Trust & data
Where your students' data goes, and the controls we operate
The controls PrepGraph operates on a school's student data, stated as facts and terms, and the five questions the US Department of Education tells a school to ask any technology provider, answered in order. Each control goes into the agreement with your school in the same words.
- Controls
- 12
- Sign-in
- Badge and PIN
- Encryption
- In transit and at rest.
The five questions
The questions the Department of Education says to ask us
The US Department of Education's August 2026 guidance tells schools to ask every technology provider five things. They are answered here in order.
- 01
1. What learning problem does it solve?
The Monday-morning one. A teacher can see, before first period, which students are stuck on which concept and which concept to reteach first, from the work each student did inside PrepGraph. Overnight, each activated student is placed in one of three bands on every concept there is evidence for; the class's concepts are ranked by which to reteach first; and the students behind each are named. A leader gets chapter-level mastery for every class in the school in one table and a single at-risk register, worst first. A misconception is read from the specific pre-tagged wrong answer a student chose, not from a raw score.
- 02
2. When should it be used?
After a unit is taught, as weekly triage, by the teacher who already teaches those students. The band is a three-level cut on an ability estimate and it feeds the reteach queue; decisions about eligibility, tiers and services stay with the instruments and the people your school already uses for them.
- 03
3. For whom?
High school to begin with: grades 9–12, math and science, in English, with your teachers choosing the topics, the order and the pacing. Our US assessment items cover the ACT and SAT, with a concept map covering 39 AP course outlines. We are not affiliated with, endorsed by or licensed by ACT, Inc. or the College Board; we name their exams to say what our content covers.1
- 1 content bank, 20 September 2026
- 04
4. For how long?
A semester if your calendar allows, with the measurement set up before the first login and a mid-point look at activation and weekly practice together. A concept shows a band as soon as there is graded evidence for it, so a fresh grid is thin and fills in as students practise; the response count behind any cell is available on request.
- 05
5. What evidence shows it works?
The pilot is built to produce yours. Four measures are recorded on the cohort — activation, weekly practice, completion of each assessment window, and score change on an instrument your school sets and scores, matched student by student — against targets agreed in writing before day one. The written review at the end states each one as met or not met, with the number of matched pairs and the exclusions printed, and you approve the dataset, the wording and whether your school is named before it is shared.
Student data
The controls we operate
Stated as facts and terms. Each one goes into the agreement with your school in the same words, so what your counsel reads here is what your counsel signs.
- Data control
- PrepGraph processes student data on the school's instructions and only for the purposes the school disclosed it for. We claim no ownership of it, do not sell it, do not use it for advertising, do not build profiles for non-educational purposes and do not market to your families from a roster you gave us. These go in the agreement verbatim.
- Who processes student work
- Student speech and text go to approved AI endpoints in Azure India and to named speech providers, and nowhere else — enforced by configuration. The named sub-processor list, and its terms, go in the agreement.
- Isolation
- Class analytics are readable only by that class's assigned faculty and by administrators of your own tenant, enforced in the query. Within class analytics, students are never shown other students' data.
- Sign-in
- A school-provisioned student signs in with a printed badge and a short PIN — no email address, no password to reset, no account a child has to create.
- Sessions and retention
- Tutoring is a live voice session. Retention and deletion terms for your school go in the agreement, and written consent where your state's law requires it goes in the pilot agreement.
- Audit log
- Every change to an assessment mark is written with the value before and after. The audit terms for your school go in the agreement.
- Advertising and tags
- Ad signals are denied everywhere; no third-party tag runs inside a child's session.
- Encryption
- In transit and at rest.
- Rostering
- A roster file your staff load. The load runs as a preview first and creates accounts only on the explicit write pass; each student then gets a printed badge and a short PIN, and each teacher a portal login.
- Price and seats
- Quoted per cohort, in writing, before a pilot starts; the school buys a seat for every student in the cohort.
- Support
- Founder-led implementation through the pilot, a named person, and the reply channel and response terms written into the agreement.
- Numbers on this site
- Every count on this site carries a footnote naming what it was counted from and when. A roster table is the production roster on that date; a content bank is our content database on that date. Ask us to re-run any of them in front of you.
Accessibility: there is no VPAT or Accessibility Conformance Report (ACR) for PrepGraph yet. Ask us where the work stands and we will put it in writing.
Ask us, and ask everyone else
The same six things about any number a vendor shows you: sample size, comparison group, duration, who funded the work, who reviewed it, and where it showed no result. The pilot is written so that the answers are on paper before day one.
Our research programme is public at research.prepgraph.com.
PrepGraph’s Founder & CEO is Umesh Sharma.