My audit came back ugly and nobody owns accessibility. Who should?

General
1

I ship AI-assisted UI work and ran an accessibility audit that came back ugly. My buttons were unlabeled everywhere, and fixing them turned into a loop where design says it’s engineering and engineering says it’s design. With the EU Accessibility Act enforceable since June 28, 2025 for anyone selling to EU consumers, plus the WCAG AA baseline in EN 301 549, this feels urgent. Who owns accessibility when a team has no dedicated role?

2

unlabeled everywhere

Nobody needs ownership. Block merges on axe CI failures, ship one fixed component library.

3

Sam’s fix sounds clean but I have watched this exact story play out with fraud controls. Automated checks catch maybe a third of real barriers. Keyboard traps, confusing screen-reader announcements, and AI-generated UI sludge all sail through green CI. Without a named owner who holds veto power and a budget, the backlog just rots while everyone assumes the tooling handled it. And enforcement has teeth now, with the first EAA suits filed in France last November and Carrefour ordered to fix its site and app or face daily fines. Gates are necessary, but who on your team can actually stop a release, Sarah?

4

Honestly Priya, nobody can stop ours. We rotate a release captain each sprint but there’s no veto checklist, and since the axe warnings are non-blocking everyone scrolls past. In fintech is that criteria a checklist or more of a severity bar?

5

It’s a severity bar, not a checklist. Any new critical blocks the release: keyboard trap, unlabeled primary action, contrast failure on a core flow. Highs can ship only with a dated fix commitment attached. Mediums ride the normal backlog. On my team we tie the bar to WCAG AA success criteria on money-movement flows, so there is no debate. Start with just no new criticals. Which flow would you put that bar on first?

6

Putting the bar on checkout first, since that’s our money flow and the messiest screens live there. The WebAIM Million 2026 report found “95.9% of home pages had detected WCAG 2 failures” WebAIM: The WebAIM Million - The 2026 report on the accessibility of the top 1,000,000 home pages with an average of 56.1 errors per page, so “no new criticals” feels like the floor. I’m proposing the bar at next sprint planning.