Draft — not yet in force. This document is being prepared and has not been finalized. It does not currently form an agreement, and the details it refers to are incomplete.
In short: We aim to meet WCAG 2.2 Level AA on the pages members of the public use — surveys, project showcases, organization portals and shared results. We are not fully conformant, and the known gaps are listed below rather than left for you to find. If something stops you taking part, tell us and we will find another way to hear from you.
This statement applies to the pages Viewshed serves to members of the public on behalf of the organizations that use it:
The signed-in authoring application — where staff build surveys and design pages — has not had the same page-by-page review, though it is covered by our automated checks and has no known failures. Section 5 says what that does and does not mean. If your organization needs the authoring side formally assessed before purchase, ask us and we will tell you honestly where it stands.
Target: Web Content Accessibility Guidelines (WCAG) 2.2, Level AA. That target also covers WCAG 2.0 and 2.1 Level AA, which is what Section 508 and EN 301 549 reference.
Status: partially conformant. Partially conformant means most of the standard is met and some parts are not. The parts that are not are in section 4.
A criterion-by-criterion Accessibility Conformance Report (VPAT) accompanies this statement, in the format US public procurement asks for. It reports the same problems this page describes, against the numbered criteria. Organizations running consultations on our platform are welcome to attach it to their own proposals.
The public pages were reviewed against WCAG 2.1 and 2.2 Level AA and rebuilt where they failed. The substantive changes:
These are the accessibility problems we know about on public pages.
| What | Who it affects | Where we are |
|---|---|---|
| Chart colours are pale. Several series sit below the 3:1 contrast the standard asks of a meaningful graphic. | Anyone with low vision judging a bar by its length, and anyone reading a projected screen from the back of a room. | Every value is printed beside its mark, every series is labelled, and each chart has a full text description — so the figures are all available. Two of the themes we ship, “High contrast” and “Charcoal”, clear the standard on every colour including the chart palette, and the editor shows the measured ratio while an organization chooses. |
| Maps can be panned and zoomed from the keyboard, but the map is a single stop in the tab order — you cannot tab between individual markers. | Keyboard and switch users who want to move through the marked places one at a time. | Everything marked on a map is listed as text beneath it, in the same order, so the markers are reachable as ordinary content even though the pins are not. |
| Signature questions accept a typed name instead of a drawn one. | Anyone who cannot draw with a pointer. | Working as intended, but worth stating: a typed name is what is recorded, and the organization asking for it decides whether that satisfies them. |
| An organization chooses its own brand colours, and may choose ones with poor contrast. | Anyone reading that organization's pages. | We show the contrast ratio and a warning while they choose, and we never override the choice. See section 6. |
| Uploaded images depend on the organization writing a description. | Screen reader users. | We provide the field and prompt for it. We cannot write it for them — see section 6. |
| No independent audit, and no testing with a formal panel of disabled users. | Everyone, potentially — self-assessment finds less than real use does. | Assessed by us, with automated checks and manual keyboard and screen reader testing. |
The signed-in application used by an organization’s own staff has not been through the same page-by-page review as the public pages. It is not unexamined, though: the automated checks that run on every change cover the whole codebase, and the two issues they found have been fixed — inline fields in the showcase and infographic editors that lost their focus indicator, and tabbed sections that did not state which panel each tab opened.
What that means in practice: there is no known accessibility failure in the authoring application, and also no claim that a full review would find none. If a member of your staff needs an accommodation to use it, contact us — we will prioritise the fix and, in the meantime, help you get the work done another way.
Accessibility here has two halves, and it is not useful to pretend otherwise. Section 5 of the Terms of Service makes the Customer responsible for its own ADA and Section 508 obligations. That is accurate for what an organization publishes. It is not a disclaimer of our own responsibility for how the software behaves, and we do not read it as one.
| Ours | The organization's |
|---|---|
| How every page is built: keyboard operation, focus order and visibility, semantics and roles, error handling, motion, reflow, and the contrast of the interface we control. | The words, images and colours they put on the page: alt text for their photographs, plain-language question wording, brand colours with adequate contrast, and captions or transcripts for any media they embed. |
| Giving them the tools to get their half right — an alt text field beside every upload, a contrast warning beside every colour, and a chart that labels itself. | Using those tools, and responding to accessibility requests from their own community about their own content. |
If you are unsure which half a problem falls into, report it to us anyway. We would rather route it than have it go unreported.
If something on one of these pages stops you taking part, we want to know — and we want you heard regardless. Email [hello@example.com] with the page address and what happened. You do not need to know the technical name for it; “I could not get past the second question with my keyboard” is a perfect report.
We aim to reply within five working days. If the barrier is on a specific project’s page and we cannot remove it quickly, we will tell you which organization is running that project so you can give them your views directly — a consultation you could not take part in is a consultation that did not hear from you, and that is the actual harm.
Self-evaluation against WCAG 2.2 Level AA, using a combination of:
Automated testing catches roughly a third of real barriers. The rest of this statement rests on the manual work, and the honest limitation is that it was done by the people who wrote the software rather than by the people who rely on it.