ANS-100/P — Consumer Internet Sector Calibration Note

Drafted 30 August 2026 · Ratified 30 August 2026, before the first consumer internet company was scored · Rater 1 · Companion to ANS-100/P v2.1 (frozen anchor set) · Sector: Eternal, Swiggy, Nykaa (FSN), Info Edge, PB Fintech, MakeMyTrip, Zoho (private), Zerodha (private)

Ratification log: B11 ratified 30 Aug 2026 — Q32 classed N for BFSI; no score changes; BFSI files final as drafted pending right-of-reply.

1. Why digital natives need their own note

These companies' revenue is already digital and their core operations are algorithmic by construction. Without calibration, they would collect free points exactly where banks would have (B3's lesson), and their genuine AI differentiation — which lives in AI-native products, agentic-commerce posture, and support/ops automation — would be under-read. Two universe members are private (Zoho, Zerodha), triggering the methodology §5 confidence machinery for the first time.

2. Proposed rulings (ratify individually: C1 ☐ … C10 ☐)

C1 — Q25 economic pair. Revenue (or GMV where revenue is accounting-distorted) per employee, read with the adjusted-EBITDA/contribution-margin trajectory. A1 unchanged: level 3 needs a number linking named AI systems to the reported metric. Accounting-model changes (e.g., 1P inventory transitions) are named confounds.

C2 — Q27 baseline exclusion. The algorithmic operation of the marketplace itself (matching, routing, ranking, dynamic pricing) is the business, not automation evidence — excluded from Q27 the way channel share was under B2. Q27 requires disclosed no-touch rates in the human-labour-adjacent processes: support resolution, seller/restaurant onboarding, catalogue generation, content moderation, KYC.

C3 — Q30 re-anchored (the B3 analog). The question becomes: is the company monetising AI natively? Level 2 = AI features material and disclosed in the product. Level 3 = AI-native monetisation disclosed as revenue or volume lines — AI products sold (including externally), AI-driven advertising or take-rate economics disclosed as such, paid AI tiers.

C4 — Q31 re-anchored: agentic disintermediation is this sector's existential exposure. AI agents ordering, shopping, booking and job-matching on users' behalf threaten the app-interface moat; LLM answer engines erode search-led traffic (acute for listings and content businesses). Mitigation = proprietary supply/logistics/data moats, agent-ready surfaces (APIs, agentic-commerce partnerships), owned demand. Level 2 = exposure acknowledged with stated mitigation; 3 = structural or quantified mitigation. Silence on agentic disintermediation is a scored absence for this sector.

C5 — Q33. Couldn't-exist-pre-AI products with disclosed volumes or external sales. Internal AI tooling productised and sold externally is first-class level-3 evidence.

C6 — Security ceiling. Same B6 discipline; consumer-scale agents touching payments, refunds, and account actions get the explicit evaluation, documented in-file.

C7 — Q18 without a regulator baseline. No RBI floor here: DPDP-readiness plus platform trust-and-safety specifics = 2; audited/certified AI management systems or AI-in-scope bug bounties = 3. Expect real dispersion — several of these companies run public bug-bounty programs.

C8 — Private-company machinery (Zoho, Zerodha). Same instrument, same anchors, same A-rules. Confidence grade does the work: if fewer than 18 questions score, the company publishes as a band-range with the standing invitation line (§5). No sympathy points and no penalty — coverage states itself. Evidence norms noted: both publish unusually rich engineering material for private firms; Zerodha voluntarily discloses financials.

C9 — Entity rulings. Eternal: consolidated (food delivery, Blinkit, Hyperpure, District) — the Blinkit 1P accounting change is a named C1 confound. Info Edge: operating platforms (Naukri, 99acres et al.); investee companies (Zomato/PB stakes) excluded from evidence. PB Fintech: Policybazaar + Paisabazaar. MakeMyTrip: NASDAQ filings are the primary base. Nykaa: FSN consolidated.

C10 — Signal library additions. Engineering blogs (first-class in this sector and A7(c)-gradeable), app release notes, founder shareholder letters, product launch documentation, developer/API portals, bug-bounty scopes.

3. What does not change

All v2.1 anchors, P-caps, evidence floor, window, A1–A10, ceilings arithmetic, evidence-language rule. Application, not amendment; the freeze holds.

4. On ratification

Score in universe order: Eternal → Swiggy → Nykaa → Info Edge → PB Fintech → MakeMyTrip → Zoho → Zerodha. Sector synthesis after all eight.

Published 9 September 2026 under CC BY-SA 4.0 as part of the Edition One replication material. Status changed from "proposed" to "ratified" on publication: the ruling was ratified on 30 August 2026 before the sector was scored, and the draft filename had not been updated. No ruling text is changed by this publication.