School Software Audit: 12 Checks Before the New Session [2026]

<h2>Your school software passed the demo. Nobody checked it since.</h2> <p>Every Indian school runs the same sequence: demo, quotation, approval, a frantic two-week setup, then silence. Two years later the accounts clerk keeps a parallel Excel sheet "just in case", teachers mark attendance on paper first, and nobody has opened the transport module since training day.</p> <p>None of that shows up as a complaint. The software works — it is just not doing the job you pay it for. A <strong>school software audit</strong> is the annual 90 minutes that closes that gap. It is not a renewal negotiation — price and notice periods are a separate exercise (see the <a href="/blog/post/school-software-renewal-india-7-checks-before-you-renew-2026/">renewal checklist</a>). It answers one question: <em>is the system earning its place in your school this year?</em></p> <h2>Why an annual school software audit beats a vendor review</h2> <p>Most principals review school software only when something breaks or a bill arrives. That timing guarantees two bad outcomes: you judge a year's use by one outage, and you decide under deadline pressure.</p> <p>An annual audit flips the order. Collect evidence at a calm point in the Indian school calendar — February to March, before admissions close and new-session data entry begins — then decide with a full year of numbers in front of you. Mid-year works too — after the Diwali break.</p> <p>Schools also change shape: 640 students in 2024, 900 and a second campus in 2026. Software configured for the old shape does not fail loudly; it quietly stops matching reality.</p> <h2>The 12 checks, in one table</h2> <p>Run these in order. Anything you cannot evidence scores as a fail, not "probably fine".</p> <table> <thead> <tr><th>#</th><th>Check</th><th>The test</th><th>Fail signal</th><th>Owner</th></tr> </thead> <tbody> <tr><td>1</td><td>Student master</td><td>Compare 20 random students: name, DOB, parent mobile, section, bus stop</td><td>2+ mismatches — one wrong number is one parent never alerted</td><td>Office</td></tr> <tr><td>2</td><td>Fee ledger</td><td>Does software outstanding match the accounts register to the rupee?</td><td>Any variance explained by "we adjust manually"</td><td>Accounts</td></tr> <tr><td>3</td><td>Attendance truth</td><td>Ask three teachers which record is official on a disputed absence</td><td>Any answer other than "the app"</td><td>Coordinator</td></tr> <tr><td>4</td><td>Real logins</td><td>Logins per role over four weeks, not four days</td><td>Under 50% of teachers weekly; zero HOD logins</td><td>Principal</td></tr> <tr><td>5</td><td>Parent reach</td><td>% of families with an install and one open in 30 days</td><td>Under 60% — a third of circulars never arrive</td><td>Front desk</td></tr> <tr><td>6</td><td>Module graveyard</td><td>Mark every paid module used weekly / monthly / never</td><td>3+ modules at "never" for two terms</td><td>IT</td></tr> <tr><td>7</td><td>Cost per student</td><td>Total annual spend ÷ enrolment, add-ons included</td><td>You cannot state it without three invoices</td><td>Accounts</td></tr> <tr><td>8</td><td>Hours returned</td><td>Ask accounts and front desk: hours per task, now vs before</td><td>Any task slower than the manual version</td><td>Principal</td></tr> <tr><td>9</td><td>Invoice match</td><td>Tick last year's invoice lines against what you used</td><td>Over 20% of spend on unused items</td><td>Accounts</td></tr> <tr><td>10</td><td>Export test</td><td>Export students, fees, attendance to CSV yourself</td><td>You need a support ticket for your own data</td><td>IT</td></tr> <tr><td>11</td><td>Access + consent</td><td>Show who opened a minor's record, and the consent basis</td><td>No audit log, or no documented parental consent</td><td>Principal</td></tr> <tr><td>12</td><td>Support reality</td><td>Tickets raised, median first response, unresolved count</td><td>Over 24-hour median, or closed-without-fix tickets</td><td>IT</td></tr> </tbody> </table> <h2>Checks 1–3: data you can actually trust</h2> <p>Almost every downstream problem traces back to the student master. A transposed mobile digit is not a data-entry inconvenience; it is a fee reminder that never arrives and a bus-change message sent to a stranger. The 20-student sample takes eleven minutes — the highest-yield test here.</p> <p>Fee reconciliation exposes parallel systems fastest. If the software total and the accounts register differ, one is fiction — usually the software holds the "official" number while a notebook holds the real one, and a human reconciles the gap monthly. That human is your hidden cost.</p> <p>Attendance is a culture question disguised as a technical one. If teachers mark paper first, the app is a reporting tool, not a record — survivable for a term, corrosive over three years, because analytics end up built on data nobody trusted enough to enter live.</p> <h2>Checks 4–6: adoption is not a login count</h2> <p>Vendors report "95% adoption" because 95% of accounts exist. You want weekly active use by role — a system used only by the accounts office is a billing tool with a school attached.</p> <p>Parent reach deserves its own check. If a third of families never open the app, every circular you "sent" reached two-thirds of the school. Paper notices are not a failure of parent discipline; they are the app failing a usability test in a Tier-2 city, where a 40 MB update is a real barrier. (Full playbook: <a href="/blog/post/school-parent-app-adoption-india-8-fixes-that-work-2026/">8 fixes for parent app adoption</a>.)</p> <p>The module graveyard is where money disappears. Schools buy the full suite because the per-module difference looks small, then never touch hostel, alumni or library. "Never" for two terms is a line item to delete, not a training problem.</p> <h2>Checks 7–9: the money check most schools skip</h2> <p>Cost per student is the only software number a committee can compare across years and vendors. Compute it honestly, including what never appears on the software invoice.</p> <table> <thead> <tr><th>Cost line</th><th>How to compute</th><th>What it exposes</th></tr> </thead> <tbody> <tr><td>Subscription</td><td>Annual invoice ÷ current enrolment</td><td>Whether you pay for last year's headcount</td></tr> <tr><td>SMS / WhatsApp credits</td><td>Annual top-ups ÷ enrolment</td><td>Whether alerts are sent, or stockpiled</td></tr> <tr><td>Hardware + AMC</td><td>Biometric, RFID, printers, server, annual AMC</td><td>True cost of the "cheap" on-premise option</td></tr> <tr><td>Staff time</td><td>Hours/week × 48 weeks × hourly cost, for duplicated tasks</td><td>The parallel-register tax — usually the biggest line</td></tr> <tr><td>Training</td><td>Vendor days + internal hours per new joiner</td><td>Whether onboarding is a product or a project</td></tr> </tbody> </table> <p>A 900-student school paying ₹1.2 lakh a year sits at roughly ₹133 per student — less than one notebook set. That same school burning 14 staff-hours a week on duplicate entry pays several times that, and the second number never appears in a quotation. That is the number to surface.</p> <h2>Checks 10–12: risk, exit and support</h2> <p>Export your own data, yourself, without a ticket. An ERP that cannot hand you a clean CSV is a system you rent with extra steps. Test students, fees and attendance in February, when the vendor is motivated — not in June, when you are not.</p> <p>Access and consent is now a compliance question. Under India's <a href="https://www.meity.gov.in/data-protection-framework" target="_blank" rel="noopener">Digital Personal Data Protection Act, 2023</a>, a school processing a child's data needs verifiable parental consent and a stated purpose. If nobody can show who opened a minor's record last month, lead the report with that. Our <a href="/blog/post/school-software-data-security-checklist-india-dpdp-act-2023-2026/">DPDP security checklist</a> covers the vendor questionnaire.</p> <p>Support is rarely measured. Ask for last year's ticket export: raised, median first response, closed without a fix. If the vendor will not share it, that is your answer. The <a href="https://dashboard.udiseplus.gov.in/" target="_blank" rel="noopener">UDISE+ dashboard</a> benchmarks enrolment against district trends; no dataset tells you whether your vendor answers the phone on a Saturday.</p> <h2>Scoring your audit</h2> <p>Score each check 0, 1 or 2 — no half points. Which group is weakest matters more than the total: each points to a different decision.</p> <table> <thead> <tr><th>Score (of 24)</th><th>What it means</th><th>Decision</th></tr> </thead> <tbody> <tr><td>20–24</td><td>Doing its job; two weak checks</td><td>Continue, re-audit next February</td></tr> <tr><td>14–19</td><td>Working, with a parallel-system tax and unused spend</td><td>Renegotiate scope, retrain two processes</td></tr> <tr><td>8–13</td><td>You pay for software the school has quietly rejected</td><td>Costed comparison of three vendors before renewal</td></tr> <tr><td>0–7</td><td>Data and adoption both broken; only the invoice works</td><td>Start a switch evaluation with a 90-day exit window</td></tr> </tbody> </table> <p>If checks 1–3 are red, do not switch yet — migration carries the mess across and you will audit the same failures next year. Fix the master data first. Our guide on <a href="/blog/post/when-to-replace-school-erp-in-india-6-signs-its-time-2026/">when to replace school ERP</a> sets the threshold where switching beats repair.</p> <h2>Who runs it, and when</h2> <table> <thead> <tr><th>Week</th><th>Activity</th><th>Who</th><th>Time</th></tr> </thead> <tbody> <tr><td>1</td><td>Collect evidence: student sample, fee ledger, invoice, ticket export, login report</td><td>Office, accounts, IT</td><td>3 hrs</td></tr> <tr><td>2</td><td>Score all 12 checks in one sitting, evidence recorded against each</td><td>Principal, accounts, IT</td><td>90 min</td></tr> <tr><td>3</td><td>Fix the three cheapest wins: data correction and one retraining session</td><td>Check owners</td><td>1 week</td></tr> <tr><td>4</td><td>Decide: continue, renegotiate scope, or evaluate alternatives — with the score in writing</td><td>Management committee</td><td>1 meeting</td></tr> </tbody> </table> <h2>What a good audit looks like a year later</h2> <p>The goal is not a perfect 24. It is a school where a disputed absence has one answer, a fee total reconciles first time, a parent in Baddi or Indore gets a circular the day it is written, and cost per student is a number the committee knows.</p> <p>That is what a school management system should buy — not a dashboard nobody opens. If this year's score lands in the bottom half, the fix is usually tighter scope and better onboarding, not a bigger suite. See how <a href="/saksham/">Saksham AI</a> covers the modules Indian schools use daily, check implementation cost on our <a href="/pricing/">pricing page</a>, or <a href="/#demo">book a free demo</a> with your scorecard — we will show which of the 12 checks the system genuinely closes.</p>