The three statements are the summary; the notes are the evidence. In a deep dive you read them all, but you read them as an extractor, not a completist: each note answers a specific question, and you take the signal and move on. Here is the extraction map for the six that matter most, with the Asian Paints read for each.
What should you extract from each note?
- Revenue recognition. Extract: when revenue is booked (point-in-time on dispatch vs over-time), gross-vs-net (agent vs principal), rights of return, rebates/discounts, and, critically for any product sold through a channel, sell-in vs sell-through. The red flag it hides: channel stuffing: booking primary sales to dealers that end demand hasn't pulled through, inflating revenue while dealer inventory quietly builds (rising receivables and dealer-scheme spend are the tells, M2.06). Asian Paints: revenue recognized on dispatch to a vast dealer network, so in any soft quarter you must ask whether reported "sales" are sell-in the dealers will digest slowly, and watch primary-vs-secondary commentary and the receivables/dealer-incentive lines. This is exactly the note that a competitive shock (Birla Opus discounting to fill dealer shelves) can distort.
Advertisement
- Related-party transactions (RPTs). Extract: every transaction with promoters, promoter entities, key management, JVs, and associates (sales, purchases, royalties, rent, loans, guarantees), and whether they are at arm's length. The red flag it hides: value siphoned out of the listed entity into promoter pockets (the Indian governance failure mode par excellence; M8.04 and the ethics thread; think of the forensic cases in M2.08). Asian Paints: multiple promoter families (illustrative; check the current promoter group), an industrial-coatings JV with PPG (a genuine related party, so check the terms and the profit-share), and any purchases from or rent to promoter-linked entities. The bar is not "are there RPTs" (there always are) but "are they material, arm's-length, and stable". And you carry the RPT map into Day 3's incentive read.
- Leases (Ind AS 116 / IFRS 16 / ASC 842). Extract: the ROU asset and lease-liability balances, the lease maturity profile, and short-term/variable lease costs kept off-balance-sheet. The red flag it hides: hidden financial leverage (leases are debt) and the EBITDA comparability trap (see normalising a ten-year build). Asian Paints: leases for depots, regional offices, and the Beautiful-Homes retail stores, modest relative to the balance sheet, but you add the lease liability to net debt for a true leverage read and you confirm the FY20 EBITDA step is the standard, not the business. (For a retailer or airline this note is huge; for Asian Paints it is a check-the-box confirmation, but you check it.)
- Contingent liabilities & commitments. Extract: disputed tax demands, litigation, guarantees, and capital commitments (capex contracted but not incurred, a leading indicator of future capex). The red flag it hides: an off-balance-sheet loss about to become an on-balance-sheet one; in India, large disputed indirect/direct-tax demands are common and occasionally real. Asian Paints: disputed tax matters (frequently large gross numbers, usually provisioned or assessed as remote; read management's assessment and the auditor's view), and capital commitments that tell you the next year's capex before the cash-flow statement does. Size the plausible loss, not the gross claim.
- Segments. Extract: revenue and result (EBIT/margin) by segment, and capital employed by segment if disclosed. This is where you find the real profit engine (often not the biggest revenue line) and where re-segmentation breaks hide (see normalising a ten-year build). The red flag it hides: a glamorous growth segment cross-subsidized by a boring cash cow, or a "diversified" group that is really one good business funding several bad ones (the SOTP/holdco problem, M3.10). Asian Paints: the segment note is decisive: decorative India carries essentially all the economics (high margin, high ROIC), while international is lumpier and lower-return, industrial is a modest JV-inflected business, and home décor is the sub-scale investment drag the ten-year history flagged. The segment note is what lets you say "this is a great decorative-paints business with some attached projects," which is a very different sentence from "this is a great diversified home-improvement company," and the reverse DCF cares enormously which one is true.
- ESOP / share-based compensation. Extract: the SBC expense, the add-back in CFO, the options/RSUs outstanding and their dilution, and the vesting/grant trend. The red flag it hides: real compensation cost hidden below the line and future dilution the share count hasn't yet felt (a huge issue for Indian new-age tech and US software, per M1.09 and the M2.04 "FCF flattered by SBC add-back" trap). Asian Paints: SBC is immaterial (a legacy management-owned compounder, not an options factory), so this note is a quick confirmation. But you confirm it, because the discipline is universal and the day you deep-dive an Indian IT or a US SaaS name, this note moves from a footnote to a first-order driver of real per-share economics. (This is why we teach it on a company where it's small: so the habit is installed before you meet a company where it's fatal.)
What else is worth a quick scan?
Two more you always at least scan: the auditor's report (opinion type, Key Audit Matters, going-concern language; India's CARO 2020; M1.10) and, for India, the shareholding pattern (promoter holding and pledge %, since a pledged promoter is a forced seller in waiting). For Asian Paints both are clean (illustrative), which is itself information: a boring audit and an unpledged promoter are the quiet luxuries of a quality franchise, and their absence elsewhere is where a deep dive earns its keep.
The notes-extract tab, done, is one line of signal per note, perhaps eight lines. It looks slight. It is the difference between an analyst who read the accounts and one who read the summary of the accounts.
Day 1 is done. You have a normalized decade on one screen, the incremental-ROIC contrast that is the compounder's real verdict, a management-credibility ledger, and eight lines of extracted note-signal. You know this is a wonderful business being tested. Day 2 places it in its industry, benchmarks it, models it forward, and prices what the market has already assumed.