Student Insight Upload Your Filled Sheet
Drop your class's filled Excel file below. Managing more than one section or batch? Drop 2 or more files at once — matching ones are compared automatically, and any that don't match still get their own individual analysis.
Click to browse or drag & drop your filled Excel file(s)
.xlsx / .xls · drop 2+ to compare sections · never leaves your browser
How would you like to start?
Create New Template
Start fresh — set up institution, class, subjects and tests, then download a blank workbook to fill offline.
Update Existing Template
Already filled Test 1? Load your workbook, add Test 2 / Test 3, then re-download — existing marks are kept, only new blank columns are added.
Initialising…
Step 1 of 10
0%
Crunching numbers in your browser — this only takes a few seconds.
Subject Averages
Class Trend
Marks vs Attendance
Performance Heatmap — Student × Subject
Excellent ≥85%
Good 70–84%
Satisfactory 50–69%
Needs Support
Early Warning Flags
Class Wellbeing & Stress Indicators
Attendance vs. Performance
Subject Weakness (Class-Wide)
Subject Correlation
₹0
Cost, forever
0
Servers involved
100%
Offline capable
MIT
Open source licence
Your Data Always Belongs to You▾
Student Insight never treats the browser as permanent storage.
Your data is stored only in files that you own.
Every project can be imported, analyzed, updated, exported, shared, archived, or backed up without depending on an online account or remote server.
If you move to another computer, your project moves with you. If you disconnect from the internet, your project continues to work.
Your data remains yours.
Built for GitHub Pages and Static Hosting▾
Student Insight is intentionally designed to run from a single static website. It works equally well whether it is opened from:
GitHub Pages
A local HTML file
A USB drive
A school intranet
An offline classroom computer
No installation. No server. No database. No subscriptions. No vendor lock-in.
Privacy by Design▾
Educational records are among the most sensitive types of information.
Student Insight is designed so that student information never needs to leave the educator's device. The application does not require:
✕ User accounts
✕ Cloud synchronization
✕ Remote databases
✕ Cookies or fingerprinting
✕ Personal data collection
✕ Background uploads of your files
The browser simply processes the data that you choose to open — your uploaded spreadsheets and student records never leave your device.
To understand how many people use Student Insight, the site uses Cloudflare Web Analytics — a cookieless, aggregate-only visit counter that does not use fingerprinting and collects no personal data. It only ever sees anonymous page-view counts, never anything from the files you upload or analyze.
On accessibility: Student Insight has not yet been formally tested against accessibility standards (e.g. WCAG). If you rely on assistive technology, it's worth verifying directly rather than assuming full support today.
Your Spreadsheet is the Source of Truth▾
Traditional systems store information inside databases. Student Insight stores knowledge inside educator-owned project files.
The application can always reconstruct the complete working environment from those files.
Nothing important depends on browser memory. Nothing important depends on server storage.
This carries through the whole academic year, not just a single test. As Test 2 and Test 3 come in, the same file grows to hold them — the Setup step can load an already-filled sheet and add the next test's columns onto it directly, so nothing already recorded is ever re-entered or discarded along the way.
Your project remains portable, transparent, and future-proof.
Designed for Every Educational Institution — and Individuals▾
Its architecture is designed to support:
SchoolsCollegesUniversities
Coaching CentresIndividual Tutors
Training InstitutesOnline Learning Programs
ParentsSelf-Prep Aspirants
Beyond Marks and Attendance▾
Student Insight is not another student management system. Its purpose is to help educators understand learning.
The goal is not simply to collect data.
The goal is to transform educational data into educational intelligence.
Offline First. Stateless by Design.▾
Every educator or parent accesses the same application.
Every user owns their own data.
Every project remains independent.
Every user owns their own data.
Every project remains independent.
No shared sessions.
No shared storage.
No hidden cloud dependency.
No shared storage.
No hidden cloud dependency.
The application is temporary.
Your project is permanent.
Your project is permanent.
Open, Portable and Future Ready▾
Because Student Insight is built around open file formats and user-owned data, everyone remains free to archive, migrate, or extend their projects without being tied to proprietary infrastructure.
"Open" here covers two separate things: open file formats (xlsx/PDF — described above) and the MIT source-code license shown in the stats bar. The source itself is not yet published at a public, stable link — this page will be updated with a direct repository link as soon as that's live, rather than leaving the license claim unverifiable.
"Student Insight is a privacy-first, offline education analytics platform where educators own their data, projects live in user-controlled files, and the application serves only as an intelligent analysis engine."
— Our Philosophy📐
Want to verify the maths yourself?
Every average, rank, percentile, trend and composite score, written out as exact formulas — for the maths/statistics/analytics people your institute will ask.
No questions match your search. Try a different word.
From the PrincipalMoney, liability, reputation▾
QIs this free forever, or a trial that starts charging later?Serious▾
It's free. There's no backend, no server cost, no subscription infrastructure — it's a single HTML file hosted on GitHub Pages, which is free hosting. There's nothing for me to "start charging" for later because there's no ongoing cost on my end to recover.
QWhat's the catch — what are you getting out of this?Serious▾
No catch in the commercial sense. I built it because I wanted a tool like this to exist and be genuinely private. There's no data collection, no ads, no lead-gen — it's a personal project, not a funded product.
QWho else is using this? Can you name another school?Serious▾
Right now, honestly — I can't point you to another school using it yet. You'd be among the first. I won't pretend otherwise.
QIf a grade comes out wrong on a report card, who's responsible?Serious▾
The calculations are transparent formulas (averages, trends, rankings) applied to the numbers entered — there's no hidden AI making judgment calls. If a number is wrong, it's traceable to either the source data or a bug I need to fix. Every report should be reviewed by a teacher before it goes out, the same as anything drafted by a junior staff member.
QDo you have a registered company?Serious▾
No. This is an independent personal project, not a registered vendor. If procurement requires a registered entity, this can't satisfy that today.
QWhat happens if you stop maintaining this next year?Serious▾
Your data isn't stored anywhere by the app — it lives in your Excel files. Even if maintenance stopped entirely, no historical records would be lost; only future analysis via this specific tool would stop. That's a direct benefit of the stateless design.
QDoes this replace our school ERP?Practical▾
No — it doesn't handle attendance, fees, or admissions. It's narrowly for turning marks into analysis and report cards, as an add-on for one specific job, not a replacement for a school management system.
QDid you build this alone? Are you even a teacher?Silly▾
Yes, alone, and no, not a teacher. This was built as a developer solving a well-defined problem — marks in, insight and report cards out — not as someone claiming pedagogical authority.
From the VP / Academic CoordinatorProcess & control▾
QDoes this integrate with our existing student database?Serious▾
No integration — you upload an Excel file each time. If your ERP can export marks to Excel, that export can be fed into this. There's no live sync.
QOur teachers struggle with Excel — will they manage?Practical▾
The app generates a ready-made Excel template based on your subjects and tests, so teachers fill a structured sheet rather than building one from scratch. That said, a teacher uncomfortable with Excel entirely will still need to do that filling-in step — this doesn't remove Excel from the workflow.
QWho trains the teachers on this?Practical▾
There's no training program or support team — it's a single developer and the app itself. The Sample Files section has downloadable examples to reduce guesswork, but there's no live onboarding available at scale today.
QCan I use this to track just my own kid?Practical▾
Yes — choose Individual mode in Setup instead of Institution mode. It's built for a parent tracking their own child or an aspirant tracking their own exam prep, with no class roster or comparison to other students.
QCan I track two children in different grades?Practical▾
Yes. Add each child as their own row in the same Individual-mode Excel sheet, then use the switcher on Insights to flip between them — each child's data and reports stay fully separate, never compared to each other.
QDo I need to be tech-savvy to use this?Practical▾
No — you just fill in a downloadable Excel template with marks and upload it back. No installation, no account, and no spreadsheet formulas to write yourself.
QCan a coordinator see all teachers' data, and teachers only their own?Serious▾
There's no login or accounts at all, so there's no "coordinator view vs teacher view." Anyone with the uploaded file and the app open sees everything in that session. Access control is entirely organizational (who has the file), not enforced by the app.
QDoes it catch teacher data-entry mistakes?Practical▾
Partially — before analysis runs, it checks for duplicate student IDs, marks rows referencing students who don't exist, and non-numeric entries in a marks cell (like "Absent" typed where a number was expected). It flags these; it doesn't guess the correct value for you.
QDoes it handle CBSE/ICSE/State board grading differences?Practical▾
Grading bands and thresholds (pass marks, drop alerts) are configurable per class in Setup, so it's flexible on numbers. It hasn't been specifically validated against every board's exact convention — worth a direct conversation about your system before relying on it as a perfect fit.
QWhat about students with different elective subject combinations?Serious▾
No — subjects are set once at the class level in Setup and applied uniformly. If students take different subject combinations, this isn't built for that today.
QCan I customize the report card layout?Practical▾
Not as a visual template editor — the layout is fixed in the code. What is customizable is the content: institution name, class, subjects, tests, and thresholds.
QDoes it work if our internet is down?Practical▾
Once the page and its libraries have loaded, core steps (reading Excel, computing, generating PDF) run in-browser, so a brief mid-session drop likely doesn't interrupt them. But it needs internet to load the app and its libraries initially — it isn't designed or tested as a true offline tool.
From the IT-in-chargeHosting & data location▾
QWhere exactly is our data stored?Serious▾
Nowhere, by design. The app doesn't save student data to any server, database, or even the browser's local storage — it exists only in the browser tab's memory while in use, and disappears when the tab closes. The uploaded Excel file is the only persistent copy, and it stays wherever it already was. One honest caveat: "nothing is stored" isn't the same as "nothing is ever visible to anyone else" — on a shared or public computer, anything genuinely in memory during an active session could still be seen via the browser's back/forward cache, open DevTools, or a screen-share/remote-support tool. That's inherent to any browser-based tool, not something this app can fix on its own — close the tab when you're done on a shared machine.
QIs this hosted on a government-approved server?Serious▾
It's hosted on GitHub Pages, a standard public web-hosting service — not on any government-empanelled cloud.
QYou said no login — so how do you stop misuse?Serious▾
There's no authentication layer. Access control is entirely on the school's side (who has the link, who has the Excel file). This tool doesn't provide role-based access.
QInstalled app, or does it run in a browser?Practical▾
Browser only — it's a website, not an install from a store. It is technically a PWA (progressive web app), so it can be "Added to Home Screen" for an app-like icon, but it's still a web page underneath.
QIs this on the Play Store / App Store?Practical▾
No, only via the web link.
QCan we host this on our own school website instead?Practical▾
The code is a single HTML file, so it can technically be hosted anywhere that serves static files, including a school's own web server — but that's not currently packaged or officially supported as a distributable install.
QWhat if our ISP blocks GitHub Pages?Silly▾
Then the app won't load. There's no fallback hosting today.
QIs this accessible to users with disabilities (screen readers, keyboard-only navigation, etc.)?Practical▾
It hasn't been specifically tested against accessibility standards (e.g. WCAG) — worth verifying directly with your assistive-technology setup rather than assuming it works cleanly.
QIs the source code available — can we audit it?Serious▾
Not yet at a public, stable link, despite the MIT license shown on the About page — that claim is about the license Student Insight is released under once published, not a working repo link today. This will be updated with a direct link as soon as one is live, rather than leaving it unverifiable in the meantime. In the meantime, the app's entire logic runs client-side and is visible via browser dev-tools (see "Where in the code do these formulas actually live?" under Exact Formulas).
From a Regular TeacherThe actual daily user▾
QHow much time will this actually save me?Practical▾
There's no validated number from a formal time-and-motion study against manual report-card creation. What's concrete: it removes the manual calculation and PDF-formatting steps once marks are in the template — the real time saved depends on the current process being replaced.
QCan I edit the AI-generated remarks, or are they locked?Practical▾
Yes — every narrative field (The Bottom Line, Strengths, At Home This Week, At School) has an editable box right in the student modal, with its own Save button. Whatever you type there is what goes into the exported PDF. The one exception is "What's Changed," which is a computed summary of the marks table, not a wording template — it isn't meant to be edited.
QDoes the report show total marks, not just percentages?Practical▾
Yes — each test row shows a Total column (e.g. "231/450") alongside how many subjects were actually attempted (e.g. "4/5 opted"), so a partially-attempted test doesn't get silently folded into an average without showing the real numbers. In Institution mode, a Class Avg row sits directly under each test for comparison.
QHow does it handle a student absent for a whole test?Practical▾
There's an absence field per test that factors into some calculations (like the stress score, which weighs absence frequency). The exact effect on overall averages is worth verifying on a real example with your data before relying on it — better to check than assume.
QCan I save my work and come back tomorrow?Serious▾
No — this is the most important limitation to know upfront. Closing the tab loses everything in that session. The source Excel file is safe, but any in-progress Setup or analysis is not saved. Re-upload and re-run analysis each time you return. This is a deliberate tradeoff for the "nothing stored anywhere" privacy model, not a bug.
QWhy do I need an app for something I can do in Excel myself?Serious▾
Fair challenge — averages and rankings could be computed with Excel formulas directly. What this adds is packaging: automatic report card layout, consolidated flags/wellbeing indicators, and PDF generation, without building and maintaining your own formulas and formatting every term.
QWill this replace my judgment as a teacher?Serious▾
No — every flag comes with a plain-language reason (e.g. a specific score drop between two named tests), meant to surface things to look at, not decide for you. Any flag can be disagreed with.
QCould the AI say something insensitive about a student?Serious▾
The "AI" here is rule-based text generated from formulas — not a language model improvising sentences — so it won't say anything unexpected outside its fixed templates. It hasn't been independently reviewed by an educator for tone, so feedback on anything that reads badly is genuinely welcome.
QCan I do this on my phone, or only a laptop?Practical▾
It runs in a mobile browser, but working with Excel uploads and multi-column setup forms is realistically more comfortable on a laptop — not recommended for the setup/data-entry steps on a phone.
QCan I fix a mistake by re-uploading, or do I start over?Practical▾
Re-uploading a corrected file re-runs the analysis fresh. Restarting the whole Setup step is only needed if the mistake was in Setup itself.
QWill parents see this directly?Practical▾
No built-in parent portal — export the PDF and share it however report cards are already shared (print, email, WhatsApp, etc.).
QDoes it support regional-language names with special characters?Practical▾
Non-Latin scripts and special characters in name fields haven't been specifically tested — worth verifying directly rather than assuming it works cleanly.
QDoes it print directly, or save PDF first?Silly▾
It generates a downloadable PDF; printing happens through your normal printer/PDF viewer — no direct "print" button skipping that step.
QCan I put my school's logo on the report card?Silly▾
Not currently — there's no image/logo upload feature in the app today.
QIs there a WhatsApp group for support?Silly▾
No — support is directly from the developer if something's broken, not a dedicated team.
Parent-Facing ConcernsWhat the principal anticipates▾
QWhy does it say my child is at "burnout risk"?Serious▾
Every flag includes a specific, factual explanation — like a defined score drop between two named tests — rather than a vague label. That reasoning can be shown to a parent directly, not just an unexplained flag.
QIs this leaking data to an ad company?Serious▾
No — there are no ads, no analytics tracking of student data, and nothing is transmitted off the device during use. This isn't a promise on faith — it's a direct consequence of there being no server for data to go to in the first place.
QIs this like a "free VPN" that secretly sells your data?Silly▾
No — those business models make money by collecting and selling data. This app has no mechanism to collect data in the first place, so there's nothing to sell.
For the Admin / OfficeBudget & paperwork▾
QWill there be a per-student or per-teacher fee later?Serious▾
Not planned — there's no infrastructure cost driving a future need to charge. "Never" isn't a guarantee that can be made absolutely, but there's no roadmap toward monetization today.
QDo we need to sign an MOU or agreement?Serious▾
No formal agreement exists today — this isn't an enterprise product with a sales/legal process behind it.
QIs there a paid version with more features?Practical▾
No — there is one version.
The Nitpicky (But Real) OnesYou'll actually hear these▾
QDoes it work on Internet Explorer?Silly▾
Almost certainly not reliably — it depends on modern browser features and is built and tested against browsers like Chrome, Edge, and Firefox, not legacy IE.
QCan it detect if a teacher marked a student unfairly?Silly▾
No — it only analyzes the numbers it's given. It has no way to know whether those numbers were assigned fairly in the first place.
QWhy "Insight" — does it predict the future?Silly▾
There is a "predicted next score" metric based on simple trend extrapolation from past scores — closer to "if this trend continues" than any real forecasting model.
QIf two teachers open it at once, will it get confused?Silly▾
No — each browser tab or session is completely independent; there's no shared server state to conflict over.
QWhat's the AI's error rate?Practical▾
There's no trained model with an ML-style error rate — every number is a deterministic formula. The more accurate framing: the calculations are exactly as accurate as the data entered and the formulas used, which can be verified by hand on any single student.
QIs this connected to ChatGPT — is my data going to OpenAI?Serious▾
No — there's no external AI API call anywhere in the app. All "AI-style" outputs are generated locally from formulas in the browser.
QWhat if I close the tab by mistake?Serious▾
Everything in that session is gone. This is worth knowing upfront rather than discovering mid-report-card-season — see "Can I save my work and come back tomorrow?" above.
Exact Formulas & Calculation LogicFor maths/statistics/analytics reviewers — every number, in writing▾
ΣHow is a test percentage calculated?Core▾
For each test:
Test % = ( Σ min(marks scored, subject max) ) ÷ ( Σ subject max ) × 100, summed across every subject the student attempted in that test, then rounded to the nearest whole number. A subject with no mark entered is excluded from both the numerator and denominator for that test (it doesn't count as zero). If an entered mark exceeds the subject's max, it's clamped down to the max for this calculation and separately flagged as a data-entry issue.ΣHow is the Overall Average calculated?Core▾
Overall Avg = ( Σ marks scored across ALL tests ) ÷ ( Σ max marks across ALL tests ) × 100, rounded once at the end. This is deliberately not a simple average of the already-rounded per-test percentages — averaging pre-rounded numbers compounds rounding error test over test. Working from raw cumulative totals and rounding only once keeps the final figure accurate to the source marksheet.ΣWhat are the grade bands?Core▾
Applied to the Overall Average:
A+ ≥90 · A ≥80 · B ≥70 · C ≥60 · D ≥ Pass Threshold · F below Pass Threshold. The Pass Threshold itself is set per-institute in Setup (default 35%), so D/F cut-offs move with whatever threshold was configured — the A+/A/B/C bands are fixed.ΣHow is class Rank decided, including ties?Core▾
Students are sorted by Overall Average, descending. Standard competition ranking ("1224") is used: students tied on Overall Average share the same rank number, and the next distinct score skips ahead by the number of students tied above it (e.g. two students tied for Rank 1 → the next student is Rank 3, not Rank 2). This matches how real exam boards report ranks.
ΣHow is Percentile calculated?Core▾
Students are sorted by Overall Average ascending. For a student at position
i (0-indexed) out of n students: Percentile = round( i ÷ (n−1) × 100 ). With only 1 student in the class, percentile is defined as 100.ΣHow is Trend (Improving/Declining/Stable) decided?Core▾
Needs at least 2 tests with valid data.
diff = (latest test %) − (first test %). Improving if diff ≥ +5, Declining if diff ≤ −5, otherwise Stable. It only compares the first and most recent test — it does not fit a trendline through every point in between.ΣHow is the Predicted Next Score calculated?Core▾
Simple linear extrapolation, needs ≥2 valid tests:
slope = (latest % − first %) ÷ (number of valid tests − 1), then Prediction = latest % + slope, clamped to 0–100. This is "if the average trend so far continues by one more step" — a straight-line projection, not a statistical forecasting model.ΣHow is SD (Standard Deviation) calculated, at class and student level?Core▾
Population standard deviation (divide by n, not n−1):
SD = √( Σ(x − mean)² ÷ n ). At class level, x = each student's Overall Average. At student level (used for Volatile and Consistency Score below), x = each of that student's own valid test percentages.ΣHow are Median, Q1 and Q3 calculated?Core▾
Class Overall Averages are sorted ascending (n values). Median: middle value if n is odd, average of the two middle values if n is even. Q1: the value at index
floor(n × 0.25). Q3: the value at index floor(n × 0.75). This is one accepted quartile convention among several used in statistics software — with small class sizes, different conventions can shift Q1/Q3 by a mark or two, which is normal.ΣHow is Consistency Score calculated?App-defined▾
Consistency = max(0, round(100 − 2 × SD)) of the student's own valid test percentages. It's mathematically the inverse of that student's own volatility — a design choice to turn "spread" into a 0–100 "steadiness" number that reads intuitively on a report card.ΣHow is Growth Rate calculated?App-defined▾
Growth Rate = round( (latest % − first %) ÷ max(first %, 1) × 100 ), clamped to ±300%. The clamp exists because a near-zero starting score (e.g. 0% → 50%) produces a mathematically correct but meaningless "+5000%" — clamping keeps the number informative instead of alarming on a parent-facing report.ΣHow is the Health Score calculated?App-defined▾
A weighted blend of four sub-scores, each 0–100:
Health = 0.4×Academic + 0.2×Consistency + 0.2×Trend + 0.2×Engagement. Academic = Overall Average (capped at 100). Consistency = as defined above. Trend = 100 if Improving, 60 if Stable, 20 if Declining. Engagement = the Engagement Index defined below. Bands: Excellent ≥80, Good ≥65, Average ≥50, Below Average ≥35, Needs Support below 35. The 40/20/20/20 weighting is this app's own design choice, not a published or standardized formula — stated plainly so it isn't mistaken for a clinical or peer-reviewed instrument.ΣHow is the Engagement Index calculated?App-defined▾
absentPct = min(100, totalAbsences ÷ (number of tests × 3) × 100), then Engagement = min(100, round( (100 − absentPct) × trendMultiplier )), where trendMultiplier is 1.1 for Improving, 0.8 for Declining, 1.0 for Stable. It is an attendance-and-trend based proxy — the app has no way to directly measure classroom participation.ΣHow is the Stress Score calculated?App-defined▾
Stress = min(100, (totalAbsences × 5) + (30 if Declining) + (40 if below Pass Threshold)). Wellbeing Flag: High ≥60, Moderate ≥30, Low below 30. Built from absence frequency and performance trend only — not a certified wellbeing or psychological assessment.ΣHow is the Early Warning Score calculated?App-defined▾
Sum of fixed point-values for whichever flags are currently active on the student, capped at 100:
At Risk +40 · Sharp Drop +20 · High Absence +15 · Volatile +10 · Burnout Risk +15. It's an additive composite of existing flags, not an independently derived statistic.ΣWhat exactly triggers each flag (Volatile, Sharp Drop, Plateau, Burnout, Resilient)?App-defined▾
Volatile: SD of the student's valid test % values > 15. Sharp Drop: any two consecutive tests where the score falls by ≥ the configured Drop Alert threshold (default 20 points). Plateau: across 3+ valid tests,
(max − min) ≤ 8 points, and Overall Average is below 70 (a high-scorer holding steady near the top isn't flagged). Burnout Risk: first valid test ≥70% AND latest valid test is more than 15 points below the first. Resilient: at any point in the sequence, a test dropped ≥10 points from two tests prior, and the very next test recovered by ≥8 points from that dip.ΣHow is a Subject Average calculated, and Subject-vs-Class Delta?App-defined▾
Subject Average = mean of that subject's per-test percentage
(marks ÷ subject max × 100, capped at 100) across every test where it was attempted, rounded. Subject-vs-Class Delta = that student's Subject Average minus the class-wide average for the same subject — a signed number showing exactly how far above/below the class norm a student sits in one subject.ΣHow is class-level Subject Weakness identified?App-defined▾
For each subject:
% of class below Pass Threshold in that subject, plus the class-wide subject average. Subjects are then ranked by that percentage, worst first — this identifies which subject needs curriculum attention, distinct from which individual student needs attention.ΣHow is the Attendance-vs-Performance correlation calculated?App-defined▾
Students are split into two groups — zero recorded absences vs. one-or-more absences — and the average Overall Average of each group is compared side by side. Only shown when both groups have at least 2 students, as a basic floor against drawing conclusions from (or identifying) a single student's data.
ΣWhere in the code do these formulas actually live?For reviewers▾
All of them run client-side, in the browser, inside three functions in the app's own source:
computeAnalysis() (per-student scores, flags, grades), computeExtraInsights() (subject deltas, rank movement) and computeClassStats() (mean/median/SD/quartiles/correlations). Anyone with browser dev-tools access can inspect the exact running code — nothing is computed on a server or hidden behind an API.Terms & Abbreviations Used in the AppWhat every label actually means▾
AaMedianStatistics▾
The middle score if every student were lined up from lowest to highest — half the class scored above it, half below. Unlike a plain average, one or two extreme scores can't pull the median up or down.
AaSD (Standard Deviation)Statistics▾
Full form: Standard Deviation. A measure of how spread out the scores are around the average. A small SD means most students scored close together; a large SD (e.g. ±22) means there's a big gap between the strongest and weakest students in the class.
AaQ1 (First Quartile)Statistics▾
Full form: First Quartile. The score below which the bottom 25% of the class falls. If Q1 is 48%, a quarter of the class scored 48% or less.
AaQ3 (Third Quartile)Statistics▾
Full form: Third Quartile. The opposite end from Q1 — 25% of the class scored at or above this number.
AaPercentileStatistics▾
Shows how a student compares to classmates — e.g. "78th percentile" means this student scored better than 78% of the class. This is not the same as a percentage score — a student can be at the 78th percentile with a 60% score, if the rest of the class scored even lower. Easy to misread, worth clarifying with parents. Below 12 students in a class, the app doesn't show a percentile number at all — it implies more precision than a tiny class actually has — and shows rank plus a plain point-difference from the class average instead.
AaRank & Tie HandlingApp logic▾
Uses standard "competition ranking": if two students tie for 1st place, both are shown as Rank 1, and the next student is Rank 3 — not Rank 2. This matches how real exam results are conventionally reported, so a "skipped" rank number is expected, not a bug.
AaHealth Score / Health BandApp-defined▾
Not a standard statistical or medical term — it's a single 0–100 number this app calculates by blending four things: 40% how well the student is scoring, 20% how consistent they are, 20% whether they're trending up or down, and 20% an engagement estimate. Those weightings are a design choice made for this app, not a scientifically standardized formula.
AaConsistency ScoreApp-defined▾
How steady a student's scores are from test to test. A high consistency score means they don't swing wildly between a great test and a bad one — it's mathematically the inverse of how "spread out" their individual scores are.
AaGrowth RateApp-defined▾
How fast a student's scores are moving up or down over time, shown as a percentage with a + or – sign.
AaEngagement IndexApp-defined▾
The app's estimate of how "involved" a student seems, inferred indirectly from attendance and score-trend data. Important to clarify: this is a proxy, not a direct measurement — the app cannot actually observe classroom participation or attention.
AaPlateau DetectionApp-defined▾
Flags a student whose scores haven't moved up or down much across 3 or more tests in a row. It means "flat," not necessarily "failing" — a high-scoring student can plateau too.
AaEarly Warning ScoreApp-defined▾
A single combined risk number built from several other signals together (score drops, volatility, absences), meant to surface students needing attention even before any one signal alone looks alarming.
AaBurnout RiskApp-defined▾
Flags a student who was previously scoring well and then dropped sharply. It describes a score pattern associated with burnout — it is not a clinical or psychological diagnosis.
AaResilience ScoreApp-defined▾
The positive counterpart to Burnout Risk — flags a student who bounced back well after a bad test.
AaVolatility / VolatileApp-defined▾
Same idea as SD above, but applied to one individual student instead of the whole class — do their scores bounce around a lot from test to test?
AaStress Score / Wellbeing FlagApp-defined▾
A number combining how often a student is absent with penalties for declining performance trends. Meant as an indicator to look into, not a certified wellbeing assessment.
AaPWA (Progressive Web App)Technical▾
Full form: Progressive Web App. A website built to feel more like an installed app — it can be "Added to Home Screen" for an icon, but it's still a web page running in the browser underneath, not a Play Store / App Store install.
AaMOU (Memorandum of Understanding)Admin▾
Full form: Memorandum of Understanding. A formal agreement document sometimes required before a school adopts outside software — not currently used for this app, since there's no vendor relationship behind it.
AaERP (Enterprise Resource Planning)Admin▾
Full form: Enterprise Resource Planning. The umbrella term for a school's main management software (attendance, fees, admissions, etc.). This app is not an ERP — it's a narrow add-on for marks analysis and report cards only.
AaCBSE / ICSEAdmin▾
Full form: Central Board of Secondary Education / Indian Certificate of Secondary Education — the two major Indian school-board systems, each with its own grading conventions the app's thresholds may need to be checked against.