The Department of Health and Human Services recently announced a one-year extension of the compliance dates for web content and mobile app accessibility requirements under Section 504 of the Rehabilitation Act. The requirements themselves are not new in substance: covered recipients of HHS federal financial assistance must make covered web content and mobile apps conform to WCAG 2.1 Level AA. What changed is the timeline. Recipients with 15 or more employees now have until May 11, 2027, and recipients with fewer than 15 employees now have until May 10, 2028. This follows an almost identical extension for ADA Title II.
For many accessibility advocates, this delay is understandably frustrating. People with disabilities have already waited far too long for equal access to digital services. Another year is not an abstraction. It may mean another year of inaccessible patient portals, appointment systems, application forms, mobile apps, PDFs, and online health information. These are not minor inconveniences. In many cases, these are the digital front doors to health care, public benefits, education, social services, and other essential programs.
It is also fair to ask why more time should be necessary. WCAG 1.0 has been around as long as anyone up to age 30 can remember. WCAG 2.0 has been with us since 2008. WCAG 2.1, the standard adopted in the HHS rule, was published in 2018. The web has been redesigned, rebuilt, replatformed, rebranded, and relaunched many times since then. Every inaccessible redesign has been a lost opportunity. Every inaccessible PDF uploaded last week is not a relic of the 1990s. Every new form that cannot be completed with a keyboard, every button without an accessible name, every low-contrast interface, and every inaccessible mobile app workflow represents a choice, whether intentional or not, to continue excluding some users.
Recent data make the frustration even sharper. As my colleague, Christopher Phillips, reported in this space last month, the 2026 WebAIM Million report found that among the top one million website home pages, the average number of WAVE-detected accessibility errors rose about 10% from 2025 to 2026. Christopher rightly notes that this means the web became less accessible than it was a year ago, at least by this measure.
That should bother us. It is difficult to ask disabled users for patience when the broader web is not merely moving slowly, but in some measurable ways moving backward. Patience is much easier to request from the people who are not blocked by the form, the portal, the document, or the checkout process.
And yet, there is another reality that accessibility professionals encounter every day. Organizations are not simply sitting on a secret pile of accessible code and choosing not to deploy it. Some website owners only recently learned what digital accessibility requires. Some built their sites quickly with low-cost tools. Some bought scheduling systems, patient portals, learning management systems, donation platforms, document management systems, or mobile apps without realizing that accessibility would become a serious barrier later. Some inherited thousands of inaccessible PDFs. Some have only one person responsible for the website, and that person may also be responsible for communications, social media, IT support, and refilling the copier paper when it jams.
This does not excuse inaccessibility. It does help explain why meaningful accessibility work can take time.
HHS itself points to practical concerns that many organizations face. The Department notes that recipients vary widely in size, resources, location, technical support, and mission. It also cites concerns about the time and cost of PDF remediation, uncertainty about how compliance will be measured, and the difficulty of ensuring that third-party contractors make web content conform to WCAG 2.1.
Anyone who has worked in accessibility consulting will recognize those issues. Accessibility is not usually fixed by one person running one automated scan on one afternoon. Real progress often requires auditing, remediation, design changes, development changes, content changes, procurement changes, staff training, vendor pressure, policy updates, and ongoing monitoring. Organizations must not only fix what already exists; they must also stop creating new barriers. That is often the harder cultural shift.
So yes, some organizations need time. But time for what? A deadline extension may be understandable as a practical matter, but it is only defensible if the additional year is used for meaningful progress. It must not become another year of waiting, deferring, debating, or hoping the problem will somehow become smaller on its own.
The HHS notice includes one comment that should stop us cold. A virtual mental health care provider argued that the 2024 rule would impose substantial financial burdens on health care providers “without providing any material benefits.” HHS explicitly disagreed, noting that especially in telehealth, if a person with a disability cannot access the web content or a mobile app from a telehealth provider receiving federal funds, this is not just a technical issue, but a denial of health care by a recipient of federal dollars.
That phrase, “without providing any material benefits,” reveals the deeper problem. Accessibility can look optional only when the people excluded by inaccessible systems are treated as invisible. For a health care provider, the material benefit should be obvious: patients can get care.
This is where organizations need to be honest with themselves. The legal deadline may have moved, but the human needs did not. A blind patient who cannot use a provider’s portal does not become less blind because the compliance date changed. A person with limited dexterity who cannot complete an online form does not receive better service because the organization has another year. A Deaf patient who needs accessible video content, a low-vision user who needs sufficient contrast and text resizing, or a screen reader user who needs a properly labeled form field is not asking for advanced features. They are asking to use the service.
The extension should therefore be treated not as a reprieve from accessibility, but as a reprieve from panic. Panic produces rushed audits, superficial fixes, inaccessible redesigns, and the kind of magical thinking that leads people to believe that one simplistic overlay can solve years of inaccessible design and development. A thoughtful year can produce something much better: a prioritized plan, trained staff, better procurement practices, improved templates, cleaner content workflows, more accessible third-party systems, and measurable reduction of barriers.
Organizations that are covered by these requirements should use the additional time wisely.
First, prioritize the services that matter most. Not every page, document, or app screen has the same immediate impact. Appointment scheduling, patient intake forms, benefit applications, payment systems, contact forms, complaint processes, emergency information, and required documents should rise to the top. If a barrier prevents someone from receiving care or participating in a program, it should not wait until the end of the project plan.
Second, stop creating new inaccessible content. An organization that spends the next year remediating old PDFs while continuing to publish new inaccessible PDFs is bailing water with the faucet still running. Templates, authoring practices, document workflows, and publishing permissions all matter.
Third, take procurement seriously. Many accessibility problems enter organizations through third-party products. Website platforms, form builders, scheduling systems, mobile apps, map widgets, payment processors, document systems, and embedded media players can all create barriers. Organizations should ask vendors direct questions, require credible accessibility documentation (such as a VPAT), include accessibility in contracts, and test important workflows before purchase. “Our vendor handles that” is not a strategy unless the vendor actually handles it.
Fourth, train the people who create and maintain the content. Accessibility cannot live only with the web developer or an outside consultant. The person who writes headings, adds links, uploads PDFs, creates social media graphics, posts videos, or builds forms can either reduce barriers or multiply them. Basic training for content authors can prevent many common problems before they reach the public.
Finally, measure progress honestly. No organization will fix everything in the first month. But by the end of the next year, organizations should be able to show what they tested, what they fixed, what remains, who is responsible, and how accessibility will be maintained. Progress does not need to be perfect, but it does need to be real.
The disability community has heard promises before. “We need more time” is only credible when it is paired with action. “We are working on it” only means something when barriers are actually being removed. The deadline moved. The civil rights obligation did not. For organizations that are serious about accessibility, the approach should not be how little they can do before 2027 or 2028, but how much.