Offline virtual practicals for Class 9–10 science, in the state's own language
TL;DR Prackit packages the practicals on each state board's Class 9–10 science list as simulations a teacher runs offline on the school's computer or smart classroom, in the local language. It records which step each class gets wrong. State education departments and CSR funders pay per school per year.
A lot of Indian students reach their board exams having read about experiments they never actually did. India has 14.71 lakh schools, 69% of them run by government[1]. The PM SHRI scheme is upgrading schools with integrated science labs, but its target is about 14,500 schools[1], which by my arithmetic is under 1% of the total. I couldn't retrieve the UDISE+ figure for science labs during research, so check it on the UDISE+ portal before relying on any number.
What has changed is screens. Schools with a computer rose from 35% in 2018-19 to 65% in 2024-25, and those with internet from 19% to 64%[1]. So plenty of schools now have a screen and no lab, and the content that would fill that gap is rarely in the student's language or matched to the board's practical list.
What it does, in regulatory terms: classroom teaching software only. Prackit makes no exam or certification claims, and by default it stores no student names, so children's data stays minimal under the DPDP Rules[2].
Each simulation follows one practical from the board's list, such as finding the focal length of a concave mirror or testing for starch in a leaf, and uses the same apparatus, steps and readings. The teacher runs it on a projector and then calls students up in turns, or pairs use the computer lab. Everything installs once and works without internet. Each run logs where a class went wrong, like misreading a scale or skipping a control. Teachers get a weekly summary and districts get a dashboard.
The seed pitched data-driven virtual labs for students, and I've reframed it. Good free simulations already exist - PhET offers research-based ones under 48 language options[3]. Selling more simulations to students means competing with free. A package matched to each board, in its language, working offline and run by the teacher is something a state can actually buy.
The users are science teachers and their students. The payers are state education departments, through ICT or science budgets, and CSR funders adopting districts. The price to test is a yearly fee per school with teacher training included, and that's my assumption. For comparison, the leading global virtual-lab vendor's college contracts typically start at US$5,000 a year[4], about ₹4.8 L, which no Indian state will pay per school.
The jump to 65% of schools with computers happened between 2018-19 and 2024-25[1], so hardware isn't the main gap any more. PM SHRI shows the government will fund science labs, but only for a small share of schools[1]. And the DPDP Rules, notified on 14 November 2025, tell vendors what consent children's data needs[2].
There's no public figure for what Indian states spend on school virtual labs. A demand signal comes from higher education - the Ministry of Education's Virtual Labs site reports 8,560,251 participants and 2,055 nodal centres[5], though that serves engineering colleges. Broader context is 24.69 crore school students[1], which is not your market.
PhET from the University of Colorado is free and widely translated, but it isn't matched to Indian board practical lists[3]. Virtual Labs, run by IIT Delhi with other institutes under the Ministry of Education, targets engineering students[5]. Commercial virtual-lab vendors sell to colleges at institutional prices[4]. The gap is a school-level, board-matched, offline product with class data that a state can roll out.
| Difficulty | Moderate (43/100) |
| Starting capital | ₹60 L–1.2 Cr |
| Time to launch | 6–9 months |
Building 40 good simulations takes months of careful work with science teachers. The hardest part is selling to government, with its tenders, long payment cycles and a pilot district that wants proof first. You also need training, because teachers who never ran the real practical may not run the virtual one either.
No licence is needed to make classroom software. Selling it to a state is where the paperwork starts.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't trust AI to write the science itself. Every apparatus step, reading and answer needs a subject teacher's sign-off.
Want to work out the AI layer for Prackit? Talk to Yash →
I think this makes sense for a founder with science-teaching depth and experience selling to state education departments or CSR foundations. If you're expecting consumer revenue then it's the wrong idea for you. First, prove that teachers in 20 pilot schools run at least two practicals a month without being chased.
Plan for roughly ₹60 L–1.2 Cr. Building around 40 board-matched simulations and surviving slow government payments take most of it.
No licence to build the software, but you'll need GeM or tender registration to sell to states, plus DPDP compliance for any children's data.
Yes, wherever schools have screens but no labs. Match the simulations to the local syllabus and language, and check the local procurement rules.
The free Prackit validation kit has the line-by-line capital breakdown, the effort scoring, the procurement and child-data checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for teachers and education officials, and 5 assumptions to test. Download the kit →
Thinking about building Prackit or something close to it? The make-or-break work is winning a first state or CSR contract without running out of cash during the wait, and I've seen plenty of founders get that timing wrong. 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.