The conversation around accessibility in social housing has too often centred on ramps, handrails and physical adaptations. But as housing providers increasingly turn to AI-powered digital tools to manage resident communications, a new frontier of inclusion — and exclusion — has quietly opened up.
At askporter, we believe that AI should lower barriers, not create them. That's why we've been investing in meaningful accessibility improvements to our platform — and why I think the wider sector needs to take this issue far more seriously.
The scale of digital exclusion is striking. Research from Acquia's 2024 survey found that 89% of users have encountered accessibility issues when interacting with digital services. And when they hit those barriers, the consequences are real: 51% seek out an alternative accessible option, and 42% simply stop using that service altogether.
In a housing context, this isn't just a UX problem. It's a compliance issue, a safeguarding concern, and — frankly — a failure of the duty of care that residents deserve.
The Tpas National Engagement Standards (4th Edition, 2024) are clear on this. As the standards set out: "Meeting the diverse needs of residents, removing barriers to engagement and providing information in plain language is non-negotiable." Accessibility in digital channels isn't a nice-to-have. It's embedded in the framework that housing providers are now held to account on under the Social Housing (Regulation) Act 2023.
"If the technology isn't built to work for the residents who need it most, the promise of 24/7 responsive service becomes, for those residents, no service at all." — Harry Dodd-Noble, Chief Product Officer, askporter
Accessibility isn't an edge case
Most people don't think about what it's like to use a webchat with a screen reader, switch control, or voice navigation. But for a significant portion of your residents, that's not a hypothetical, it's how they're trying to reach you. These are often the same residents who most need reliable, accessible digital services. Our webchat is built to work for all of them.
Page titles
Take something as simple as a page title. If you're a blind resident with several tabs open, trying to find the right window, a vague or missing title means guessing. A clear one means finding what you need without the friction.
Screen reader announcements
Consider what happens when a bot replies. For a sighted user, the response just appears. For someone using a screen reader, the old behaviour meant they'd have to manually move away from the input field to discover a reply had arrived — breaking the flow of a conversation that might involve a repair request or something more urgent. That's fixed now. Responses are announced automatically, and the conversation moves as it should.
Structure without styles
We also looked at how the chat holds up when styles are stripped away entirely — which is how some assistive technologies experience the web. The structure now makes sense on its own, without relying on visual design to carry meaning.
What this means in practice
None of these changes are visible to most users. That's rather the point.
All updates have been tested with VoiceOver on Apple devices and Google's screen reader on Android. We'll keep going — accessibility is a standard, not a milestone.
There is a structural problem here that extends beyond any single platform. While AI adoption in housing and property management is accelerating, the focus on ethical and accessible implementation is lagging. Fewer than one in five organisations in the built environment have implemented AI policies that explicitly address ethical implications or accessibility considerations.
That gap has real consequences. Research by Fable found that half of disabled users who have engaged with AI technology encountered barriers or challenges because of their disability — and only 7% believe there is adequate representation of disabled people in the development of the AI tools they're being asked to use.
Housing associations have a particular responsibility here. The residents most likely to rely on AI-powered communication channels — for repairs, queries, complaints — are often those with the greatest support needs. If the technology isn't built to work for them, the promise of 24/7 responsive service becomes, for those residents, no service at all.
There is also a straightforward business case. The Tpas framework makes clear that meeting diverse resident needs and removing barriers to engagement is non-negotiable. As AI becomes a primary contact channel for housing providers, the technology underpinning those channels must reflect that standard.
At askporter, we're committed to continuing to test, iterate and improve. As Tpas Chief Executive Jenny Osbourne MBE puts it: "Excellent tenant engagement is a journey, not an outcome." The same is true of accessible technology — it is not a project with an end date, it's an ongoing commitment. But I also want to be honest with the sector: these improvements require deliberate effort, specialist testing, and a genuine willingness to prioritise inclusion.
The question for housing providers evaluating AI technology is not just "does it work?" It should be: "Does it work for everyone?"
Harry Dodd-Noble is Chief Product Officer at askporter, an AI operations platform built for housing and facilities management. Find out more at askporter.com.