Assess Against Standards, Not Just Percentages

Tag your questions with the objectives they assess, and a result stops being “62%” and becomes “not yet met: solving two-step equations — and here is what to do about it”.

No credit card required · Works with your existing question banks

A Percentage Does Not Tell a Student What to Do Next

Two students score 62% on the same paper and need completely different things — and the mark says nothing about which. Learning objectives are what turn a score into a diagnosis: each question carries the standards it assesses, each result reports attainment against them, and the material a student needs is released to them automatically. Not as a reading list bolted onto a result page, but only for the objectives they have not met yet.

How Standards-Based Assessment Works

1

Write your objectives

Add the standards you assess against — a title, an official code, and the mastery threshold that counts as met. Paste a standards document and have them drafted for you.

2

Tag your questions

Attach objectives to questions across every bank, one at a time or with AI proposing tags in bulk for you to review.

3

Report and remediate

Every result breaks down by objective, and students who fell short are given the practice material for exactly those.

What Makes the Report Worth Reading

  • Objectives live above the exam, not inside it

    Standards are written once and tagged onto questions across every bank, so "how is this class doing on this standard" is answerable across assessments rather than only within one paper.

  • Remediation is released, not listed

    A student sees the material for the objectives they missed and nothing for the ones they met. Links, videos, documents, practice sets or a written note — attached to the objective, delivered with the result.

  • A mastery threshold you set

    Each objective carries its own bar for what counts as met, so a foundational standard and a stretch one are not judged identically.

  • Build a paper by standard

    Draw questions tagged with a chosen objective straight into an exam, and see what the draw would actually find before you save — including the step that emptied it, when a filter matches nothing.

  • A result is a record of what was assessed at the time

    The objective title and code are copied onto the result, so renaming or retiring a standard next year does not rewrite what a student was told last year.

  • AI proposes; a teacher decides

    AI can draft an objective from a phrase, pull many out of a pasted standards document, and suggest tags across a whole bank. Every one of them is a proposal you accept or reject — nothing is tagged on your behalf, because a mastery judgement has to be defensible to a parent.

  • Invented standard codes are discarded, not shown

    A code the model cannot vouch for as a real standard is dropped rather than displayed. A teacher reads a code as authoritative and will never check it, so a plausible invention is worse than a blank field.

  • Item analysis alongside it

    See how each question performed — difficulty and percent correct — so a standard that looks unmet across a class can be told apart from a question that was simply badly written.

Who Assesses This Way

K-12 Schools

Report against state or national standards, with practice material released to the students who need it.

Districts & Networks

One set of objectives shared across schools, so attainment means the same thing everywhere it is reported.

Universities

Map assessment to programme learning outcomes and evidence them for accreditation.

Professional Training

Competency frameworks where a candidate must demonstrate each one, not average across them.

Frequently Asked Questions

Reporting what a student can and cannot yet do, rather than a single percentage for the whole paper. Each question is tagged with the objectives it assesses, and the result reports attainment against each of them separately — so two students on the same mark can be given different next steps.

Yes. Paste a standards document and AI extracts the objectives from it for you to review and accept, rather than typing them one at a time. You can also write them by hand, each with a title, a code, and its own mastery threshold.

No — that is the usual blocker, so AI will propose tags across a whole bank and you review them. Only proposals it is confident about are pre-ticked, because pre-ticking everything turns review into a rubber stamp, and it is told to leave a question untagged rather than guess: no tag is better than a wrong one, which would drag a question into a mastery judgement it does not belong in.

No. Mastery is a straightforward comparison against the threshold you set for that objective, and on the line counts as met. AI is only ever used to propose objectives and tags for a teacher to accept — the judgement itself is arithmetic you can check.

Their result, broken down by objective, showing which were met and which were not — and, for the ones that were not, the practice material you attached to those objectives. Material for objectives they met is not shown, so the list they get is short and is about them.

Yes. A bank draw can be filtered to questions tagged with a chosen objective, and the builder shows what that draw would actually find before you save — naming the filter that emptied it, if it comes up short.

Past results are unaffected. The objective title and code are copied onto each result when it is graded, because a result is a record of what was assessed at the time — changing this year’s wording must not rewrite what a student was told last year.

Turn Marks Into Next Steps

Tag your questions with the standards they assess, and let every result say what a student should do next. Free to start, and no credit card required.

Get Started Free