Appearance

Competitive landscape & SWOT ​

Part of Strategy: competitor analysis & market positioning. Evidence-class legend and the measured commercial baseline are there.

1 · The competitor set, corrected ​

📘 The vision document's competitor list is Google My Business, Linktree, Wix/Squarespace and Zomato/Swiggy. 🔎 Three of those four are the wrong contest. Wix and Squarespace are Western website builders a Dadar stall vendor has never heard of; Linktree is a creator tool; Zomato is an aggregator in a category 📘 explicitly out of scope.

🔎 The real contest, ranked by how likely each is to be the reason a merchant does not sign up:

RankCompetitorCategoryWhy it outranks the documented list
1WhatsApp Business (+ Meta Business AI)Messaging, catalogue, payments🌐 78% of Indian small businesses already use it. Free. 15M+ have catalogues. 🌐 Meta shipped Business AI for Indian small businesses in May 2026
2Google Business ProfileDiscovery, reputationFree, on Maps and Search, and owns reviews — the trust primitive we have none of
3JustdialPaid local discovery🌐 56.1M listings, 639,200 paid campaigns, ~₹1,220 Cr annualised revenue, field sales force, and it has already launched an end-to-end business-management suite for SMEs
4ONDCPublic commerce infrastructure🌐 500M cumulative transactions, 3 lakh+ sellers, 100+ buyer apps, 400+ cities. A structural threat and our biggest unexamined opportunity
5Vyapar · myBillBook · Tally · MargSMB billing/ledgerThe tool a merchant with money already pays for. 🌐 Vyapar ₹69 Cr FY25
6Khatabook · OkCreditDigital khataFree, huge installed base, and the direct precedent for our own khata thesis
7Digital-card vendors — MeCard, Uniqode, HiHello, Tapect, TapMo, VCRDSDigital visiting cards🌐 Dozens of them, price-collapsed. This is the category an outsider will mistake QRSETU for
8Dukaan · Instamojo · Bikayi/BIKSMB storefront builders🌐 Dukaan pivoted after failing to find fit in kirana tech; Bikayi moved upmarket to enterprise
9The notebook, and nothingStatus quo🔎 Statistically the incumbent we actually lose to. 📘 The vendor tracks 1,500 idols in a notebook
10Vertical SaaS per segment — Petpooja, salon POS, PractoDeep vertical toolsBecomes the competitor the moment we enter their vertical seriously

Two entries deserve their reversal called out

🔎 ONDC is filed as a threat and is probably an opportunity (§4). And 🔎 "the notebook" is the honest number-one: the festival vendor did not reject a competing app, they had never used one. That changes the sales conversation from "we are better than X" to "here is what you stop doing by hand", which is the pitch differentiation §5 recommends.

2 · Dimension-by-dimension comparison ​

Scale: 🟢 strong · 🟡 partial · 🔴 weak or absent · ⚫ not applicable. QRSETU's column is 🧮 measured against live Dev on 2026-08-18, not against intent — where a capability is designed but unbuilt it scores 🔴 and the note says so.

DimensionQRSETU (measured)WhatsApp BusinessGoogle Business ProfileJustdialONDCVyapar / myBillBookKhatabook / OkCreditDigital-card vendors
Merchant onboarding friction🟡 self-serve, ~2 min claimed; 🧮 5 workspaces exist🟢 already installed🟢 minutes🟡 field rep🔴 needs a seller app🟡 app install🟢 minutes🟢 minutes
Public page / card🟢 SSR, edge-cached, manifest templates🟡 a profile, not a page🟡 a listing🟡 a listing page⚫🔴🔴🟢 that is the whole product
Catalogue depth🟢 unbounded, faceted, masonry designed🔴 500-item hard cap, 25 average🟡 basic products🟡🟢 via seller app🟡 inventory, not shoppable🔴 shut down🔴
Consumer discovery🔴 no directory live; in-app feed just gaining a read path🟡 no discovery, only contacts🟢 best in market🟢 192.9M quarterly users🟢 rides Paytm/PhonePe⚫⚫🔴
SEO🔴 landing is a placeholder; area/category pages not built🔴🟢🟢🟡🔴🔴🟡
Reputation / reviews🔴 no table exists at all🔴🟢 decisive🟢 ratings🟡⚫⚫🔴
Orders🟢 live, 🧮 19 orders, buyer UI shipped🟡 chat-mediated🔴🟡 leads🟢🟡 invoices🔴🔴
Payments🟢 Route split, commission at source, 🧮 18 payments🟢 UPI in-chat, 🌐 65% adoption🔴🟡🟢🟡🟡🔴
Handover proof / collection code🟢 designed, and unique in this table🔴🔴🔴🔴🔴🔴🔴
Engagement / notifications🔴 push does not exist; email broken since ~1 Aug🟢 the channel merchants and buyers already live in🟡🟡🔴🟡🟡🔴
Analytics for the merchant🔴 nothing recorded; beacon 404s (QRS-734)🟡🟢 insights🟢 lead reports🟡🟡🟡🟡 view counts
Proactive nudges / assistant🔴 principle only; no substrate🟡 🌐 Business AI, May 2026🔴🔴🔴🔴🟡 payment reminders🔴
Vertical workflow depth🟡 architecture 🟢, substrate 🔴 — 🧮 6 of 11 primitives have no tables🔴🔴🟡🔴🟡 accounting only🔴🔴
Khata / credit ledger🔴 no balances table🔴🔴🔴🔴🟢🟢🔴
Subscriptions (merchant billing)🔴 no collection path; manual bank transfer⚫ free⚫ free🟢 prepaid⚫🟢🟡🟢
Enterprise / seats🟡 tree + oversight designed; 🧮 organizations has 0 rows, no org-admin portal🔴🟡🟢🟡🟡🔴🟡
Promotional monetisation🔴 inert by design; fails closed forever🟢 Meta ads🟢 Google ads🟢 the model🔴⚫⚫🔴
Cross-platform reach🟢 one codebase → Android, iOS, web🟢🟢🟢🟢🟢🟢🟡
Cost to the merchant₹0 free · ₹9,999/season the only priced plan₹0₹0🌐 ~₹19,000/yr average paid campaign₹0 + commission~₹1,000-4,000/yr₹0₹500-3,000/yr

The three readings that matter most ​

🔎 (a) We are strong exactly where nobody competes, and weak exactly where everybody does. Catalogue depth, handover proof and money-split at source are 🟢 and near-unique. Discovery, SEO, reputation, engagement and analytics are 🔴 — and those five are what a merchant means by "will this get me customers?"

🔎 (b) Justdial's paid conversion is the most sobering number in this document. 🌐 639,200 paid campaigns against 56.1M listings is 1.14%, achieved with a field sales force, at ~₹19,000 per paid campaign per year. If the largest local-commerce database in India converts 1.14% with feet on the street, a two-person self-serve funnel should plan for less, not more. See revenue-model.

🔎 (c) Justdial's ARPU is 2× our top plan. ₹19,000/year versus ₹9,999/season. Indian merchants will pay real money for local demand. They pay it for leads, which is the thing we cannot yet deliver — which is an argument for building the discovery surface, not for lowering price.

3 · SWOT ​

Strengths 🧮 unless marked ​

  • The money loop works on real money. 🧮 19 orders, 18 payments, 2 Route transfers, 1 platform settlement, a payout account, and 📘 a live order with the provider fee and a mixed-rate tax snapshot recorded. Detect-only reconciler, and a watchdog asserting a sweep examined something (manual-run only since 2026-09-23, when all scheduled CI was removed to save Actions quota).
  • Four-layer platform model, verified rather than asserted. 📘 Applicability derives from primitive composition with zero grant rows. A new vertical is a row.
  • Engineering governance far beyond a two-person norm — 24 gates, 26 ADRs, a load-bearing release system, mutation-tested tooling. 🔎 This is the reason one developer can hold this surface area at all.
  • Real vendor discovery, one industry at a time. 📘 The festival brief overturned four of its own assumptions after one session; 📘 the direct-seller brief was falsified by owner input and says so.
  • Anonymous-first public surface, structurally, not by policy.
  • Three-surface parity by construction, with static gates encoding past incidents.
  • A named, dated, external launch deadline with 📘 12 personally-visited vendors. 🔎 Rare focus.

Weaknesses 🧮 unless marked ​

  • ⚠ Nothing is measured. No analytics table; the card beacon posts to an archived function (QRS-734); mobile has zero analytics wiring. 🔎 Every merchant-facing value claim, every proactive nudge and every renewal conversation depends on data that is not being collected.
  • ⚠ No notification transport on any channel. 📘 Push does not exist by decision; email OTP has been broken since ~1 August (SMTP 535). 📘 The festival brief's own words: "a booking nobody hears about during a four-week season is worse than no booking at all."
  • ⚠ No reputation capability. 🧮 No reviews or ratings table. 🔎 The single most valuable local-commerce asset, entirely absent, and not in the eleven primitives.
  • ⚠ Six of eleven primitives have no substrate. 🧮 parties, schedules, ledger, balances, assets, campaigns do not exist. So CRM, bookings, khata, AMCs and campaigns — most of the retention story — are architecture, not product.
  • ⚠ No subscription collection path. 🧮 pro.price_minor is NULL; the only priced plan is ₹9,999/season, activated by hand after a bank transfer. 📘 Documented as deliberate for 12 vendors, and it does not scale past them.
  • ⚠ Zero content on the platform. 🧮 0 media rows, 0 card links, 0 card hours, 6 catalogue items. A card with no photographs is not shareable, and 🔎 photographs are what the vendor's buyers choose on.
  • No discovery surface live. Landing is a placeholder; the public directory twin is unbuilt; the in-app feed is only now getting a read path.
  • The DOM half of the design system does not exist. 🧮 apps/web/src/ui/ holds only a README, so component parity across the two idioms is unverified by construction.
  • Willingness to pay is unmeasured in every vertical. 📘 Festival Q21 returned NA.
  • 🔎 Bus factor is 1. One developer holds the schema, the gates, the release system and the money path. The gate discipline mitigates knowledge loss; it does not mitigate absence.
  • 🔎 Support is the dominant marginal cost and it is unmodelled. 📘 QRS-487 says so explicitly: infra is ~2% of a ₹9,999 plan while an informal vendor uploading 1,500 photos in a four-week crunch is human time that dwarfs servers.

Opportunities ​

#Opportunity🔎 Why it is realEffort
O1Seasonal/festival micro-retail🌐 Structurally outside WhatsApp's 500-item cap; nobody builds for a 28-day businessIn flight
O2Google review funnel — QR that drives reviews to the merchant's Google profile🔎 Merchants pay for reviews. We already mint QR codes. Complement Google instead of losing to itDays
O3WhatsApp as the share-and-notify rail🌐 78% merchant penetration, ~universal consumer reach. Solves the broken-transport problem with no push infrastructureDays
O4Done-for-you catalogue setup📘 The measured blocker is ~50 hours of data entry, not the card. 🔎 Highest-conviction revenue in the setZero engineering
O5ONDC as a seller app🌐 3 lakh sellers, 100+ buyer apps. Demand without building a consumer base — see §4Research first
O6Reseller / association channel🔎 The only mechanism by which two people reach 1,000 merchants. 🌐 It is how Justdial and Vyapar actually grewMarketing-led
O7MLM upline seats📘 The upline is compensated on downline volume, so the buyer's incentive is pre-aligned and there is no procurement26.0.2
O8AI catalogue ingestion🔎 Photo → draft name/size/price turns 50 hours into 5. Attacks the measured #1 adoption blockerMedium
O9Create-and-share consumer artifacts📘 QRS-531: supply-free, so it is the only consumer-acquisition loop that works before inventory existsLarge; a second product
O10Physical QR placement in tier-3 towns📘 QRS-608's own conclusion: buy the channel, do not build a vertical to get it. Per-panel slugs make it measurableCash, not code

Threats ​

#Threat🔎 SeverityEarly-warning signal
T1WhatsApp lifts the catalogue cap, or Business AI absorbs the assistant ideaExistential to the wedge. 🌐 Registrations in India +61% YoY; Business AI shipped May 2026Watch Meta's India business-product releases quarterly
T2Google Business Profile remains the only discovery merchants needHigh. Free, installed, and owns reviewsAsk every new merchant where their customers find them
T3Justdial extends its SME suite downmarketMedium-high. 🌐 They already launched one, and they have the sales forceTheir quarterly results name campaign growth
T4ONDC commoditises the storefrontMedium. A merchant reachable from PhonePe and Paytm needs no card of ours for demandSeller-count growth in our cities
T5Commission leakage is structuralCertain, and 📘 already conceded. QRS-489: discovery-only platforms leak by construction📘 The pre-committed rule: online share under ~10%
T6Two-sided cold start never breaksHigh, and 📘 acknowledged as a chicken-and-egg argument running in the direction we can least pushConsumer registrations after Ganapati
T7Seasonality: only one Ganapati inside the target windowHigh, and structural. See revenue-model §2The calendar. It will not move
T8Regulatory load exceeds two peopleMedium-high. 📘 Direct-selling income and health claims, DPDP, E-Commerce Rules seller identity, IT Rules grievance clock — with no moderation tooling builtThe first takedown request
T9Key-person concentrationHigh and unmitigatedAny week the developer is unavailable
T10Support load at 500+ merchantsMedium-high, and 📘 named as the dominant marginal costTickets per merchant per week

4 · ONDC: the reversal worth arguing ​

🧮 ONDC appears zero times in the portal. 🔎 It should appear as a strategic decision, because it cuts directly at the hardest problem in the plan.

The problem it solves. 📘 The Marketplace spec concedes that on day one there are zero registered consumers, and 📘 QRS-528 records the flywheel as "a chicken-and-egg argument running in the direction we are least able to push." Building a consumer base with two people and no marketing budget is the single least tractable item on the roadmap.

What ONDC offers instead. 🌐 500M cumulative transactions, 100+ buyer apps including Paytm and PhonePe, 400+ cities. A seller app publishes a merchant's catalogue into that demand. QRSETU already has the hard half — a structured, typed, per-industry catalogue with variants, units, stock and tax — which is precisely what an ONDC seller integration needs and what a WhatsApp catalogue cannot provide.

Why it fits the architecture unusually well. 📘 The Fulfilment primitive exists, orders and payments are live, and 📘 industry scope's §6 already lists "Marketplace / discovery" as riding existing hooks with no foundation impact. An ONDC seller app is a new outbound projection of the catalogue, not a schema change.

The counter-arguments, because this must be a debate and not a pitch:

  • 🌐 Adoption is metro-concentrated and tier-2/3 awareness is low — which is where our merchants are. So the demand may not be where our supply is.
  • 🌐 Mobility and metro ticketing dominate the transaction count. The 500M number is not 500M retail orders, and reading it that way would be exactly the kind of unbacked inference this repo's third rule forbids.
  • It contradicts "own your customer." An ONDC order arrives from a buyer app that owns the relationship. 🔎 That is a real strategic cost and it is smaller than it sounds, because our verticals are not being disintermediated by anyone today.
  • Integration is a genuine protocol project — catalogue mapping, order and fulfilment callbacks, network participant compliance. ❓ Cost unestimated. 🔎 Weeks, not days.
  • Our launch vertical cannot use it. 📘 Festival idols are collection-only, one-of-a-kind, and the season is four weeks. ONDC would serve the year-round goods verticals (kirana, boutique, sweet shop), which are not built.

🔎 Recommendation: do not build it in this horizon, and do not leave it undecided either. Spend one day producing a written feasibility note — participant requirements, catalogue mapping effort, which of our industries could list, and whether buyer-app demand exists in our target pincodes. Then record a decision. ONDC is the only available answer to "where does demand come from" that does not require us to build an audience, and leaving the largest strategic option in the market unexamined is worse than rejecting it. Logged as QRS-747.

5 · Where competitors are stronger, and where we can realistically win ​

They win, decisivelyDo not contest
Discovery and SEO — Google, JustdialCompete by complementing: drive reviews to Google (O2), publish an SEO directory for long-tail area queries nobody optimises for
Reputation — GoogleDo not build a rival review graph. Surface the merchant's Google rating on their card and help them earn more of it
Messaging reach — WhatsAppDo not replace it. 🔎 A merchant will not abandon the channel their customers already use, and 🧮 conversations having 0 rows is early evidence
Billing and accounting — Vyapar, Tally📘 Correctly out of scope. Keep it that way
Distribution muscle — Justdial's field force, Vyapar's ₹102 Cr salary lineCannot be matched. Rent it via resellers and associations (O6)
We can realistically winWhy
Catalogue at 500+ items🌐 A structural cap in the market leader's product
Seasonal businessesCommercially unattractive to everyone with a shareholder
Handover and collection proofNobody else has it, and it attaches to a moment of real anxiety
Per-vertical workflow at micro scale🔎 The four-layer model's actual payoff: the twentieth vertical is a row
Proactive nudges grounded in the merchant's own data🔎 Requires only measurement and transport — the two cheapest gaps in product-gaps

Sources ​