What goes in the caption rows of a VPAT

A VPAT — filled in, it becomes an Accessibility Conformance Report — asks, criterion by criterion, what your product supports. The caption rows are where video-heavy organisations stall, because the honest answer depends on evidence most teams do not have in a form they can cite.

Which rows are the caption rows

A VPAT covers the whole of the standard, but on a video library these are the rows a reviewer reads first — and the ones most often left blank.

Cited from W3C Web Content Accessibility Guidelines 2.1, Level AA (W3C Recommendation, 5 June 2018) — read the standard.

Which level the evidence actually supports

A row takes one of four levels: Supports, Partially Supports, Does Not Support, Not Applicable. The temptation is to write Supports whenever a scan came back clean. That is the single most common way a caption row becomes indefensible, because for most caption criteria a clean scan is not a proof — the file can show the criterion failing and cannot show it passing.

This table is computed by calling the same function that grades a CaptionPass conformance report, so it cannot describe a mapping the product does not apply.

Not Evaluated, not Not Applicable

When nothing you hold can settle a criterion, the row is Not Evaluated, with the reason — not Not Applicable. WCAG 1.2.4 is entirely applicable to an organisation that streams live; writing N/A there is exactly the unjustified row that gets a response thrown out.

Not Evaluated is not one of the four standard levels, and that is the point. It sits visibly outside the vendor’s own scoring, which is the honest place for it, and the remark says who has to answer the question instead.

Reviewers in higher education and government are trained to read blank caption rows and unexplained N/A as the place where vendors hide. A stated limitation reads as candour. A gap reads as evasion — and the reviewer decides which, not you.

Getting the evidence

The rows have to come from somewhere. A per-asset receipt records what one delivery was checked against; a conformance report rolls a library up into exactly the table above, failures included. If you are filling this in for a public body under a deadline, start from evidencing ADA Title II captions.

What this is not. A conformance receipt is evidence of what was checked, by which engine, against which version of which standard, and what it found. It is not a legal opinion, an audit, or a safe harbour. CaptionPass reads the caption file and never the audio, so anything that needs the audio — above all whether the captions match what was said — is reported as not evaluated rather than assumed.