Why Indian Coaching Institutes Are Still Running on Excel and WhatsApp

Walk into any coaching institute in India at the start of a fee cycle and the same scene plays out. A laptop open to a sprawling spreadsheet. A phone scrolling through hundreds of WhatsApp chats. A pile of paper receipts. And one person (usually the owner, or the one admin everybody trusts) holding the entire financial memory of the business in their head. This is not an edge case. It is the default operating system for an industry that moves enormous amounts of money.

The scale of a market that software forgot

India has roughly 1.4 million registered coaching institutes. Together they collect over Rs.15 lakh crore in fees every year, a figure large enough to rank among the country's biggest informal financial flows. Yet the overwhelming majority of that money is reconciled by hand, in tools never designed for it.

The shape of the market explains why. Around 85% of these institutes are micro-academies with fewer than 500 students. They are the neighbourhood JEE and NEET coaching centres, the spoken-English classes, the commerce tutors, the music and dance academies. They are profitable, sticky, and deeply local, but individually too small to attract the attention of enterprise software vendors, and too specific to be served well by generic tools.

So the segment fell into a gap. Too small for the ERPs that chase universities and large chains. Too particular for the horizontal SaaS that sells the same dashboard to a dentist and a logistics firm. The result is an entire industry running its core revenue process on improvisation.

Why ERPs failed this segment

It is tempting to assume the problem is that owners have not "gone digital." Many have tried. The graveyard of abandoned education ERPs is full of institutes that bought a license, sat through onboarding, and quietly drifted back to Excel within a quarter.

The ERPs failed for three concrete reasons. First, they were priced and built for scale these institutes do not have: implementation fees, annual contracts, and feature sprawl designed for organisations with dedicated IT staff. A 200-student academy does not have an IT department; it has a founder who also teaches.

Second, they modelled the wrong object. Enterprise systems are built around the institution and its hierarchy. Coaching institutes are built around the student and the fee cycle: a student who switches from a weekday batch to a weekend one, drops physics but keeps maths, repeats a year, or carries a sibling discount. Generic systems force this living reality into rigid tables that break the moment a student does anything other than enrol once and pay on time.

Third, they solved everything except the one thing that actually hurts. Attendance modules, exam modules, library modules, and a fee module bolted on as an afterthought, with no automation, no reminders, and no intelligence. The pain was never a missing library catalogue. It was always money: who has paid, who has not, and how to collect without spending evenings chasing parents.

The coaching segment did not reject software. It rejected software that was never designed for how a coaching institute actually runs.

The real workflow, today

Strip away the theory and look at what genuinely happens every cycle. The picture is consistent across thousands of institutes.

Excel is rebuilt from scratch, every single cycle. Fee schedules, installment plans, late fines, and discounts are re-entered by hand each term because last cycle's sheet is a tangle of merged cells and ad-hoc formulas that no longer reflect this cycle's batches. Hours disappear into a task that adds no value. And the sheets are fragile: one wrong row, one dragged formula, and the numbers silently go wrong, usually discovered only when a parent disputes an amount.

Collection happens one WhatsApp message at a time. The admin opens each chat, checks whether that parent has paid, types a reminder, attaches a payment link or UPI ID, and moves to the next. Multiply by several hundred families. There is no single view of who paid, who owes, and who is simply ghosting. The same parent gets reminded twice, or not at all. The "system" is the admin's short-term memory.

One person holds everything. The fee logic, the exceptions, the informal "pay next week" promises, the running mental tally: it all lives with a single admin. There is no audit trail, no handover document, no second source of truth. When that person takes leave, falls ill, or quits, the institute does not lose a file. It loses its financial memory.

Cash flow is invisible until month-end. No owner can answer "what is my collection rate right now?" with confidence. They find out when the month closes, long after the moment they could have acted on a shortfall. Planning is replaced by reacting.

The cost, in numbers

This is not a minor inconvenience tax. Across the segment, institutes spend 15 to 20 percent of staff bandwidth on fee administration, work that should take minutes a month, stretched into days. That bandwidth is the single most expensive resource a small institute has, and it is being spent on copy-paste, reconciliation, and reminder typing instead of teaching, recruiting students, or improving outcomes.

There is a second, quieter cost: lost revenue. Without timely, structured reminders, payments slip, late fees go uncollected because no one tracked them, and defaulters are noticed only after they have defaulted. A few percentage points of leakage on a multi-lakh annual collection is real money, and it compounds, cycle after cycle.

What good actually looks like

The fix is not "a better spreadsheet" or "a discipline drive." It is a different model of how fee operations should work: one built around the student and run on automation.

• A permanent student identity. One persistent profile that survives batch changes, course switches, and academic years. The record follows the student instead of being rebuilt from zero every term.
• Automated fee cycles. Set the templates and installment schedules once. Every cycle thereafter, invoices generate, fines apply, and receipts send themselves, with no manual re-entry, ever.
• AI-timed reminders. Reminders that learn when each parent actually pays and fire at the optimal moment, rather than blasting everyone on day one and hoping. The chasing stops being a human job.
• Defaulter prediction. Payment-history patterns flag the students likely to pay late before the due date, so admins can act early instead of chasing after the fact.
• Real-time collection health. A live dashboard that answers "where do we stand right now?" instantly, turning month-end surprises into mid-cycle decisions.

This is a category gap, not a feature gap

It would be easy to read all of this as a checklist of features some incumbent forgot to ship. It is not. No one owns the fee lifecycle for this segment. The enterprise players ignore it because the accounts are too small. The generic SaaS tools underserve it because they were never modelled around the student and the cycle. And manual processes persist by default because nothing has credibly replaced them.

That distinction matters because it changes what winning requires. Closing a feature gap means shipping one more module. Closing a category gap means owning the entire workflow (onboarding, cycles, reminders, payments, receipts, intelligence) well enough that going back to Excel feels like going back in time.

What changes in the first two hours

The most striking thing about replacing this workflow is how fast the change arrives. This is precisely what Feezy is built to do, and the transformation is measured in hours, not quarters.

In the first onboarding session, students are bulk-imported once (names, batches, courses, fee plans) into permanent profiles. A payment gateway such as Razorpay or Cashfree is connected, and fee templates and installment schedules are configured. That entire setup is designed to be done in under two hours.

From that moment, the manual cycle stops. Fee cycles generate automatically. Reminders fire over WhatsApp at the right time. Fines apply on their own. Receipts send the instant a payment lands on a branded, OTP-secured page. And AI begins surfacing likely defaulters before due dates, while a real-time dashboard shows collection health at a glance.

The before-and-after is stark: an admin who spent days each cycle rebuilding sheets and typing reminders now spends minutes reviewing a dashboard. The single-admin risk dissolves into a system with a full audit trail. And the owner can finally answer, in real time, the only question that ever mattered: how much have we collected, and how much is still owed.

That is the whole point. The work that consumed 15 to 20 percent of an institute's bandwidth does not get optimised. It gets automated away. The fees, as the team likes to put it, start collecting themselves.