The Scorafy blog
Practical guides on better feedback.
Practical guides on rubric-based assessment, AI marking with assessor review, and feedback people can act on.
More articles
- 14 August 2026
Is AI Grading Reliable? Consistency, Validity, and the Numbers Vendors Quote
A straight look at AI grading reliability: why consistency is not validity, why accuracy percentages without a method are a red flag, and how to test a tool on your own material.
9 min read - 14 August 2026
What Should an Audit Trail for AI-Assisted Assessment Contain? A Buyer Checklist
A vendor-neutral checklist for evaluating the audit trail of any AI-assisted assessment tool: config versioning, override ledgers, release records, model identification and reconstructability.
9 min read - 14 August 2026
Rubric Design for Workplace Competency Assessment
How many criteria, how to write level descriptors, when weighting helps and when it hides problems, plus the four failure modes that break workplace rubrics - with worked examples.
10 min read - 14 August 2026
AI Assessment and GDPR: What Actually Matters
The GDPR questions that genuinely bite when AI is involved in assessment: lawful basis, residency, retention, respondent rights, and human oversight of consequential decisions.
10 min read - 27 June 2026
AI Assessment Tool vs AI Quiz Generator: What's the Difference?
The plain difference: a quiz generator uses AI to write questions, an AI assessment tool uses AI to mark the answers. Here is why that distinction decides which one you need.
6 min read - 27 June 2026
How to Grade Open-Ended Responses With AI Against a Rubric
A step-by-step method for grading written, open-ended answers with AI against a rubric, with cited evidence and a human sign-off, so the result holds up.
7 min read - 27 June 2026
Should You Build Your Own AI Grader or Use One?
A practical build-versus-buy guide for AI grading. The prompt is the easy 20%. Rubric handling, evidence grounding, review workflow, and compliance are the 80% that sinks projects.
7 min read - 27 June 2026
AI Rubric Marking: How It Works
How AI rubric marking actually works, step by step: you define criteria and levels, the model reads each response, maps it to the rubric, cites evidence, and a person signs off.
6 min read - 27 June 2026
Grounded AI Feedback: How to Stop the Model Inventing Strengths
AI feedback drifts into inventing strengths a learner never showed. The fix is grounding: make the model quote the actual answer as evidence and never credit what is not there.
6 min read - 27 June 2026
GDPR-Compliant AI Assessment: Data Residency and Isolation
What GDPR-compliant AI assessment requires in practice: lawful data handling, EU data residency, tenant isolation, encryption, and not using submissions to train models.
7 min read - 27 June 2026
Formative vs Summative Assessment: The Difference and When to Use Each
Formative assessment guides learning while it happens; summative assessment measures it at the end. A plain-language guide to the difference, with examples and how to use both well.
7 min read - 27 June 2026
Criterion-Referenced vs Norm-Referenced Assessment Explained
Criterion-referenced assessment measures you against a fixed standard; norm-referenced assessment ranks you against other people. A clear guide to the difference and when each is right.
7 min read - 27 June 2026
Recognition of Prior Learning (RPL) Explained
Recognition of Prior Learning lets a learner gain credit for skills and knowledge they already have, evidenced rather than re-taught. A plain guide to how RPL works and what makes evidence valid.
7 min read - 27 June 2026
Bloom's Taxonomy in Assessment: A Practical Guide
Bloom's Taxonomy ranks thinking from remembering facts to creating new work. A practical guide to using its six levels to write assessment questions that target the right cognitive depth.
8 min read - 27 June 2026
Validity and Reliability in Assessment: What They Mean and Why Both Matter
Validity is whether an assessment measures what it claims to; reliability is whether it measures consistently. A clear guide to the difference, why you need both, and how to improve each.
7 min read - 27 June 2026
Competency-Based Assessment Explained
Competency-based assessment judges whether someone can perform a task to a defined standard, not how they rank or how long they studied. A clear guide to how it works and what it requires.
7 min read - 27 June 2026
Authentic Assessment Explained: Testing Real-World Performance
Authentic assessment asks learners to perform real-world tasks rather than answer abstract questions. A clear guide to what it is, why it works, and how to mark it at scale.
7 min read - 27 June 2026
Assessment Validation for VET: What It Is and How to Do It
Assessment validation is the quality check that confirms VET assessment tools and decisions are sound and consistent. A clear guide to what validation is, how it differs from moderation, and how to run it.
8 min read - 30 May 2026
Best AI Assessment Tools in 2026
An honest roundup of AI assessment tools across three categories, with five decision criteria and who each tool is actually best for.
8 min read - 30 May 2026
Can AI Grade Open-Ended Answers? What It Can and Can't Do in 2026
A practical look at where AI grading of open-ended answers works, where it fails, and why a human still has to sign off.
7 min read - 30 May 2026
AI Grading and the EU AI Act: What Assessment Teams Need to Know
How the EU AI Act treats AI used in education and assessment, why grading is often high-risk, and what your team should put in place.
8 min read - 30 May 2026
Human-in-the-Loop AI Assessment: Why a Person Still Has to Sign Off
What human-in-the-loop really means in AI assessment, the difference between rubber-stamping and real oversight, and how to design sign-off that holds up.
7 min read - 30 May 2026
Rubric-Based Grading vs Auto-Scored Quizzes: Which Actually Measures Skill
When auto-scored quizzes are the right tool, when they fail, and why rubric-based grading is the only way to measure applied skill.
7 min read - 30 May 2026
How to Write an Assessment Rubric AI Can Grade Against
A step-by-step guide to writing assessment rubrics that are clear enough for AI to apply consistently and for a human to defend.
8 min read - 30 May 2026
AI Assessment Tools for RTOs: A 2026 Buyer's Guide
A practical guide to choosing AI assessment tools for RTOs in 2026 - what to look for, what to avoid, and how to stay compliant under ASQA and the EU AI Act.
6 min read - 30 May 2026
Grading Bootcamp Project Submissions Without Burning Out Instructors
Project submissions are the heart of a coding bootcamp and the biggest grading drain. Here is how to mark them consistently and fast without exhausting your instructors.
6 min read - 30 May 2026
What an Assessment Audit Trail Should Contain (and Why Appeals Depend on It)
When a learner appeals a result, your audit trail is the whole case. Here is exactly what a defensible assessment audit trail needs to contain - and the gaps that sink appeals.
6 min read - 30 May 2026
TestGorilla vs Rubric-Based Assessment: When Standardised Tests Fall Short
Standardised tests like TestGorilla are fast and consistent for screening. But for open-ended work judged against your own criteria, they hit a ceiling. Here is where the line sits.
6 min read - 30 May 2026
Multi-Modal Assessment: Grading Video and Audio Answers, Not Just Essays
Text is not the only way to show competence. Video and audio answers reveal skills writing cannot - but they are punishing to grade by hand. Here is how multi-modal assessment works.
6 min read - 24 February 2026
Why Employee Engagement Surveys Fail (And What to Do Instead)
Billions are spent on engagement surveys every year, but engagement has not improved. Here is why aggregate data fails - and how individual AI-powered assessments change the equation.
7 min read - 22 February 2026
How to Write Better Assessment Questions (And Why It Matters for AI Analysis)
A practical guide to writing assessment questions that produce richer AI-generated reports - covering question types, common mistakes, optimal length, and how question design affects the analysis.
7 min read - 19 February 2026
The Difference Between a Score and a Story: Why Assessment Reports Need to Be Personal
Two people score 78 percent but their journeys are completely different. Why reducing someone to a number misses the point - and what happens when reports tell the whole story.
6 min read - 17 February 2026
How I Direct AI to Build a Production SaaS: What Actually Worked
After years spanning sales, consulting, and teaching high school Computer Science - without a traditional engineering career - I built and launched Scorafy by directing AI development tools. Here is the honest version of that story.
8 min read - 14 February 2026
Employee Engagement Surveys Are Broken. Here's What to Do Instead.
Annual engagement surveys suffer from low response rates and generic insights. Assessments that give employees personalised feedback change the equation entirely.
7 min read - 12 February 2026
What Happens When You Actually Read Every Student's Answer
Most assessment tools score responses but ignore what students actually wrote. Here is what changes when AI reads every answer and responds to it.
6 min read - 10 February 2026
Assessment Platforms for Schools and Educators
How schools and educators can use modern assessment platforms for learning styles, wellbeing check-ins, career guidance, and course feedback.
6 min read
See it live
See AI-powered assessments in action.
Try the interactive demo - no sign-up required.