The wrong peer — why a set can be built badly
Half the skill in comps is knowing what does NOT belong. Below is a deliberately bad set: a chip anchor sitting next to a fund, a bank, and a company from a different sector entirely. The table shows what happens — some members refuse outright, others ground but shouldn't be read as peers.
Diagnose it: say why each member is or isn't a valid comparable to the anchor. The mentor grades the diagnosis — the reasoning about comparability, not a recitation of numbers.
A comparable is a company in a similar line of business. A fund isn't a company at all — it's a basket, with no operating statements. A bank files a different income statement, so the walk metrics don't apply. A large company in another sector files everything but has different economics. Comparability is about business, not size.
| Metric | NVDAFY2026 | VOO | JPMFY2025 | KOFY2025 |
|---|---|---|---|---|
| Revenue growth (YoY) | +65.5% vs FY2025 | — | — | +1.9% vs FY2024 |
| Gross margin | 71.1% | — | — | 61.6% |
| Operating margin | — | — | — | — |
| Net margin | 55.6% | — | 31.3% | 27.3% |
| Return on equity | — | — | — | — |
NVIDIA CORP 10-K vs a deliberately mismatched set (NVDA · VOO · JPM · KO), each cell cited or refused per filer, via SEC EDGAR
Educational use only — not investment advice. Figures come from public SEC filings; Echelon teaches you to analyze data, it never recommends buying or selling any security.