The Organisation That Never Prioritised It
Engineering fails. Design fails. But both of those failures have a common source: organisations that never valued accessibility in the first place.
The first piece in this series was about engineering. The second was about design. But engineering doesn’t hire itself. Design leadership doesn’t set its own values in a vacuum. Both failures have a source. And the source is the organisation above them.
This is where the problem actually lives.
The Agency Problem
Start with agencies and consultancies, because the failure mode there is the most naked.
A client brief arrives. It has a budget, a timeline, a list of deliverables. It does not mention accessibility. The agency does not add it. Not because they’ve weighed it up and decided it’s out of scope. Because nobody in the room has the experience to know it should be there. Accessibility isn’t being deprioritised. It’s simply not part of how they think about the work.
That’s a skills problem dressed up as a process problem. And it runs deeper than a single project.
Because when a client does ask for it (and increasingly, under the EAA, some of them are), the agency has a problem. There’s nobody in-house who can deliver it properly. No front-end engineer with genuine accessibility depth. No designer who knows the difference between annotating for accessibility and designing for it. What happens instead is a scramble: a subcontractor brought in late, an audit bolted on at the end, a compliance report produced by someone who ran a tool and called it done. The client asked the right question. The agency couldn’t answer it.
They ship something anyway. It goes live. The audit report goes in the appendix. The people who can’t use it never appear in the case study.
The Enterprise Problem
Then there’s the enterprise.
Different failure mode. Same outcome.
Large organisations don’t scope accessibility out — they bury it. Not in a single decision but in the accumulated weight of process, governance, and the particular kind of leadership that gets rewarded for not changing things.
The thinking isn’t modern. It doesn’t need to be. Nobody has required it to be. Digital transformation arrived at these organisations as a phrase before it arrived as a reality, and in many of them it still hasn’t fully landed. There are people in decision-making positions at major organisations who genuinely do not understand what a screen reader is, why it matters, or that there are legal obligations around it. They have survived this long without knowing. The organisation rewarded them anyway.
Below them, accessibility teams exist. Sometimes. They produce guidelines. They run workshops. They write standards documents that nobody reads. They push from the bottom of the hierarchy and find, repeatedly, that the organisation above them is not structured to respond.
Because the process wasn’t built for it. Governance cycles that predate WCAG. Procurement frameworks that don’t assess for it. Sign-off chains where accessibility never appears as a criterion. You can have the most capable accessibility specialist in the country embedded in an enterprise and watch them achieve almost nothing, because the organisation they’re embedded in was never designed to act on what they find.
That’s what glacial looks like from the inside. Not stupidity. Structure. A set of processes and incentives and reporting lines built for a different era and never fundamentally questioned.
The engineer who doesn’t understand semantic HTML didn’t appear from nowhere. They were hired by an organisation that decided front-end specialism wasn’t worth the cost. The creative director who removed focus outlines didn’t operate in a vacuum. They worked in an organisation that never made accessibility a value — never put it in a brief, never raised it in a crit, never included it in what counted as excellent work.
The disciplines fail because the organisation permits the failure. Sometimes actively, by scoping it out. More often passively, by never requiring anything different.
Some organisations are getting this right. Not many, but some. The ones that are tend to share a common trait: somebody at a senior enough level decided accessibility was non-negotiable, and built the conditions for that to be true. Not a workshop. Not a policy document. A sustained, structural commitment that changed what the organisation hired for, what it measured, what it rewarded.
That’s not a small thing. It requires leadership that understands what it’s committing to, and that’s rare in organisations where digital literacy thins out quickly above a certain level.
For the majority — the agencies still leaving it out of scope, the enterprises still running governance processes designed for a pre-digital world — the European Accessibility Act provided a legal nudge. Some of them will respond to that nudge. More of them will find the minimum viable compliance response and treat the problem as solved.
It isn’t solved. It’s documented.
An accessibility statement in the footer and a completed audit checklist do not make a product accessible. They make it defensible. That distinction has been the subject of this series from the start. But it starts here, at the organisational level, before a brief is written, before a designer opens Figma, before an engineer writes a line of code.
The organisation that never prioritised accessibility will not produce accessible products. Not through any individual failure of the people inside it, but because the system was never built to. And systems produce what they’re designed to produce.
Unless someone inside decides to build differently anyway. Designers who ask the question the brief didn’t ask. Engineers who fix what wasn’t required. Accessibility specialists who refuse to let the governance cycle be the last word. They’re not fixing the system. But they’re doing it anyway — side of desk, under the radar, between the things the organisation actually asked for. That matters. It’s just not a substitute for the organisation doing its job.