An empty trial cannot demonstrate a closed-loop system.
So we ship one pre-seeded.
Most software is a system of record — uploading data already proves something. Mentron is a system of action: misconception tags, generated remediation, mastery deltas and outcome attainment all appear after graded work has been processed. None of that is visible in a fresh tenant on day one.
The sandbox is a fully provisioned synthetic institution. Rooted rosters, real-looking submissions, multi-week history, and the loop already in motion. Evaluation starts from a populated state instead of an empty one — which is the only honest way to evaluate a system whose value is longitudinal.
Five reasons free trials fail for closed-loop LMS evaluation
A free trial is the right evaluation model for many products. It is the wrong one for a platform that needs prior work to produce signal.
You start from zero
A free trial begins with an empty database. There are no courses, no rosters, no submissions, no history. The interface looks correct, but there is nothing to act on yet — and populating it requires uploading your real student data before any agreement is in place.
AI needs prior work to be useful
The interesting parts of Mentron — misconception tagging, targeted remediation, mastery deltas, outcome analytics — only appear after graded work exists. In an empty trial, the AI has nothing to read and the loop cannot close.
You cannot evaluate risk honestly
A free trial shows what the system does in the happy path. It does not show how the approval gate behaves under disagreement, what happens when a rubric is ambiguous, or how attainment reporting reads across cohorts — because none of those conditions have been triggered yet.
Real value takes weeks, not minutes
Mentron is most useful after a full instructional cycle: assessment, remediation, return, retest. A 14-day trial covers at most one assessment before the clock runs out. Longitudinal value cannot be demonstrated inside a trial window.
Generative cost falls on you
A genuinely free trial for an AI-native platform would expose the institution to unbounded LLM cost on first contact, before any procurement guardrail is in place. The sandbox keeps the cost model predictable while still letting you exercise the product.
Free trial vs Mentron sandbox
The same evaluation question, answered two different ways.
| Capability | Typical free trial | Mentron sandbox |
|---|---|---|
| Starting state | Empty tenant. No courses, no students, no history. | Pre-seeded synthetic institution with rosters, courses, submissions and multi-week history. |
| Time to first insight | Days — assuming you upload data and create assignments first. | Minutes. The loop is already populated. |
| AI feedback you can see | Not available. No graded work exists yet. | Rubric-aligned evaluations across hundreds of submissions, pending review. |
| Misconception & mastery analytics | Empty charts. | Real distribution across topics, students and cohorts. |
| Approval-gate workflow | Configurable, but untested. | Live. Review, edit and approve sample evaluations. |
| Outcome & Bloom reporting | Empty matrix. | Filled CO × Bloom coverage heatmap with attainment thresholds. |
| Data privacy posture | You have to upload real records to test anything realistic. | Synthetic records only. No real student data ever leaves your evaluation. |
| Cost model during evaluation | Free to sign up, but unbounded LLM usage is on the institution. | Provisioned and metered. Predictable for both sides. |
What you can experience inside the sandbox
Four flows, each one set up so a first-time visitor can complete it without guidance.
Walk the teacher flow
Sign in as a course instructor. Review a queue of AI-drafted rubric evaluations, edit the feedback, and approve or send back. See how the approval gate writes back to mastery and the gradebook — without releasing anything to the student before you are ready.
Walk the student flow
Sign in as a student. Open the flashcards, mindmap and notes that were generated from your specific misconception tags. Step through the spaced-repetition queue and watch the mastery delta update after each review.
Walk the leadership flow
Open the department dashboard. Inspect the CO × Bloom heatmap, attainment percentages against the 70/50 KHDA/MoE thresholds, and the at-risk cohort signal that surfaces weeks ahead of assessment.
Walk the configuration flow
Ingest your own materials into a parallel sandbox slot. Define a rubric, generate a quiz, and observe the AI evaluation against your specific criteria rather than the synthetic demo data.
Common questions about sandbox evaluation
See it running, not described
The sandbox is a full synthetic institution — rosters, submissions and multi-week history already in place. Walk the teacher, student and department flows end to end. Access is provisioned per email address, so start with a short request.
No student data required. The sandbox runs entirely on synthetic records.