Accessibility statement
Last updated: 25 September 2026. Privacy policy · Legal notice
Volunteers do not choose this platform: their organisation chooses it for them. It seems only fair that they should be able to sign up with a keyboard, a magnifier or a screen reader. This page says where we stand, how we measure it, and what the measurement leaves out.
Conformance status
Volunteo is fully conformant with the French RGAA 4.1, by self-assessment: 100% of the applicable criteria are met, 87 out of 87, the other 19 not applying to this product. The RGAA covers WCAG 2.1 level AA.
"Self-assessment" is the word that matters: the grid was drawn up by us, not by a third party, and 49 of the 106 criteria are judged by hand, because no machine decides whether a label or an alternative is apt. The other 38 are measured on every run, over the 190 screens the product has. The breakdown is below, and so is the list of what this conformance does not say: it matters as much as the figure.
This statement is voluntary. The French accessibility obligation (article 47 of the law of 11 February 2005) applies to public bodies and large companies, not to a software publisher of our size. It does apply to some of our customers, local authorities and publicly funded associations, whom a tool that says nothing about its accessibility does not help.
Measurements
Self-assessment of 25 September 2026, re-run whenever a screen changes. Sample: the 190 screens of the product, that is all of them, not a selection of pages.
| Status | Criteria |
|---|---|
| Conformant | 87 |
| Non-conformant | 0 |
| Not applicable (no time-based media, no CAPTCHA, no time limit, no cryptic content, no titled table) | 19 |
The automated measurements behind that grid:
| What is measured | Screens | Result |
|---|---|---|
| WCAG 2.0 and 2.1 rules, levels A and AA, that a machine can check (axe-core): contrast, accessible names, ARIA attributes, heading and form structure | 54 | no violation |
| Full keyboard walk: tab order, visible focus, traps, skip link, mouse-only controls, meaningless alternative text | 190 | none |
| Script errors and failed requests, which leave a screen half usable | 77 | none |
What is in place on every screen
- a "Skip to content" link as the first tab stop, and a
mainlandmark for screen readers; - a visible focus indicator, verified by comparing two images of the same element with and without focus;
- no keyboard trap, and no control that only a mouse can reach: where dragging exists (the availability calendar, the global schedule), it doubles a form that does the same thing;
- text contrast of at least 4.5:1, including on the team colours the server computes on the fly;
- animation switched off when the visitor's system asks for less motion (
prefers-reduced-motion); - pages readable at 375 px wide, with no horizontal scrolling;
- the site and the application in French, English and Breton.
What this conformance does not say
This is the part that matters most in an honest statement. A hundred per cent of the criteria met is not a hundred per cent of the way: the reference measures what can be checked on the rendered code.
- No audit by a third party. The grid is a self-assessment, and 49 of its 106 criteria are judged by hand.
- No testing with an actual screen reader (NVDA, JAWS, VoiceOver) has been done so far.
- No testing with disabled people. A machine tells you whether a journey is possible, not whether it is workable.
- Label quality cannot be measured. "Edit" is an accessible name; "Edit Paul Martin's permissions" is a good accessible name, and no tool tells them apart. We review these by hand, with no guarantee of having seen them all.
- The rich text editor used for emails to volunteers is a third-party library (CKEditor 5) whose internal accessibility is not ours to control.
- Generated documents (badges, run sheets, PDF exports) are laid out for paper and have not been checked as accessible documents.
- Content written by organisers in their sign-up pages, emails and documents is not ours: alternative text missing from an image they upload stays missing.
How we measure it
The measurements are scripts in the application's repository, run in a real browser against a
complete data set: rgaaAudit.js, which holds the grid of the 106 criteria,
axeAudit.js (axe-core 4.13), keyboardAudit.js and
consoleAudit.js. They are re-run before any release that touches a screen, and the
inventory of screens is kept up to date from the application's routes, so that no screen can
escape measurement for want of data to open it.
The criteria wordings come from the official file published by the French DINUM, not from a paraphrase: two criteria changed meaning between RGAA 4.0 and 4.1, and quoting them from memory would have given a false result.
A criterion is declared not applicable by measurement, not by decision: a probe looks for its subject across the 190 screens, and the criterion is set aside only if it finds nothing anywhere.
Reporting a problem
If you have trouble using Volunteo, write to us through the contact form. Tell us the page, what you were trying to do and, if you know it, the assistive hardware or software you use. We answer every message, and an accessibility problem goes to the front of the queue.
An organiser who has to document their own tools for RGAA purposes can ask us for the detail of these measurements through the same form.