Ward-level site plans for rain gardens that clear known waterlogging spots
TL;DR Soakmap tells a city exactly where to put rain gardens, soak pits and park storage so that a known waterlogging street stays passable after ordinary rain. It sells costed, ward-by-ward site plans to municipal engineers and to the consultants who write their project reports.
Indian cities flood after storms that are not extreme. Chennai's 2015 floods caused insured losses of ₹10,500 Cr, and the 2005 Mumbai flood did more than ₹13,500 Cr of damage[1]. The everyday cost is quieter - the same underpasses and junctions go under water every monsoon. Delhi's master drainage plan was last prepared in 1976[2], so drains were sized for a city with far less concrete. Widening drains is slow and costly, and in dense wards there is often no room. Cities need water to soak in or wait before it reaches the drain.
Soakmap starts with a city's own list of waterlogging spots. For each one it traces the small catchment that drains to it, then ranks every usable patch of public land inside that catchment, like road medians, park edges, school grounds and footpath verges. It sizes a rain garden, soak pit or shallow storage area for each, estimates how much runoff each can hold, and adds them up until the spot should clear a typical storm. The output is a site list with sketches and cost bands that an engineer can drop into a project report.
I narrowed the seed, which promised green infrastructure that tackles urban flooding in general. The evidence says that is too much. A US study of bioretention planters found they held back all the runoff in 38% of storms but released water in 96% of the larger events[3]. So Soakmap targets the frequent, nuisance flooding where soak-and-hold features work, and says plainly that big floods still need drains.
Residents and commuters benefit. The payer is a municipal corporation or smart-city company, usually through the consultant preparing its project report, at a fee per ward. A second buyer is any large campus or tech park that floods its own roads, and it pays directly and decides faster. No public price exists for this kind of plan, so test prices in the pilot.
Cities have more money for water work than before. AMRUT 2.0 carries an indicative outlay of ₹2,99,000 Cr and has approved 3,032 water bodies for rejuvenation[4]. Since 2025, cities from Gurugram to Vijayawada began working with flood-modelling start-ups and drone surveyors[2][5]. Engineers who now have flood maps still need to know what to build and where.
There is no public figure for what Indian cities spend on planning green-blue infrastructure. The nearest neighbour - AMRUT lists storm-water drainage as a funded component and has sanctioned ₹2.73 lakh crore of projects across water, sewerage and drainage over ten years[4]. That is context for a small planning slice, not your market.
A flood-modelling start-up incubated at IIT Gandhinagar models where floods will happen and how deep, and Gurugram and Indian Railways use it[5]. A Bengaluru drone-survey firm maps flood-prone zones from the air for Bengaluru and Vijayawada[2]. Both answer "where does it flood". The free US EPA SWMM model can test low-impact designs[6], but someone has to feed it local data. Nobody sells the next step, which is a buildable, costed list of soak sites per waterlogging spot.
| Difficulty | Moderate (44/100) |
| Starting capital | ₹60 L–1.1 Cr |
| Time to launch | 6–9 months |
Getting usable data is the first hurdle, since drain maps are often missing or out of date, and land ownership in a ward is messy. The second is credibility, because engineers will not trust a runoff estimate until one built site has worked through a monsoon. Selling to cities is slow and often runs through consultants who want the credit.
No sector licence applies to selling plans, but public works bring their own checks.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't trust a model's runoff estimate without a built site that has worked through a monsoon. Engineers sign the design, not the software.
Want to work out the AI layer for Soakmap? Talk to Yash →
Here's what can go south -
I think this suits you if you have hydrology or civil engineering training and actually enjoy fieldwork and slow government sales. If you're hoping to sell software without walking flooded streets then it's a poor fit. Before raising money, get one ward's sites built and show the spot staying clear through one monsoon.
Plan for roughly ₹60 L–1.1 Cr, mostly engineers, fieldwork, data and a long city sales cycle.
No sector licence. Designs for public works need a qualified civil engineer to sign off, and you'll usually sell through tenders or consultant panels.
Yes, many cities fund green infrastructure. Check local engineering sign-off rules and procurement first.
The free Soakmap validation kit has the line-by-line capital breakdown, the effort scoring, the procurement and approvals checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for city engineers and campus managers, and 5 assumptions to test. Download the kit →
Thinking about building Soakmap or something close to it? The make-or-break work is getting one ward's sites built and measured before you sell the next city, and that's the kind of plan I help founders put together. 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.