Accessibility
Version 1.0 · Last tested 6 September 2026
1. What we are aiming for
Sorted Permits targets WCAG 2.1 Level AA. That is the standard Victorian councils hold themselves to, and if a council links to this tool from its own website, our accessibility becomes part of theirs. We would rather tell you exactly where we stand than claim a conformance level nobody has checked.
2. How we tested
Testing was done on the built pages as they deploy, not on a staging version, on 6 September 2026:
- Automated. axe-core against the home page, the app, the guide hub, a guide topic page, and both legal pages, each in light and dark colour schemes at 1280px and 360px wide.
- The app in use. The app was driven through the states an organiser actually reaches: the intro, two wizard steps, the results screen and each of its six tabs, an expanded requirement, and an open form panel. A tool that passes on its landing screen and fails once you use it has not been tested.
- By keyboard. Twenty-two focus stops checked for a visible focus indicator, the skip link, and arrow-key operation of the results tabs.
- Under stress. Text resized to 200%, a 320px viewport, reduced-motion preference on, and JavaScript disabled.
3. What we found and fixed
Three things failed and all three are fixed:
- The results tabs were announced but not operable. The app's results screen declared itself a tab list without giving screen readers the roles and states to work with, and arrow keys did nothing. The tabs now carry the correct roles, the selected tab is announced, arrow keys and Home and End move between them, and only the open panel is exposed to assistive technology.
- Six pieces of text sat below the contrast minimum, between 3.17:1 and 4.27:1 where the standard asks for 4.5:1. These were muted greys on the legal pages and the amber badge on guide callouts. Every replacement colour was measured against its actual background rather than eyeballed.
- Most of the home page was invisible without JavaScript. Forty-five of fifty-one content blocks waited on a scroll animation that never ran. The animation is decoration; the words are the point, and they now show either way.
4. What is still not right
At browser windows narrower than about 328 pixels, the home page reports roughly eight pixels of horizontal overflow. We have checked that no text, control or link is cut off by it: the overflow comes from a decorative element behind the content. It is still a technical failure of 1.4.10 Reflow and we would rather list it than leave it out. It is on the list to fix.
Two further limits worth stating plainly. We have not had an independent accessibility audit, and we have not tested with real assistive technology users. Automated tools catch perhaps a third of what matters. When a council pilots Sorted Permits, testing with your own accessibility team is something we would welcome rather than resist.
5. Documents we generate
The app generates Word documents from your answers. Those documents follow the council's own template structures, and their accessibility is inherited from those templates rather than designed by us. If you need generated documents to meet a particular standard, tell us and we will work to it.
6. Where we know we depend on others
Two third-party components appear on the site: privacy-preserving analytics, which renders nothing visible, and Cloudflare Turnstile on the enquiry forms, which is a deliberately low-interaction alternative to a captcha. If Turnstile ever becomes a barrier for you, every form on this site keeps a plain email link beside it that reaches us just as well.
7. Tell us when we get it wrong
If any part of Sorted Permits is difficult or impossible for you to use, email hello@sortedpermits.com.au and say what happened and what you were trying to do. We will reply within five business days, tell you honestly whether and when we can fix it, and give you another way to get the same result in the meantime.
8. Keeping this honest
This page is re-tested whenever the site or the app changes materially, and the date at the top is the date of the last test, not the date it was written. If it says something that is no longer true, that is a bug and we want to hear about it.