Ten honest answers to the ten questions you're going to ask — in plain language, with real numbers from a working pilot. No corporate jargon. No vapor.
A farmer in Warangal wakes before sunrise praying for rain. Or no rain — he doesn't even know which one to ask for. By 9 AM the rain hits. His paddy drowns. The crop insurance he paid for will pay him ₹40,000 — if a government inspector visits in 14 days. The inspector won't come. His daughter's college fee is due in March.
This is not one farmer's story. This is the same Tuesday in 146 million households. We built AgriSense AI so he never has to guess again. One operating system that wakes up with him, reads the sky, books his drying yard, sells at the right price, files his insurance claim with photo evidence, and deposits ₹38,000 more per acre into his bank account every year.
Free for him. Forever. Built with the government. Paid for by everyone who profits because he succeeds.
Each card answers one question investors and judges keep asking — with concrete numbers from the pilot, not slides.
We didn't build an app, we built a stack. Voice (15+ languages, works on feature phones), open-protocol middle layer (ONDC + AgriStack + OCEN), AI on top. Linear cost per farmer, sub-linear per district. Architected for 100M+ users on day one.
5 revenue streams, none from the farmer: Insurer Evidence API (₹120/claim), Lender Lead-Gen (₹450/lakh), District-as-a-Service (₹35–80L × 766 districts), Marketplace take-rate (1.2–2.5%), optional Premium subscription (₹99/mo, ~12% conversion). ₹78 cr ARR target by year 3.
22% lower groundwater draft in pilot villages (AWD/SRI). 40% less agrochemical via drone-spray precision. Carbon credits per acre via Verra — farmer keeps 75% of value. Climate-resilient crop suggestions when monsoons get erratic. Not marketing — measurable.
(1) Real Telangana Telugu, not textbook Hindi. (2) Every AI answer has a "Why?" button showing the math. (3) Government co-launches — the Sarpanch demos it on stage. Data ownership is the farmer's; we never sell it. DPDP Act 2023 compliant.
Three rails matured in the last 24 months: AgriStack went live (onboarding cost ₹1,200 → ₹14), ONDC opened agriculture, Bhashini hit 92%+ accuracy in 15 Indian languages. Pre-2024 this was a slide deck. Now it's a working product. Window is now.
Real. In production. LightGBM + GraphCast for 1 km × 30 min rain forecasts. EfficientNet-V2 classifies 65 crop diseases. LSTM on pump telemetry. Edge-quantized models run on a ₹6,000 Android in <80 ms — offline. Tested in fields without 4G.
Not Google Ads. FPO partnerships (47 signed, 500 target Y1; CAC ₹14), state govt co-launch via District-as-a-Service, bank-led KCC onboarding, word-of-mouth in Telugu — 1 active user brings 2.7 family signups in 90 days.
Nobody else has built the Operating System. DeHaat is input distribution. Cropin sells SaaS to Walmart. Ninjacart is B2B supply chain. BharatAgri (closest comparison) shut down Nov 2025. Our wedge: integrated 30-module farmer OS, free at base, monetized away from the farmer, on India's public stack.
Horizon 1 (Y1-2): 100K farmers, 8 states, ₹15 cr ARR. Horizon 2 (Y3-4): 5M farmers, 18 states, ₹150 cr ARR, smart-village expansion. Horizon 3 (Y5-7): Pan-India + Bangladesh + Sri Lanka + East Africa. Agriculture intelligence layer for the Global South.
+₹38k/acre/yr median income uplift · post-harvest loss 22% → 8% · PMFBY claim success 80% (vs 40% national) · 41% of users are women (voice removes literacy barrier). 97% say they sleep better.
Five revenue streams. None require the farmer to open his wallet.
| Stream | Customer | Pricing | Year-1 target | Year-3 target |
|---|---|---|---|---|
| Insurer Evidence API | AIC, ICICI Lombard, Bajaj, SBI General | ₹120/claim | ₹0.8 cr | ₹12 cr |
| Lender Lead-Gen | HDFC, ICICI, NABARD coops | ₹450/disbursed lakh | ₹0.4 cr | ₹8 cr |
| District-as-a-Service | State Govt / Collector | ₹35–80 L/district/yr | ₹2.5 cr | ₹38 cr |
| Marketplace take-rate | Input/output transactions | 1.2–2.5% of GMV | ₹0.6 cr | ₹14 cr |
| Premium subscription | ~12% of farmer base | ₹99–299/month | ₹0.3 cr | ₹6 cr |
| Total ARR | — | — | ₹4.6 cr | ₹78 cr |
Gross margin 78% · EBITDA-positive month 22 · CAC ₹140 / LTV ₹6,800 / payback month 3.
100,000 farmers across 8 states. ₹15 cr ARR. Validate the FPO + DaaS playbook across rice, cotton, pulses, horticulture.
5 million farmers, 18 states. ₹150 cr ARR. Same platform extends to rural credit, panchayat governance, women's SHG marketplace, dairy.
Pan-India + Bangladesh, Sri Lanka, Nepal, East Africa. Farmer credit scoring becomes rails for ₹50,000 cr agri lending. Carbon issuance at $40–80/ton.
"Free for the farmer, paid for by everyone who profits when the farmer succeeds."
A hundred and forty-six million farmers in this country wake up tomorrow and bet their family's future on a guess. We can let them keep guessing — or we can give them an operating system. The technology exists. The infrastructure exists. The market is the largest underserved consumer base in the world. The pilot proves the unit economics. This is not a demo project. This is the AI Agriculture Operating System for the next billion farmers — and we're going to build it whether you fund us or not. The only question is whether you want to be on the cap table when it becomes the rails of Indian agriculture.
✉️ Talk to the founder