranges for key metrics. What's "good" in one sector is "terrible" in another:

Which metrics belong in a peer comparison table?

A well-structured comps table covers four dimensions: Valuation, Profitability, Growth, and Financial Health. Each dimension tells a different story about the company's competitive position.

How to normalize for fair comparison

Raw numbers across companies of different sizes aren't comparable. A company with $100B revenue and $10B EBITDA looks different from one with $5B revenue and $1B EBITDA — but their EV/EBITDA multiples tell you which is valued more expensively relative to earnings.

Key normalization techniques:

• Use per-share metrics (EPS, FCF per share, book value per share) — removes size difference
• Use LTM (last twelve months) data — consistent time period across all peers
• Use forward estimates — same analyst forecast period (typically next fiscal year)
• Adjust for non-recurring items — include or exclude one-time charges consistently across the table

How to read the output

Once the table is built, the analysis follows a pattern: identify the company that looks cheapest relative to peers, then ask why it's cheap. Is the discount warranted (structural problem, declining market) or unwarranted (temporary headwind, sentiment overshoot)?

The most valuable insight from a comps table is often not the cheapest company — it's finding a company in the middle of the pack that has significantly better margins or growth than the cluster suggests. That combination — middle-of-road valuation + above-average quality — is where many of the best investments hide.

Red flags in peer analysis

Real-world example: FAANG 2022 — peer analysis after the tech selloff

By mid-2022, rising interest rates had triggered one of the worst tech selloffs since the dot-com bust. Meta, Alphabet (Google), Netflix, Amazon, and Apple — the FAANG stocks — had all fallen 40-65% from their 2021 highs. At first glance, everything looked cheap. But a peer comparison revealed very different fundamentals beneath the uniform selloff.

The peer table revealed stark differences: Meta traded at just 8x forward P/E — the cheapest in the group — but investors needed to understand why. Meta was facing a dual headwind: TikTok was taking users and advertising share, and management had committed tens of billions to the metaverse with no clear path to returns. The cheapness was partially warranted.

Alphabet at 14x P/E looked moderately valued with industry-leading search economics (~30% operating margins from Google Search), strong Google Cloud growth, and massive FCF generation. The selloff had compressed the multiple to levels seen during the 2009 financial crisis — arguably unwarranted for a near-monopoly business.

Netflix at 25x P/E looked expensive relative to peers, and the FCF yield of ~2% offered little margin of safety. Netflix was still in investment mode — spending on content and building its ad-supported tier. The comps table correctly identified that Netflix's premium multiple required execution investors weren't confident in.

Investors who built this peer table in mid-2022 and bought Alphabet while avoiding Netflix outperformed significantly over the following 18 months. Alphabet's stock recovered 60%+ as markets recognized the unwarranted discount; Netflix underperformed until it demonstrated margin improvement.