Education & TrainingFree
Accessibility Checklist for Course Materials
Send the documents, slides and media in the course, and get the findings grouped by fix rather than listed by file.
Every accessibility checker reports per file, because that is the unit it opens. Open a document, list its problems, move to the next document. Thirty-four files come back as thirty-four sections and the summary line says 211 issues, which reads as 211 separate jobs. Nothing in that report tells you that eighty of those issues share three causes. So the remediation gets sequenced the way the report happened to be written, one document at a time, and the same decision gets made over and over.
BIOL 210 has eighteen PDFs, nine slide decks, five videos and two spreadsheets. Grouped by remediation type the 211 issues collapse into nine types, and three of them are 141 issues, or 66.82 percent of the total. Alt text is 74 of those. But the 74 occurrences are only 41 unique images, because the course reuses figures across decks and readings, so 33 of them are the same picture already described somewhere else.
That changes the cost. Priced per occurrence at three minutes each, alt text is 222 minutes. Priced per unique asset it is 139.5. The nine contrast failures trace to four slide masters, so 225 minutes becomes 100. Twenty-eight of the 41 heading failures come from one publisher template. File by file the course is 21.05 hours of work. Grouped by fix it is 15.525 hours, a saving of 5.525 hours, and the arithmetic prints its rates.
Two point three seven percent of the issues are the ones that actually block a student
Count and impact are different axes, and a per-file report conflates them. Five of the 211 issues are uncaptioned videos, which is 2.37 percent of the list and the only category that stops a student using the material at all. The 74 alt-text issues degrade the experience badly and none of them locks anyone out. So the priority order is not the frequency order, and the review reports both columns rather than sorting by the one a checker happens to produce.
The obligation is now dated rather than aspirational, which changes how the queue gets planned. Under the ADA Title II rule, public entities of 50,000 or more must have web content conforming to Level A and Level AA of WCAG 2.1 beginning April 26, 2027, and smaller entities from April 26, 2028. Section 508 already reaches this material directly: electronic content including educational or training materials shall conform to Level A and Level AA of WCAG 2.0.
Grouping by fix also changes who does the work. Alt text for 41 unique figures is a subject matter task and it has to be the instructor, because a description of a mitotic spindle is not generic. Correcting a style map, adding table header rows and repairing document metadata are mechanical and can go to anyone. Captions and an image-only scanned PDF need a specialist or a service. Three different queues, and the file-by-file report mixes all three on every page.
How it works
Send the course
The documents, slides, media and spreadsheets, plus any checker output you already have.
Group by fix
Findings reorganised by remediation type, with the files each type appears in listed.
Collapse the repeats
Duplicate assets and shared templates identified, so a single correction closes many findings.
Order by impact
What blocks access first, what degrades it next, and what is a referral rather than a task.
What you get
- Every finding grouped by remediation type rather than by the file it was found in
- Repeated assets deduplicated, so one description closes every occurrence of the same figure
- Findings traced back to a shared template or master where one correction fixes many
- Priority by student impact reported as its own column, separate from priority by count
- The work split into three queues: instructor judgement, mechanical correction, specialist referral
- An estimate per remediation type, with the per-item rate it was computed from
Common questions
Does this replace an automated accessibility checker?
No, and it works better with one. A checker finds issues reliably and reports them in the wrong shape for doing the work. This takes that output, or the files directly, and reorganises it around causes and fixes. Where you already have a report, sending it saves the detection pass entirely and goes straight to the grouping.
Why does grouping by fix save so much time?
Because remediation has setup costs and repeated decisions, and file-by-file ordering pays both every time. On BIOL 210 the 74 alt-text findings are 41 unique images, nine contrast failures are four slide masters, and 28 heading failures are one publisher template. The same work priced by cause is 15.525 hours instead of 21.05.
Which issues should go first?
The ones that block access, not the ones there are most of. Five uncaptioned videos are 2.37 percent of the findings and the only category that shuts a student out completely. The review reports impact and count as separate columns so the ordering is a decision rather than an artifact of whatever the checker sorted by.
Can it fix the files, or does it only find things?
It writes the remediation specification and the content that needs judgement, including draft alt text per unique figure for you to correct. Rewriting the files themselves is a separate step, and where a whole course is being repackaged LMS migration and package conversion is the one that moves the artifacts.
What about captions and audio description?
They are named as specialist referrals with the video list attached, rather than estimated alongside the mechanical work. Caption quality is its own discipline and an auto-generated transcript is a starting point rather than a deliverable. The review states which items need it and keeps them out of the internal hours estimate.
Does the same approach work outside a course?
Yes, and it is the same arithmetic. For a product interface rather than a set of documents, the accessibility audit and remediation pack carries the equivalent structure with component-level grouping instead of asset-level. The grouping principle is what transfers, not the checklist.
Should this happen during design or after?
Before, if there is any choice, because a template corrected once never generates the finding again. The course design pack is where those decisions sit, and the student workload calculator is the other check worth running on a course before it is published rather than after.
Accessibility Checklist for Course Materials
Fill in the form and your workspace opens with the work already underway.