A paper-first test that shows which earlier-grade skills each child is missing
TL;DR Gapfinder helps budget private schools find out which Class 3, 4 or 5 skills their Class 6–9 students never learned. Children take a short two-stage paper test, the teacher scans the sheets with a phone, and the school gets level groups and a six-week catch-up plan. Schools pay per student per year.
Middle-school teachers in India teach to the textbook for their grade while a lot of their students sit several grades behind it. In ASER 2024, only 45.8% of Std VIII children in rural India could solve a three-digit by one-digit division, which is a primary-school skill[1]. Std VIII reading is better but still leaves about a third of government-school students below Std II level text[1].
So why don't school exams catch this? The Economic Survey 2025-26 says Indian school assessments are built for certification and promotion, so they rarely produce the diagnostic evidence a teacher needs to fix a gap[2]. A failing mark in Class 7 algebra tells a teacher nothing about whether the child can subtract with borrowing.
What it does, in regulatory terms: a learning diagnostic for schools only. Gapfinder does not tutor students or sell coaching, and it processes children's data as the school's vendor under the DPDP Act and Rules[3].
In week one, every child in Classes 6–9 sits a 20-minute screener on paper covering number sense, operations, fractions and reading comprehension. The teacher photographs the sheets, and the software reads the marks and picks a second, shorter sheet for each child that probes only the weak strand. That's the dynamic part, done without a device per child. The teacher then gets three or four level groups per class, a daily 30-minute catch-up routine for each group, and a retest at week six.
The seed pitched an AI assessment platform, and I've reframed it. Students in budget schools rarely have their own devices, and a diagnosis on its own changes very little. Teaching at the Right Level trials found that training or materials alone didn't move learning, while regrouping by level with a set routine did[4]. So Gapfinder sells the diagnosis and the routine together.
The user is the class teacher. The payer is the owner or principal of a low- or mid-fee private school, who needs results to show parents at admission time. The test price is a flat fee per student per year covering two diagnostic rounds, and that's an assumption you'll need to validate. The best-known adaptive maths product does not publish school prices[5], and I found no public price for diagnostic tests sold to Indian schools.
The Economic Survey 2025-26, tabled in January 2026, asks the system to go past aggregate scores and find the specific gaps and confusions behind them[2]. PARAKH Rashtriya Sarvekshan 2024 tested 21.15 lakh students in Grades 3, 6 and 9[2], which gets parents and schools used to talking about grade-level skills. And the DPDP Rules, notified on 14 November 2025, give vendors a clear consent framework to build to[3].
There's no public figure for school spending on diagnostic assessment in India. The bottom-up pool is big though - private schools make up 26% of India's 14.71 lakh schools and enrol 41% of 24.69 crore students, according to UDISE+ 2024-25 as cited in the Economic Survey[2]. Not all of them are budget schools, and I found no split by fee level.
An Indian assessment company's adaptive maths software is the best-known product here. A Delhi trial found students offered it scored 0.37 standard deviations higher in maths over 4.5 months[6], but it needs a screen per learner. A large education nonprofit's level-based teaching method is free and runs through governments, mostly for primary grades[4]. PARAKH measures states and districts, not individual children. The gap is a cheap, paper-based, per-child diagnosis for the middle grades.
| Difficulty | Moderate (39/100) |
| Starting capital | ₹45–85 L |
| Time to launch | 5–8 months |
Reading a phone photo of a marked sheet is solved technology. The hard work is the item bank, meaning hundreds of questions that each isolate one prerequisite skill, tested in Hindi and two or three other languages. The second hard part is teacher behaviour, because the catch-up routine only works if teachers actually run it every day for six weeks.
No education licence is needed to sell a diagnostic to schools. The weight sits on children's data.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't trust AI to label a child as having a learning disability or to decide promotion. It flags skill gaps, and teachers make the calls.
Want to work out the AI layer for Gapfinder? Talk to Yash →
I'd back a founder who has taught in or run a low-fee school and can write good test items. A pure software founder will struggle with the classroom side, no two ways about it. Before raising money, prove one thing in 10 schools - that the week-six retest shows clear movement in at least half of the lowest group.
Plan for roughly ₹45–85 L. Most of it goes into building and trialling the item bank in several languages and into teacher support.
No education licence. You need company and GST registration, and you must follow the DPDP Act and Rules on children's data, ideally as the school's processor.
Yes, wherever middle-school students sit grades behind their textbook. Rebuild the item bank for the local curriculum and check your own child data protection law.
The free Gapfinder validation kit has the line-by-line capital breakdown, the effort scoring, the child-data consent checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for school owners and teachers, and 5 assumptions to test. Download the kit →
Thinking about building Gapfinder or something close to it? The make-or-break work is getting teachers to run the catch-up routine after the report arrives, and I'm happy to work through that with you. 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.