Student 1 receives a paragraph of careful feedback.
Student 27 receives two sentences.
Same assessment. Same rubric. Same teacher.
The difference is often not professional judgement. It is the amount of attention still available after reading, assessing and commenting on dozens of submissions.
A recent Koncepts post described the familiar cycle:
“Student 1 got a paragraph of thoughtful feedback. Student 27 got two sentences.”
For an MYP teacher, that problem is amplified by the way assessment works. You are not simply checking whether an answer is correct and recording a percentage. You are reading evidence against subject specific criteria, looking across strands, deciding which achievement descriptor best represents the work and then explaining that decision in language a student can actually use.
Do that once and it is thoughtful assessment.
Do it for an entire class, across several criteria, and it becomes a workload problem.
Why MYP Assessment Takes Time
The MYP deliberately places professional judgement with teachers.
The IB explains that school based assessment uses tasks created and marked by:
“classroom teachers who are well-equipped to make judgments about student achievement.”
Teachers assess prescribed subject group objectives through the assessment criteria for the relevant subject and year of the programme.
That is fundamentally different from marking a simple answer key.
A student's performance may not fit perfectly inside one achievement band. One strand might resemble the 7 to 8 descriptor while another sits closer to 5 to 6.
Current IB examiner guidance describes the best fit approach as finding:
“the descriptor that most accurately describes the level attained”
and notes that different strands within the same criterion may appear at different levels.
So the mental process behind one assessment can look like this.
Read the submission.
Check the relevant criterion.
Compare the evidence with one achievement band.
Check the neighbouring band.
Return to the student's work.
Compare individual strands.
Decide which descriptor best represents the evidence.
Record the level.
Explain why.
Identify what the student should do next.
Then open the next criterion.
Then the next student.
The rubric is not the problem.
The repetition is.
Marking Workload Is a Real System Problem
The pressure is not unique to IB schools.
The OECD's TALIS 2024 results found that 40 percent of teachers across participating OECD education systems identified excessive marking as a source of work related stress.
OECD TALIS 2024: The Demands of Teaching
That does not prove that feedback inevitably becomes worse for the twenty seventh student.
It does explain why consistency becomes harder to maintain when assessment is concentrated into long marking sessions.
A teacher may begin with detailed comments such as:
Your explanation identifies the relationship clearly, but your justification does not yet establish why it will hold for every case.
Several hours later, the same underlying judgement can become:
Good pattern. Needs more justification.
The second comment may not be wrong.
It simply carries less information.
That matters because the purpose of feedback is not to prove that marking happened.
It is to help learning move forward.
Good Feedback Is Not the Longest Feedback
More writing does not automatically produce better feedback.
The Education Endowment Foundation recommends feedback that is appropriately timed, focused on moving learning forward and designed so that students have an opportunity to act on it.
This gives MYP teachers a useful test.
A good comment should help the student understand three things:
What evidence did I demonstrate?
What is currently limiting my achievement?
What should I do differently next time?
Consider a Criterion B comment in MYP Sciences.
Weak version
Good investigation. Try to make your hypothesis clearer.
More useful version
Your research question identifies both variables clearly. Your hypothesis predicts a relationship, but it does not explain the scientific reasoning behind that prediction. Next time, connect the prediction directly to the concept you expect to cause the change.
The second comment does not need to be dramatically longer.
It is more useful because it connects evidence, the strand expectation and a next step.
Why Two Students With the Same Level Need Different Feedback
Achievement levels summarise performance.
They do not explain how a student arrived there.
Two students may both receive a 5 in the same criterion for completely different reasons.
Student A might demonstrate one strand very strongly but provide weak evidence for another.
Student B might show moderate evidence across every strand.
The final level may be identical.
The useful feedback should not be.
This is one reason feedback generated only from a criterion number is limited.
Giving an AI system:
Criterion B: 5
does not tell it what the student actually did.
At best, it can paraphrase the descriptor.
Useful assessment feedback needs evidence from the submission.
The existing Koncepts guide to MYP grading explains the distinction clearly. The final criterion level tells a student where their achievement currently sits, while the pattern across strands provides much more information about what should improve next.
Read: MYP Grading Explained, Criteria A to D and the 1 to 7 Scale
The Blank Page Is Part of the Workload
There are really two tasks happening during marking.
The first is professional.
What does this work demonstrate against the criterion?
The second is linguistic.
How do I turn that judgement into a clear paragraph for this student?
Only the first task absolutely requires the teacher to begin from professional judgement.
The second is where a structured first draft can help.
This is the distinction behind the Koncepts Feedback Writer.
The purpose is not to make the final assessment decision for the teacher.
It is to reduce the repeated work involved in moving from evidence and descriptors to readable feedback.
That distinction also aligns with the direction the IB is taking on AI.
The IB's 2026 draft AI Design Principles state:
“professional judgement remains central.”
The same guidance says that high stakes decisions require human oversight and that AI tools should support human decision making rather than replace it.
The IB also acknowledges that AI may help educators:
“free up valuable time”
for work with students.
IB AI in Learning, Teaching and Assessment
That is a useful boundary for classroom assessment too.
AI can prepare evidence for review.
AI can draft language.
AI can organise criterion information.
The teacher makes the judgement.
From Writing Feedback to Reviewing Feedback
There is a meaningful difference between these two workflows.
In the traditional version, a teacher opens the submission, interprets the work, chooses an achievement level and then writes every sentence from a blank field.
The next student starts the process again.
With a first draft workflow, the teacher still reviews the submission and criterion evidence, but the repetitive language construction no longer starts at zero.
The job changes from:
Write twenty seven feedback sets
to:
Review and refine twenty seven feedback drafts
That sounds like a small difference.
Across an entire class, it changes where teacher attention is spent.
More time can go into checking whether the judgement is accurate and whether the suggested next step is useful.
Less time goes into finding a fourth way to phrase “you need to justify this more clearly.”
How the Koncepts Feedback Writer Works
The current Koncepts Feedback Writer is designed specifically around MYP criteria and strands rather than a generic comment prompt.
In the Assessment workflow, the teacher selects the student, MYP year and subject, then uploads the student's submission.
According to the current product documentation, Koncepts reads the PDF and returns a suggested achievement band, strand level reasoning and supporting quotations from the student's own work.
The teacher reviews those suggestions, changes any score they disagree with and confirms the final judgement before student facing feedback is generated.
Koncepts MYP Feedback Writer Guide
That final step matters.
A suggested 6 does not become the student's 6 simply because the system produced it.
The teacher can change it.
Koncepts describes the workflow as a review process rather than an automated grading pipeline.
Its Terms also state that AI generated assessments and feedback should not be treated as definitive indicators of assessment performance.
What a Useful First Draft Should Contain
A first draft only saves time if it reduces work without removing specificity.
For MYP assessment, that means the draft should be grounded in the correct subject criterion and its strands.
It should identify evidence from the actual submission.
It should explain why the evidence fits the suggested level.
It should identify a realistic next step.
A generic comment such as:
Excellent effort. Continue developing your analytical skills.
does almost none of this.
A stronger draft might say:
You compare both sources and identify relevant differences in perspective. Your evaluation becomes less precise when discussing reliability because the origin of each source is described but not connected to how trustworthy its claims are. In your next response, explain how origin or purpose changes the value of the evidence.
Now the teacher has something worth reviewing.
Perhaps the judgement is correct and only the tone needs changing.
Perhaps the feedback missed a piece of evidence.
Perhaps the suggested level is wrong.
The teacher edits it.
That is much faster than discovering the judgement and constructing the paragraph at the same time.
Consistency Does Not Mean Identical Comments
The goal should not be to make every student's feedback the same length.
Some students require more explanation.
Some need one precise correction.
Some submissions contain conflicting evidence that deserves a longer comment.
Consistency means something different.
Student 27 should receive the same quality of assessment reasoning as Student 1.
Both should be judged against the same criterion expectations.
Both should receive feedback connected to their actual work.
Both should leave knowing what to do next.
A system that simply produces equally long paragraphs has not solved the problem.
A system that makes it easier for the teacher to apply the same level of attention to the underlying evidence has.
What About the Time Saving Claim?
The LinkedIn post gives a practical example of an assessment process moving from 6 to 8 hours to 3 to 4 hours.
That should be treated as an example rather than a universal benchmark.
The actual saving will depend on:
-
Class size
-
Length of submissions
-
Number of criteria
-
Subject
-
Teacher experience
-
Complexity of the assessment
-
How extensively each draft needs to be edited
The stronger claim is therefore not a guaranteed number of hours.
It is the workflow change.
Writing from scratch becomes reviewing from evidence.
That can be measured by every teacher with their own next summative task.
A Simple Test for Your Next Assessment
Take one class and record how long the full feedback process actually takes.
Include reading, criterion matching, scoring and comment writing.
Then test a draft and review workflow on the next comparable task.
Record the time again, but also check quality.
Ask:
-
Do the comments still reference the individual student's work?
-
Are criterion and strand references accurate?
-
Did I change suggested levels when the evidence required it?
-
Does each student have a useful next step?
-
Would I be comfortable explaining the final judgement to the student, parent or coordinator?
If the process becomes faster while those answers remain yes, the tool is removing administrative repetition rather than professional judgement.
That is the part worth automating.
Student Work Still Requires Responsible Handling
Uploading student submissions into any AI assisted system also introduces a second responsibility: data protection.
The IB's draft AI principles describe learner data rights and privacy as important considerations and call for transparency about what AI systems do and where their limits lie.
Schools should therefore follow their own data protection, safeguarding and technology approval policies before placing identifiable student material in any external platform.
Koncepts currently explains its data practices through its Privacy Policy.
Read the Koncepts Privacy Policy
Teachers should always confirm what information is being uploaded, how it is processed and whether the workflow complies with school policy before using student submissions.
Feedback Should End With Student Action
There is one final risk with any feedback system.
Teachers become more efficient at producing comments that students never use.
A beautifully written paragraph still has little value if the learner reads the achievement level, closes the document and moves on.
The EEF recommends planning explicitly for how learners will receive and use feedback.
For MYP teachers, that could mean giving students ten minutes to:
-
Identify one strand to improve
-
Rewrite one section of the response
-
Annotate where they addressed the feedback
-
Correct one misconception
-
Set one target for the next assessment
The feedback loop is not:
Submit → grade → comment → finish
It should be:
Submit → assess → feedback → act → improve
Technology can make the middle of that sequence faster.
It cannot make the final step happen for the student.
The Judgement Is the Valuable Part
Criterion based assessment asks a lot from teachers because the judgement is nuanced.
That is also what makes it useful.
Reducing MYP assessment to a percentage would make marking simpler, but it would lose information about what the learner can actually demonstrate against the subject objectives.
The better question is where teacher expertise is genuinely needed.
Teachers need to interpret evidence.
They need to decide which achievement level is defensible.
They need to notice when a student's performance does not fit neatly into one band.
They need to decide what feedback will help that learner most.
They do not necessarily need to construct every sentence of every comment from an empty text box.
The IB's own direction on AI offers the right principle: keep professional judgement central while using technology where it can support the work around that judgement.
Student 27 should not receive weaker feedback because the teacher has already assessed twenty six submissions.
The goal is not automated judgement.
It is enough support around the judgement that the twenty seventh student still gets the teacher at their best.
Explore the Koncepts Assessment and Feedback tools for MYP teachers.
Related Koncepts Guides
MYP Grading Explained: Criteria A to D and the 1 to 7 Scale
Useful for understanding achievement levels, criteria and the difference between criterion scores and final grades.
IB MYP Feedback Writer: Strand Level Feedback
A detailed walkthrough of the Koncepts Feedback Writer workflow and how feedback is linked to student work.
Read the Feedback Writer guide
The Three Koncepts Tools Every IB MYP Teacher Should Know
Useful for understanding how the Unit Planner, Assessment Generator and Feedback Writer connect.
Read the MYP teacher tools guide
AI Tools for IB MYP Teachers
Explains why generic AI tools often struggle with MYP terminology, criteria and programme structure.
Read the AI tools for MYP teachers guide
IB MYP Assessment Generator
Useful for connecting better feedback with criterion aligned assessment design.
Read the MYP Assessment Generator guide
MYP Mathematics Criteria A to D
A subject specific guide showing how criterion evidence differs across Knowing and understanding, Investigating patterns, Communicating and Applying mathematics in real life contexts.
Read the MYP Mathematics criteria guide
Sources and Further Reading
Table of contents
Share
Related reading
- How to Write MYP Feedback Students Can Actually UseWrite actionable MYP feedback that connects criteria, evidence and next steps while helping IB teachers reduce report writing time without losing quality.
- IB MYP Feedback Writer for TeachersCut MYP feedback time with Koncepts Feedback Writer. Create strand-aware assessment feedback, quick comments and report comments faster. Try it free today.
- IB MYP Assessment Generator: Build Criterion-Tagged Questions with the Right Command TermsKoncepts Assessment Generator enforces MYP command terms by level band, loads criterion strand descriptors and exports separate student and teacher PDFs.
- IB MYP Feedback Writer: Strand-Level Feedback That Reads the Student's Actual WorkKoncepts Feedback Writer reads student PDFs, quotes their work to justify strand scores and writes report comments for a whole class in one batch.
- MYP Math Criterion B: Are Tasks Capping Level 4?Learn how to design MYP Math Criterion B tasks that give students access to Level 7–8 through method choice, verification, justification and proof.
