ShelfRadar

1. One-liner

Catches dark-store stockouts and ranking drops on Blinkit, Zepto and Instamart before a brand loses a day’s sales.

2. Trend signal — why now?

Quick commerce is no longer a side channel in India — it’s where discovery happens. Blinkit hit ~650k daily orders by May 2026, ahead of Zepto (~550k) and Instamart (~112M orders in Q4 FY26). Funding into the sector ran $271M across 9 rounds by March 2026, up ~1000% YoY. Every D2C brand is now being pushed onto these apps.

And every one of them is bleeding silently. The mechanic is brutal and well-documented: availability is decided per dark store, per pincode — being listed nationally means nothing. When an SKU shows “out of stock,” the platform algorithm deprioritizes it, and even after restock it takes 2–3 weeks to rebuild ranking. Miss fill-rate targets in the first 48–72 hours of a live listing and your rank takes a hit that takes weeks to recover. 50% of consumers switch to a rival app entirely when their preferred brand is OOS — so a stockout doesn’t lose one sale, it leaks high-intent customers permanently.

The kicker: small brands have no alerting. As one industry write-up put it, “It’s a common frustration for brand managers to look at their sales dashboard and see a sudden dip, only to find that their product was ‘out of stock’ for half the city.” Enterprise tools that solve this (42Signals, DataWeave, GobbleCube) are priced and built for category-manager teams at large FMCG brands. The thousands of ₹2–25cr brands newly forced onto q-commerce check listings by hand, daily, or not at all.

Provenance:

• Signal 1 (demand): Stockouts trigger algorithmic deprioritization, 2–3 week ranking recovery, 50% of customers switch platforms when OOS; brands discover dips only after the fact with “no automatic alerts” — https://www.42signals.com/blog/visibility-on-quick-commerce-platforms/ and https://codexxa.net/blog/how-to-sell-on-zepto-and-blinkit/ — 2026-06-22
• Signal 2 (feasibility): Pincode/dark-store-level availability + share-of-search scraping with WhatsApp/Telegram alerts is a proven, commoditized pattern (42Signals, Actowiz); cheap AI to convert raw scrapes into a prioritized action list is the new layer — https://www.actowizmetrics.com/darkstore-data-analytics-instamart-zepto-blinkit.php — 2026-06-22
• Signal 3 (economic): Q-commerce funding $271M / 9 rounds by Mar 2026 (+~1000% YoY); Blinkit ~650k daily orders; small brands already spend on the channel but bleed (founder: “spent over a million capital on these platforms within three months, not even 10% of sales”) — https://www.deccanherald.com/business/with-87-companies-in-indias-quick-commerce-segment-raises-404-million-in-2025-3684626 and https://www.storyboard18.com/brand-marketing/quick-commerces-ad-fee-toll-are-platforms-squeezing-out-small-d2c-brands-73952.htm — 2026-06-22 Category: Underserved niche

3. The opportunity

The incumbents (42Signals, DataWeave, GobbleCube, Actowiz) built a real category — hyperlocal digital-shelf analytics — but they built it for mid-market and enterprise FMCG. Their product is a dashboard for an analyst: deep, broad, multi-platform, priced for a category-management team that has the headcount to stare at it.

The small brand doesn’t have that headcount. The founder is the category manager, the performance marketer, and the ops lead. They don’t want a BI tool with 40 charts. They want a phone buzz that says: “Your bestseller went OOS in 38 of 120 Bengaluru dark stores 3 hours ago — and your search rank for ‘protein bar’ dropped from #4 to #11. Call your replenishment contact.”

ShelfRadar is the action layer, not the analytics layer. Same underlying scraped data, radically narrower job-to-be-done: detect the three things that actually cost money (stockouts, ranking drops, undercut pricing), rank them by lost-revenue impact, and push a one-line “fix this now” to WhatsApp. Priced for a ₹2–25cr brand, not a Britannia.

4. Target market

• Primary customer: Founder or head-of-growth at an Indian D2C brand doing ₹2–25cr annual revenue, live (or going live) on at least one of Blinkit/Zepto/Instamart, with 5–60 SKUs. Think a regional snack brand, a personal-care line, a supplements brand — bootstrapped or lightly funded, no dedicated e-commerce analyst.
• Why they buy (in their words): “Sellers are unaware whether a spike in orders was caused by paid advertising, organic traction, or something else entirely” (Sonalika Sabharwal, SouLilly Toys). “The ROAS rarely goes beyond 1.2x to 1.5x for small brands” — they cannot afford to also leak organic sales to silent stockouts.
• Rough TAM reasoning: India’s q-commerce seller base is tens of thousands and growing; the platforms onboard new brands continuously. Even 10,000 addressable small brands × ₹3,000/mo ACV ≈ ₹36cr (~$4.3M) ARR ceiling on the entry tier alone — and that’s before multi-platform and pricing-intelligence upsells.
• Why now for them: Q-commerce flipped from optional to mandatory for discovery in the last 12 months. Brands that ignored it in 2024 are scrambling onto it in 2026, paying 35–50% of selling price in platform + ad fees, and cannot afford to lose the free sales on top.

5. Product sketch (MVP)

• Connect a brand by handing over its SKU list + the cities/pincodes it cares about (no platform login or API needed for v1 — public app data).
• Per-dark-store / per-pincode availability tracking across Blinkit, Zepto, Instamart, refreshed multiple times a day.
• Instant WhatsApp + Telegram alert when an SKU crosses a stockout threshold (e.g. “OOS in >20% of tracked stores”).
• Daily share-of-search tracking for the brand’s 10–20 priority keywords; alert when organic rank drops sharply.
• Competitor price + availability watch on a handful of named rival SKUs; alert on undercut.
• A weekly “Monday fix list”: the 5 highest-impact issues ranked by estimated lost-sales, in plain language, with a suggested action each.
• No 40-chart dashboard. One screen: today’s leaks, this week’s fixes.

6. AI angle — what’s load-bearing

The scraping is plumbing — not the moat, not the AI. The AI does two load-bearing jobs:

1. Noise-to-signal triage. Raw availability data across hundreds of dark stores is overwhelming. AI clusters it into a ranked, lost-revenue-weighted shortlist (“these 5 things matter, ignore the other 200 fluctuations”) and writes each as a one-line instruction a non-analyst founder can act on. Without this, you’ve just rebuilt the enterprise dashboard the customer can’t use.
2. Anomaly + cause attribution. Distinguishing a real ranking collapse from normal daily churn, and tying a sales dip to its cause (OOS vs. rank drop vs. competitor promo) — exactly the “was it paid, organic, or something else?” question founders say they can’t answer.

Remove the AI and you have a raw data feed nobody on a 3-person brand team has time to read. The product is the triage.

7. Localization angle

India-first by construction. The platforms (Blinkit/Zepto/Instamart), the dark-store-per-pincode availability model, and the buyer wallet are all India-specific. Delivery is WhatsApp-first because that’s where Indian SMB founders live. Pricing in ₹ at a ₹2,500–8,000/mo band that global digital-shelf tools (priced in enterprise USD) structurally cannot match. The same pattern could later port to LatAm/SEA q-commerce, but the wedge is India.

8. Business model — path to $1M–$5M ARR

• Pricing: Starter ₹2,500/mo (one platform, one city, up to 15 SKUs). Growth ₹6,000/mo (all three platforms, up to 5 cities, 50 SKUs, competitor watch). Scale ₹15,000/mo (unlimited cities, pricing intelligence, agency seats).
• ACV: ₹55,000–70,000 ($700–850) blended.
• Math to $1M ARR (~₹8.3cr): ~1,200 brands at a ~₹58k blended ACV. Reachable inside the small-brand segment alone.
• Math to $5M ARR: ~6,000 brands, or fewer brands + an agency/reseller tier (agencies managing 20–50 brands each on a Scale plan) + a pricing-intelligence upsell. The agency channel is the realistic accelerant past $1M.
• Expansion path: more platforms/cities/SKUs → competitor pricing intelligence → ad-spend efficiency module (tie organic rank to paid spend) → light replenishment-suggestion layer.

9. Go-to-market wedge — first 100 customers

• Scrape the shelf, then pitch the leak. The product’s own engine finds, today, which named small brands are OOS across many dark stores. Cold-DM/email those founders a 30-second Loom showing their live stockout map: “you were out in 38 stores this morning.” That’s not a pitch, it’s a diagnosis. Target 5% reply on a list of 2,000.
• Q-commerce agencies and onboarding consultants (the firms running “Blinkit/Zepto listing optimization” services) — partner/reseller deals; they have the brand relationships and want a retention tool. 10–15 agencies each bringing 10+ brands.
• D2C founder communities — the relevant WhatsApp/Slack groups, the Inc42/Storyboard18 readership, and niche subreddits where these exact complaints are posted. Seed with a free “is my brand leaking?” one-time stockout audit as the top-of-funnel hook.

If a brand sees its own live OOS map, the demo is the close. That’s why distribution is the strong axis, not the weak one.

10. Build complexity — justification

Medium. The scraping/availability-tracking infra across three platforms at pincode granularity is real engineering — anti-bot handling, data freshness, scale — but it’s a known, off-the-shelf-adjacent problem (multiple vendors already do it, and scraping toolchains are mature). The AI triage and alerting sit on commodity LLM APIs + WhatsApp/Telegram. No novel ML, no hardware, no compliance bureaucracy. A 2–3 person team ships a credible v1 (one platform, two cities) in ~10–12 weeks, full three-platform coverage in ~4 months.

11. Gating checklist

Gate	Pass?	Note
Legal in target market	✅	Scraping public app data; stays clear of login/ToS-gated seller data in v1. Watch platform ToS as it scales.
Ethical — no harm / dark patterns	✅	Helps small brands compete; no deception.
Market exists (evidence above)	✅	Enterprise incumbents + documented small-brand pain + funded sector.
1–5 person team can build this	✅	2–3 people, ~4 months to full coverage.
Launchable with <$50K / ₹40L	✅	Scraping infra + LLM + messaging costs; no capex.

All five pass.

12. Feasibility score

Axis	Weight	Score	Notes
Problem intensity	20	16/20	Silent revenue leak, felt daily, directly costs money and compounds (ranking decay). Just short of hair-on-fire because some brands haven’t quantified the loss yet.
Demand evidence	15	12/15	Funded enterprise incumbents prove willingness-to-pay; multiple documented small-brand complaints; sector funding. Gap: no direct proof small brands will pay this price for this narrow tool yet.
Build feasibility	15	12/15	Scraping at scale is the real work but well-trodden; AI layer is commodity.
Distribution clarity	15	12/15	The “show them their own live stockout map” cold open is unusually strong; agency reseller channel is concrete.
Revenue mechanics	15	11/15	Pricing benchmarked below enterprise tools, ACV reasonable; ₹2,500 entry needs high volume to matter — agency tier carries the upside.
Time to first revenue	10	8/10	Diagnostic-led selling can close in days once a single-platform/single-city MVP is live; ~8–10 weeks to first paying brand.
Defensibility	10	3/10	Scraping is copyable; incumbents could trivially launch a cheap tier. Moat is speed, SMB-shaped UX, and the agency relationships — execution-only.
Total	100	74/100

13. Qualitative modifiers

Founder-fit tags

technical-heavy (scraping at scale + freshness is the hard part) · domain-expertise-required (must understand q-commerce ranking mechanics to weight the action list correctly).

Key assumptions to validate (3–5)

1. Assumption: Small brands (₹2–25cr) will pay ₹2,500–6,000/mo for stockout/ranking alerts. How to test: Pre-sell to 20 founders off a free live stockout audit; require a card/UPI mandate, not a “yes I’d buy.”
2. Assumption: The “your own live OOS map” cold open converts at ~5%. How to test: Run 200 personalized diagnostic Looms, measure reply + demo-booked rate.
3. Assumption: Pincode-level scraping stays reliable and affordable across three platforms at scale. How to test: Build the single-platform tracker first; measure data freshness, breakage rate, cost per 1,000 SKU-store checks over 4 weeks.
4. Assumption: Agencies will resell rather than build it themselves. How to test: Pitch 10 q-commerce agencies a rev-share; get 2 signed LOIs.

Risk flags

1. Platform dependency / ToS: Entire product reads from three platforms that can change layouts, throttle, or block scraping. Mitigate with multi-platform spread and graceful degradation; never depend on a single app.
2. Incumbent fast-follow: 42Signals or GobbleCube can launch a ₹5k SMB tier in a quarter. Mitigate by owning the SMB UX + agency channel before they notice the segment.
3. Market timing: Platforms may improve their own seller dashboards (Blinkit already offers SKU/city attribution), eroding the gap. Mitigate by being cross-platform — no single platform’s dashboard spans all three.

14. Structured verdict

Score:                  74/100
Verdict:                GO
Confidence:             Medium
Best-fit builder:       Technical founder who can run scrapers at scale, paired with a q-commerce domain advisor
Time to revenue:        8–10 weeks (single-platform, single-city MVP + diagnostic-led sales)
Capital to launch:      ₹6–12 lakh ($7–14K)
Top 3 assumptions to validate first:
  1. Small brands pay ₹2,500–6,000/mo — pre-sell 20 off a free live stockout audit with a payment mandate
  2. "Your own live OOS map" cold open converts ~5% — 200 personalized diagnostic Looms
  3. Pincode scraping stays reliable + cheap at scale — 4-week single-platform reliability/cost test
Kill criteria:
  - Abandon if <3 of 20 audited brands convert to paid after seeing their own live stockout map
  - Abandon if per-SKU-store scraping cost or breakage makes the ₹2,500 tier unprofitable at 4-week test
  - Abandon if Blinkit + Zepto + Instamart all ship adequate free cross-platform alerting before v1 launch

15. Next step — 1-week validation sprint

• Day 1–2: Hand-build (even semi-manually) a live stockout + rank snapshot for 15 named small brands in one city on one platform. This is the demo asset.
• Day 3–4: Send each founder a 30-second personalized Loom showing their own leak map. Ask for a 15-min call and a ₹2,500/mo pre-commit (UPI mandate / card).
• Day 5: Decide. Go if ≥3 of 15 put down a real payment commitment off nothing but the diagnostic. No-go if founders nod (“useful!”) but none will pre-pay — that means the pain isn’t priced, and the ₹2,500 tier won’t survive contact.

Falsifiable: payment mandates, not enthusiasm.