B2B Procurement Platform for Kirana Shops in India

Who does it:

• The kirana owner (sub-500 sq ft general store, neighbourhood market or residential colony) calls or WhatsApps the distributor's salesman with their order — typically a replenishment list spoken into a voice note or typed as a short text.
• The distributor's field salesman (one per 40–80 kiranas, in tier 2 cities often one per 100+) visits each kirana daily or alternate days, collects口头 orders on paper or a physical order book, returns to the depot, and enters them into a computer or passes them to a supervisor who does the data entry.
• The distributor owner or purchase head reconciles orders against inventory and places a pooled order with the brand/supplier.

What they use:

• Primary channel: WhatsApp voice notes + text. Kirana owners send a list like "20 L daal, 5 kg rice, 2 Handel oil" in a text or voice note. Salesman screenshots or screenshots it.
• Secondary: Plain phone calls for urgent or large orders.
• Distributor back office: Excel sheets, Tally for accounting, sometimes an ERP for mid-size distributors (annual revenue ₹5 crore+).
• Paper: Many small distributors (sub-₹2 crore revenue) still use physical order books. The transition to Excel is recent and incomplete.

Where time and money leak:

• Salesman time — a field visit to collect an order that takes 3 minutes of actual conversation and 45 minutes of travel is the core inefficiency. A salesman can physically visit 10–15 kiranas per day; order collection is typically 2–5 minutes per stop.
• Order error rate — manual data entry from paper or voice notes into Tally/Excel produces a documented error rate of 5–15% on SKU, quantity, or price in small distributor operations. Returns and re-delivery eat margin.
• Stockout cycles — because ordering is reactive and infrequent (driven by salesman visits, not real-time signals), kiranas under-order fast-moving SKUs and over-order slow ones. Distributors cannot plan pooled brand orders without accurate downstream signals.
• Salesman attrition — field salesman churn is high in tier 2/3 cities (annual attrition cited at 30–60% in trade press, unverified). Institutional memory walks out the door.
• No data for the kirana — the kirana has no visibility into their own purchase patterns, no reorder reminders, and no leverage to negotiate. They are flying blind on their own stock.

---

Who profits from it staying manual:

• Field salesmen — their job security depends on being the information bridge. A system that lets kiranas order directly threatens their role. In practice, many salesmen informally discourage kiranas from using direct-ordering tools.
• Unorganised brokers — small commission agents who sit between some distributors and kiranas in FMCG-light categories (masala, oil, pulses) take a 1–3% cut. They benefit from opacity.
• Distributors who prefer pricing opacity — a manual system lets distributors offer kirana-specific pricing (loyalty discounts, volume waivers) without audit trails that brands can see.

Who is hurt:

• Kirana owners — spend 45–90 minutes per day on procurement communication across calls and WhatsApp. They have no data on what they should reorder. Stockouts on high-margin fast-moving goods cost estimated 8–12% of potential sales (no reliable verified figure, this range appears in trade journalism).
• Honest distributors — lose to distributors with better information and lower cost-to-serve. A distributor with a direct WhatsApp ordering channel can undercut on price because their cost per order is lower.
• Brands and large FMCG companies — they fund field force (average cost per store visit ₹80–150 for a primary sales call, varies by company and city) to collect orders that a digital channel could do for a fraction. Britannia, Hindustan Unilever, and Parle have each run pilot programmes to digitise the last-mile order.

Who would pay to change it:

• Distributors with 50+ kirana accounts are the primary target. Their cost-to-serve per order is visible (salesman salary + travel, typically ₹15–40 per order in tier 2 cities), and they have enough volume to amortise a cheaper channel.
• Brands in categories with high reorder frequency — edible oils, staples, biscuits, personal care — have the strongest incentive to fund distributor adoption because their margin depends on shelf availability at the kirana.
• Kirana owners themselves would pay ₹200–500/month for an app that auto-reminded them to reorder and tracked their purchase history, but only if the friction of using it was near zero. Historical evidence (ShopKirana, others) shows this cohort has very low willingness to change existing behaviour.

---

The narrowest product/service/agent that delivers a visible win on day one:

Order Taker Agent (WhatsApp-native)

• The kirana owner sends a WhatsApp message (text or voice note with their order) to a phone number or WhatsApp Business account.
• An AI agent (or a human operator in the agency model) parses the message, maps products to SKUs, confirms the order with the kirana, and routes it to the distributor's Tally, Excel, or WhatsApp group in a structured format.
• The distributor gets one structured incoming order per kirana, replacing the paper handoff from the field salesman.
• The kirana gets a written order confirmation and an estimated delivery time.

Day-one scope (deliberately narrow):

• Only handle kirana → distributor order submission.
• One distributor partner, 20–30 kirana accounts.
• One product category (say, edible oil + pulses — high reorder frequency, simple SKU list).
• One city, one language.

Who pays:

• The distributor pays a monthly fee.
• Not the kirana. Do not charge the kirana in phase one.

Pricing SHAPE:

• Per verified order placed — ₹3–8 per order, depending on order value. A verified order = one the kirana confirmed and the distributor received in a usable format. This is the right SHAPE because it aligns value (the distributor avoids a salesman visit or a data-entry error) with cost (the agent processes it).
• Do not start with a per-seat SaaS fee. The distributor will not pay per kirana seat until they see the order quality improve.
• Floor: ₹8,000/month minimum per distributor (at ~1,000 orders/month, ₹8/order). Ceiling: ₹25/order if the agent also handles returns and reconciliation. For 90-day test, offer a flat ₹15,000/month lock-in to align incentives during the pilot.

---

Kirana-facing ordering apps (kirana orders from a distributor or brand app):

• ShopKirana — operates in 50+ cities; connects kiranas to brands and distributors via an app. Kiranas place orders through the app; delivery fulfilled by distributor. Has raised funding (total cited at ~$70M across rounds, Crunchbase). Reality check: active monthly kirana users is not publicly disclosed; industry estimate is low tens of thousands against 12+ million kiranas nationally.
• JioMart Partner app — Reliance's kirana digitization play. Covers groceries and staples. Heavy subsidy from Reliance; unit economics unclear.
• Amazon Easy Store / Flipkart SmartBuddy — kirana partners fulfil last-mile delivery; not a procurement tool for the kirana's own restocking.
• Zopper — B2B marketplace for electronics and appliances; not FMCG/staples.
• Udaan — B2B marketplace covering a wide range; used by some small retailers. Has raised significant capital (over $200M disclosed). Primarily serves modern trade and online sellers, not kirana-focused in staples.

Distributor management software:

• Konydash (formerly GoBillion) — works with FMCG distributors; SaaS for order management.
• OrderBoat — distributor ERP and order management SaaS.
• Fraxiton — unverified; appears in trade listings.
• Tally — used by nearly all small distributors for accounting; not designed for order collection from kiranas.

Direct competitor to the proposed wedge:

• WhatsApp Business API with automation — several distributors in Mumbai, Delhi-NCR, and Bangalore have experimented with WhatsApp Business with auto-reply flows or third-party integration. This is the closest existing workaround. It is partial: it handles order receipt but not order verification, reconciliation, or error correction.

The gap these all leave open: None of the above eliminates the field salesman for order collection in the 80% of kiranas not on any app. The wedge is not building a better app — it is building a channel that works with the kirana's existing WhatsApp behaviour and delivers structured output to the distributor.

---

Kill condition 1: Distributors will not pay for digital order capture because their field salesmen are free or nearly free.

• What this means: If the distributor's salesman cost per order is below ₹5 (possible in very dense urban markets where a salesman visits 50+ kiranas/day and is on a fixed salary of ₹12,000–15,000/month), there is no cost incentive to switch.
• How to check cheaply: Interview three distributors in the target city. Ask: "What do you currently pay per order to collect it from a kirana?" Include salesman salary, travel, phone. Calculate it. If it is below ₹5/order, the economics of a ₹5–8/order agent do not work.
• Pass: Cost per order above ₹8.

Kill condition 2: Kirana owners will not use WhatsApp to order — they need the salesman to visit in person for credit and relationship.

• What this means: In many kirana-to-distributor relationships, credit (akhri or udhar — 15 to 30-day credit) is extended based on the salesman's relationship and judgment. A digital order channel that bypasses the salesman may result in the distributor cutting off credit to the kirana, which the kirana cannot absorb.
• How to check cheaply: Ask 10 kirana owners: "If you could order by WhatsApp and get the same credit terms as today, would you?" Observe their hesitation. If the answer is "I need bhaiyya to come" in the first sentence, the credit-dependency problem is real.
• Pass: Kirana owners say they would order digitally if credit terms were maintained. Distributor confirms they would extend credit based on order history, not relationship.

Kill condition 3: One distributor cannot generate enough order volume to pay for the agent's operating cost in a single city.

• What this means: If a distributor with 30 kirana accounts generates 25 orders/day (one every other day per kirana), that is 750 orders/month. At ₹8/order, monthly revenue is ₹6,000. The agent processing cost (human or AI) needs to be below ₹4/order to leave margin. If it is not, the model does not work.
• How to check cheaply: Ask the distributor partner for their actual daily order volume from kiranas. Calculate revenue per order at target pricing. Estimate agent cost (AI API calls at ₹0.2–0.5 per order for simple parsing + human verification at ₹2–4 per order).
• Pass: Gross margin per order above ₹3 at the target price.

---

Budget: ₹25,000

• Weeks 1–2: Distributor partnership and kirana recruitment

- Find one mid-size FMCG/staples distributor in one city (target: Jaipur, Indore, Lucknow, or Surat — decent kirana density, not oversaturated with tech). - Sign a revenue-share or guaranteed minimum pilot agreement. - Select 20 kirana accounts from the distributor's existing base. - Give each kirana a WhatsApp number to message orders to. Use a dedicated SIM + WhatsApp Business app. Do not build anything yet. - Cost: ₹0 software, ₹500 for a second phone number and WhatsApp Business setup.

• Weeks 3–6: Human-run agent (agency mode)

- One part-time operator (₹8,000/month or a college student on internship) receives WhatsApp messages, parses them, confirms orders, and forwards structured text to the distributor's Tally-ready format via WhatsApp. - The operator uses a simple Google Sheet to log each order — no software built. - Distributor validates: is the order accurate? Is it on time? Does it replace a field salesman visit? - Cost: Operator — ₹4,000 for the 4-week period. Phone + data — ₹500. Distributor relationship travel — ₹2,000. - Pass mark: 150+ verified orders placed, order error rate below 5%, distributor says they would continue without the field salesman on these 20 accounts.

• Weeks 7–10: Introduce AI assistance

- Add a simple LLM (OpenAI or Gemini API) to parse voice notes and free-text messages into structured JSON. - Human operator reviews and corrects each output before sending to distributor. - Measure: what percentage of messages does the AI parse correctly without human correction? - Cost: API calls — ₹1,500 for the 4-week period. Operator time reduces by ~40% if AI accuracy is above 80%. - Pass mark: AI parses 80%+ of orders without human correction, error rate still below 5%.

• Weeks 11–12: Distributor payment decision

- Present the 90-day order log to the distributor. - Ask them to sign a ₹12,000/month retainer for continued service (covers 400+ orders) OR move to a ₹8/order variable model. - If they say yes: you have a paying customer and a validated unit economics model. - If they say no: understand exactly why and whether the problem is fixable. - Cost: ₹500 for a one-page summary report.

Budget summary:

• Operator (4 weeks): ₹4,000
• API calls (4 weeks): ₹1,500
• Phone/data/WhatsApp Business: ₹500
• Travel and report: ₹2,500
• Contingency: ₹16,500
• Total: ₹25,000

Pass mark for the 90-day test:

• Distributor renews at ₹12,000/month or higher.
• At least 200 verified orders placed across 20 kiranas.
• Order error rate below 5%.
• Operator time below 2 hours/day.

---

AGENCIFY first, PRODUCTIZE second, AI-FY in parallel as a cost lever — not a selling point.

The kirana-distributor ordering problem is a distribution and behaviour problem before it is a software problem. Every prior software play in this space (kirana apps, distributor SaaS, B2B marketplaces) has failed to break the field salesman's role because none of them solved for the one thing the salesman does that software cannot: extend credit and build a personal relationship that keeps the kirana loyal. A human-run agent that handles order capture — working within the existing WhatsApp behaviour of the kirana — sidesteps the adoption problem entirely and proves the economics before a single line of product code is written. AI is the cost reduction layer inside the agency, not the product. Only after one distributor pays for 90 days does a product roadmap make sense.

Availability confirmed against the .in registry (RDAP) on 2026-09-21. Prices and ownership read from our own intelligence tables. Nothing here is estimated.

Single-word, available now

• fatehs.in — available
• fateh.co.in — available
• fatehs.co.in — available
• procurements.in — available
• procurements.co.in — available

Also available (compound)

• fatehhub.in
• fatehmart.in
• fatehkart.in
• fatehmandi.in
• fatehbazaar.in
• fatehdirect.in
• fatehsupply.in
• fatehconnect.in

Taken and developed — do not chase

• fateh.com · entropy 6.41
• kiranas.co.in · entropy 4.67
• myprocurement.in · entropy 4.85
• kiranamandi.in · entropy 4.91
• gokirana.in · entropy 5.06

---

Generated 2026-09-21 04:41 UTC. Topic from our research queue; no market-size figure appears here unless a source is named. The domain block above is read from our own intelligence tables and confirmed at the .in registry (RDAP); the model wrote the analysis, not the domain facts.