Start with who has to use it every day
School software is bought by a correspondent or a principal and used by teachers, parents and students. Those are different people with different tolerances. A system that impresses in a demo and takes a teacher three minutes to mark attendance will be abandoned within a term, and the data it holds becomes fiction shortly after.
The single best predictor of whether a school ERP survives is whether the daily actions — marking a register, publishing a circular, checking a fee balance — take seconds. Ask to see those three specific things during the demo, timed, on a phone. Not the dashboard.
Ask what happens to your data
Under India's DPDP framework, a school is responsible for the personal data of children in its care, and "our vendor handles it" is not a defence. Ask three concrete questions: where is the data physically stored, is personal data encrypted at the field level or only in transit, and how is one school's data separated from another's on shared infrastructure.
"We use row-level security" and "each school has its own database schema" are meaningfully different answers, and the second is much harder to get wrong. SkoolBook gives each school its own Postgres schema, applies row-level security on top of that, encrypts personal data at the field level, and keeps data India-resident by default.
Separate what ships from what's on a slide
The most expensive mistake in this category is discovering during implementation that a module on the feature list is a placeholder. It is also completely avoidable: ask the vendor to demonstrate each module you care about on live data, not on a slide, and ask directly which listed features are not yet built.
A vendor who answers that question honestly is telling you something valuable about how they will behave when something breaks in year two. For our part: SkoolBook does not yet have online fee payment collection, WhatsApp or SMS delivery for announcements, biometric or RFID attendance, board-specific report-card templates, or an LMS with online examinations. Everything else on this site is live and demonstrable.
Understand what onboarding really costs
Ask how long it takes to bring up a new school, and what happens to your existing data. Systems that require a per-school deployment take weeks and make every subsequent change slow; systems where onboarding is a data operation on shared infrastructure take minutes.
Also ask what happens if you leave. Data portability is boring right up until the moment you need it, and a vendor who has never been asked will not have an answer.
Check the unit economics you're being sold
If a quote seems implausibly cheap for the feature list — particularly one including AI — ask how it is funded. Uncapped AI on a low per-student price is not a bargain; it is a promise that gets withdrawn, throttled, or repriced once usage is real.
SkoolBook meters AI with a fair-use session cap per student per month, on every plan including Enterprise. That is a deliberately unglamorous answer, and it is why the number quoted in year one is still the number in year three.
The checklist, short version
- Time the three daily actions
Mark a register, publish a circular, check a fee balance — on a phone, timed. If any takes over 30 seconds, teachers will stop using it.
- Ask where the data lives
Physical location, field-level encryption, and how one school is isolated from another. A schema per school beats a shared table with a filter.
- Ask what isn't built yet
Every vendor has gaps. The ones who name them are the ones you can plan around.
- Ask to see it on live data
Not a slide, not a sandbox with three perfect students. A real school's messy data.
- Ask how AI is metered
Uncapped AI at a low per-student price is a repricing waiting to happen.
- Ask how you'd leave
Data export, in a format something else can read. Ask before you sign, not after.
Frequently asked
How much should school management software cost in India?
It is normally priced per student per year, and the honest answer is that the range is wide enough that a number without context is meaningless — it depends on your headcount, which modules you need, and whether AI features are included. What matters more than the headline figure is whether the price is sustainable for the vendor: an unprofitable vendor reprices or disappears. We quote per school rather than publishing a number, because the same figure would be wrong for a 200-student school and a 3,000-student one.
What's the difference between a school ERP and a school management system?
In practice, nothing — the terms are used interchangeably in India. "ERP" tends to signal the administrative side (fees, HR, inventory) while "management system" or "SIS" leans academic (students, attendance, marks). Any serious product covers both.
Do we need a mobile app, or is a web portal enough?
For parents, an app. A web portal that parents must remember to log into will be checked at report-card time and never otherwise, which defeats the purpose of communication and transport tracking. SkoolBook ships a mobile app for families alongside the web portal.
Can we migrate from our existing school ERP?
Usually yes — student, staff and structural data can be brought across. What rarely migrates cleanly is historical assessment data in a proprietary format. Ask specifically about that during evaluation rather than assuming it.