Criteria-Based Assessment System

Students can submit documents (PDF, DOCX), plain & formatted text (TXT, RTF), presentations (PPTX), audio & video (MP3, WAV, M4A, OGG, WEBM, MP4), code projects (ZIP), rich text entries, and portfolios with multiple files. Each format has a specialized processing pipeline that extracts all content including embedded images, speaker notes, code files, and audio transcripts. If a file can't be read (e.g. a legacy binary .doc/.ppt or a scanned PDF), it's flagged with a clear message and never silently graded as blank.

Unlike percentage-based scoring, threshold grading evaluates each criterion as Met or Not Met. For Pass/Merit/Distinction frameworks, a student must meet ALL Pass criteria to achieve a Pass, ALL Pass and Merit criteria for Merit, and ALL criteria at every level for Distinction. Four-tier frameworks (IB, AQA/OCR, NAAC/NBA) add a top High Distinction band, and the overall grade is shown in each framework's own vocabulary (an SQA top grade reads “Grade A”; an NVQ pass reads “Competent”). This matches exactly how awarding bodies like Pearson BTEC require assessments to be graded. The system applies the correct logic automatically based on the framework you select.

Yes. When you upload an assignment brief, the AI extracts criteria, learning outcomes, and scenarios. You can then review, edit, add, or remove any criteria before running evaluations. You retain full control over the assessment framework and can adjust it at any time.

Billing uses the same tiered pricing as Evaluation Studio, charged per student submission evaluated. Batch evaluations are priced per student in the batch. This means you only pay for evaluations you actually run. Contact your account manager for volume pricing on large cohorts.

There is no hard limit on batch size. The system processes submissions in parallel with real-time progress tracking. Typical batches range from 10 to 200+ students. Larger batches automatically include consistency analysis showing per-criterion pass rates across the cohort and grade distribution breakdowns.

Individual student reports are available in PDF, DOCX, and CSV formats. Batch reports can be downloaded as ZIP bundles containing all student reports in your chosen format(s). Reports include per-criterion judgments with evidence, overall grade, strengths, areas for improvement, and recommendations.

Yes. Assignment briefs can contain multiple parts (Part A, Part B, etc.), each with different submission types and criteria coverage. For example, Part A might require a written report covering Pass criteria, while Part B requires a presentation covering Merit and Distinction criteria. Each part is evaluated independently, and results are aggregated for the overall unit grade.

Teacher QC is a human-standardisation side-channel: teachers record their own per-criterion verdicts and written feedback, and the system compares and scores them against the AI — the AI grade always stays the official input and is never overridden. Comparative mode shows where teachers agree or differ with the AI (and whether they mark stricter or more lenient); advisory mode rates the quality of each teacher's written feedback (Specific / Generic / Weak) with pinpoint, evidence-grounded coaching. You can use the built-in rubric or define your own — custom axes with their own scoring instructions, weights, and named performance levels — set as an organization default or per batch, and export a batch QC report to PDF or Word.