Season-ahead water shortfall outlook and backup plan for each factory site
TL;DR Thirstline gives a manufacturer a monthly forecast of whether each of its plants will run short of water in the next three to six months, and what to do about it. It joins the plant's own borewell and tanker data with public groundwater and reservoir figures. Multi-plant companies pay per site.
Factories in India draw on groundwater that is shrinking unevenly. The 2025 national assessment classed 730 of 6,762 blocks as over-exploited, with a further 201 critical[1]. Rules are tightening too, and groundwater guidelines now impose penalties for drawing without a no-objection certificate and bar new large industries in over-exploited areas[1]. Coca-Cola's own plan lists nine Indian plants in high or extremely high water stress and puts the possible rise in yearly water costs at US$180,000–2.7 million (about ₹1.7–26 Cr)[2]. Yet most plant managers only learn of a shortage when the borewell yield drops or tanker prices jump.
Each month Thirstline produces a one-page outlook per plant - likely supply against demand for the next three to six months, the main reason for any gap, and the action to take now. It uses the plant's borewell levels and meter readings, the official status of its groundwater block, storage in the reservoir that feeds the local supply, and the seasonal rainfall outlook. Simple models learn how each plant's borewells respond to rain and pumping. When risk rises, the outlook triggers a pre-agreed plan to book tankers early, shift water-heavy batches, or fix recycling first.
I narrowed the seed, which pitched drought prediction in general. District drought monitoring is public work and free, and nobody pays a start-up for it. A plant that stops for lack of water loses real money, so that is where a paid forecast makes sense.
The plant head uses it. The payer is the corporate head of EHS or sustainability at a company with several plants in dry regions, like beverages, textiles, pharmaceuticals and food processing. They already buy water metering from industrial water-monitoring firms[3] and report water data each year. Price it per site per year. No public benchmark exists, so test it.
Disclosure and enforcement both moved. SEBI made the BRSR sustainability report mandatory for the top 1,000 listed companies from 2022-23[4], and its 2023 BRSR Core adds assurance and value-chain reporting[5]. Reuters reported in January 2026 that a court had directed authorities to enforce the ban on new industrial water permits in over-exploited areas around Alwar[2]. Water risk is now on the board agenda, not just the plant's.
There is no public figure for spending on site-level water-shortage forecasting in India. The firmest proxies are counts - 1,000 listed companies file BRSR[4], and 931 blocks are over-exploited or critical[1]. One such firm says more than 120 companies use its water monitoring[3], which shows plants will pay for water data.
Industrial water-monitoring firms sell sensors and a dashboard that track water use inside industrial sites[3]. WRI's Aqueduct atlas gives a free basin-level picture of water stress[6], and many companies cite it in reports. Neither tells a plant manager what supply will look like next April. The gap is a forward, site-specific outlook tied to an action plan, and it could sit on top of metering a company already has.
| Difficulty | Moderate (40/100) |
| Starting capital | ₹60 L–1.2 Cr |
| Time to launch | 6–9 months |
The hardest part is forecasting skill. Borewell behaviour varies from site to site, public data arrives late and coarse, and a forecast that misses one dry season loses the customer. You need a year of a plant's own data before the outlook is sharp, so early contracts must be priced as pilots.
No licence is needed to sell water forecasts. Don't present them as certified compliance advice, though.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't trust a forecast from a site with less than a season of its own data, and keep plant shutdown decisions with the plant team.
Want to work out the AI layer for Thirstline? Talk to Yash →
Here's what can go south -
I think this fits you if you have hydrogeology or data-science skills and can also sell to corporate sustainability teams. You'll need patience too, because skill only shows after a full season. First prove that your outlook would have called last year's shortfall at three real plants, using their own records.
Plan for roughly ₹60 L–1.2 Cr, mostly data science, field data setup and pilots priced low while the models prove themselves.
No. Your customers handle CGWA groundwater permissions and SEBI disclosures, and you need company and GST registration.
Yes, factories in water-stressed regions face the same risk. Local groundwater data quality decides how hard it is.
The free Thirstline validation kit has the line-by-line capital breakdown, the effort scoring, the groundwater and disclosure rules checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for plant and sustainability heads, and 5 assumptions to test. Download the kit →
Thinking about building Thirstline or something close to it? The make-or-break work is showing your forecast would have caught a real past shortage at a real plant, and that's the kind of test I help founders design. 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.