Camera pheromone traps and spray alerts for cotton programmes, run as a service
TL;DR Mothtrap installs solar camera traps that count pink bollworm moths every hour and turns the counts into village-level spray alerts. Cotton programmes, ginners and seed companies pay per trap per season to cover their farmer groups. A government pilot cut sprays by more than a third, but it reaches only a few dozen villages.
Pink bollworm has beaten Bt cotton. Resistance to the first Bt trait showed up in Gujarat in 2010 and to the two-gene trait by 2014[3]. One environment magazine reports average yield losses of at least 20–25%, with severe attacks across Punjab, Haryana and Rajasthan[2]. Pest control already eats up over 43% of what Indian farmers spend to grow cotton[3].
Pheromone traps are cheap, and a lure and funnel trap sells for ₹130 online[5]. Reading them is the trouble. Farmers usually set five traps per hectare, yet the head of ICAR-CICR told that magazine that farmers rarely count what the traps catch[2]. Without counts people spray on a calendar or after the damage shows, which is too late.
What it does: pest monitoring and advisory only. Mothtrap does not sell pesticides, and its alerts point only to products labelled for cotton.
I reframed the seed, which pitched integrated analytics against crop yield loss. That market is crowded and vague. One pest in one crop, with a proven sensor design, gives a founder something to deliver and measure.
Each trap carries a pheromone lure, a sticky liner, a camera, a small computer and a solar panel. It photographs the liner every hour and a trained model counts the moths. ICAR-CICR has shown this design works in the field[1]. Mothtrap's job is the unglamorous part research projects skip - installing traps across a buyer's villages before sowing, changing liners every two weeks and lures monthly[1] and fixing dead units within days. When counts cross the threshold it sends a short alert in Punjabi, Hindi, Marathi or Telugu. A season-end report shows catches, sprays and boll damage against nearby villages without traps.
Farmers get the alerts for free. The payers are organisations that answer for a group of cotton farmers - sustainability programmes that need proof of fewer sprays, ginners and spinners buying traceable cotton, seed companies whose brands suffer when bollworm hits, and state agriculture departments. The per-trap season fee is something to test, since I found no public price. CICR's own report names subscriptions and data sales to agri firms as revenue routes[1].
CICR's 2024–25 Punjab pilot cut sprays from an average of 10.1 to 6.2 and held boll damage under 10%, against 30–65% in earlier years[1]. The Union agriculture minister formally launched the trap, and the network grew to 40 villages in three states in 2025–26 with central government support[1][4]. That still leaves most of India's 130.61 lakh hectares of cotton uncovered[3].
There's no public figure for spending on cotton pest surveillance in India. The narrowest firm numbers are agronomic, a 38.6% cut in pesticide use in the pilot[1] and 130.61 lakh hectares under cotton[3]. The public network covers 40 villages[1]. Will private buyers pay to close that gap? That's the question to test.
ICAR-CICR runs the AI Smart Trap network with the plant protection directorate, free to farmers but limited to selected districts[1]. Plain pheromone traps are sold online by small agri-input sellers[5]. Advisory apps identify pests from photos and sell inputs[6]. Camera traps are sold abroad, by a European firm for one. I couldn't find an Indian firm running camera traps for cotton programmes as a paid seasonal service.
| Difficulty | Moderate (50/100) |
| Starting capital | ₹1.2–2.2 Cr |
| Time to launch | 6–9 months |
The model and trap design already exist. The hard parts are field upkeep across hundreds of traps in heat and dust, and keeping counts accurate when other moths and debris land on the liner. Sales are seasonal too, and if you miss the budget cycle before sowing then you lose a whole year.
No sector licence for monitoring, but the traps and the alerts bring a few checks.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't let AI choose the pesticide or dose. Alerts say the threshold is crossed, and labelled products and an agronomist decide the rest.
Want to work out the AI layer for Mothtrap? Talk to Yash →
I'd back a founder with hardware field-service experience and contacts in the cotton trade or seed industry. It's not for someone who wants to sell devices and walk away. Prove first that one paying buyer will fund 100 traps for a season, and that alerted villages spray at least 25% less than comparison villages.
Plan for roughly ₹1.2–2.2 Cr for traps, field upkeep and the alert system. A 100-trap season for one paying buyer is the first test.
You need WPC radio approval for connected traps, and alerts must stick to products labelled for cotton under the Insecticides Act. Add the usual company registration and GST.
Yes, wherever pink bollworm or similar pests hurt cotton. Check local radio approval and pesticide advice rules first.
The free Mothtrap validation kit has the line-by-line capital breakdown, the effort scoring, the radio and data-protection checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for cotton buyers and farmers, and 5 assumptions to test. Download the kit →
Thinking about building Mothtrap or something close to it? The make-or-break work is keeping hundreds of field traps running all season at a cost buyers accept, and that's the kind of operations puzzle I enjoy. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.