Coaching institute management software: what to buy and what to ignore

Coaching software demos all look the same because the demo covers the part every vendor has built. Batches, attendance, fees, a parent app, a test module. Forty minutes, all boxes ticked, and you have learned nothing that separates one from another.

The differences are in three places the demo skips.

First: does the system understand a batch the way you do

A school ERP models a student as belonging to one class and one section. Coaching does not work that way. A student is in the Class 12 Physics batch, the Sunday crash batch, and the weekday doubt session. They joined in September when everyone else started in April. They moved from the evening batch to the morning batch after a month.

Products built for schools handle this by creating the same student three times, which breaks attendance, fee reconciliation and the test history simultaneously. Ask to see one student in three batches with one fee ledger and one performance record. If the answer involves “you can create a group,” keep asking.

Second: does the fee module handle the cases that actually happen

Coaching fee collection is harder than school fee collection, because the amounts are larger, the instalment plans are individually negotiated, and the discount is often the reason the student enrolled at all.

The cases that break systems are the ones the counsellor agreed to in the admission room: a scholarship on the tuition component but not the material component, a sibling concession, a four-instalment plan with the third instalment moved because of the father's posting, a partial refund on withdrawal after two months, and a cheque that bounced in January which the family says they have already settled in cash.

Any of these done outside the system means your receivables number is fiction. Ask the vendor to enter all six, live, during the demo. It is a twenty-minute test and it is the most useful twenty minutes of the evaluation.

Third: what happens to the test series data

This is the one that matters and the one nobody asks about, because the test module looks fine in a demo. Questions go in, students answer, ranks come out.

Here is the question that separates products. Your institute runs weekly tests for nine months. At the end of March, can you pull up a student and see which concepts they have missed repeatedly — not which tests they scored badly in, but which specific ideas have failed in October, again in December and again in February?

Most systems cannot, and the reason is structural. They store a mark against a test. To answer the question you need a mark stored against a concept, which means every question must be tagged to a concept before the test is administered. If that tagging is not in the question-authoring flow, it will never happen, because nobody retro-tags four hundred questions in March.

This matters more than it used to. Institutes built their reputation on selection results and their operations on batch volume. Enrolment is no longer something any institute can take for granted, and free explanation is now abundant and good. What is not abundant is a precise answer to a parent asking what has changed for their child since August. That answer is the product.

What is genuinely table stakes

Do not spend evaluation time on these. Every serious vendor has them, and having them is not a reason to choose anyone.

• Batch creation, rosters, timetables and substitutions.
• Attendance on a phone, marked in under thirty seconds for a batch of sixty.
• Fee invoicing, online payment, receipts and reminders.
• An app or portal where a parent can see attendance and results.
• Test creation, OMR or online administration, ranks and score reports.
• Enquiry and admission tracking.

Two things worth more than they look

Data export. Ask what you get if you leave. Students, fee ledger, attendance and every test result with per-question responses, in a structured format. A vendor who offers PDFs is telling you that your test history is theirs. Nine months of per-question response data is the most valuable thing your institute produces.

Speed on a mid-range phone. Faculty mark attendance between batches, in a corridor, on whatever handset they own. If attendance takes longer in the app than on paper, it will be done on paper and entered on Sunday from memory, and your attendance data will be decorative.

Evaluating in one week

1. Write down your six ugliest fee cases from the last admission cycle. Make every shortlisted vendor enter all six live.
2. Take one real test paper. Ask each vendor to load it, tag the questions, administer it to five students and show you a per-concept report. Not a rank list.
3. Put the app on two faculty phones for a week of real attendance. Ask them afterwards whether they would keep using it. Believe them.
4. Ask for the export file. Open it. If you cannot make sense of it in a spreadsheet, neither can your next vendor.
5. Ask who answers the phone on the morning of a test when nothing loads, and what time they start.

The rest is preference. Pick the one that gets the fee cases right and stores tests at the concept level, and the rest of the feature list will be close enough between vendors that it does not decide anything.

Related reading: what Kota gets right about testing and throws away afterwards and why free explanations changed what an institute can charge for.