Open in app

Sign up

Sign in

Write

Sign up

Sign in

There’s no such thing as fully automated web accessibility

Protect yourself and your company from buying an ineffective product that claims it makes websites accessible.

Anna E. Cook, M.S.
UX Collective

--

An overlay over a website showing a list of toggle options that users can turn off and on for accessibility accommodations. A wand is nearby with glitter shooting out of its star.

The Overlay

It was the end of another long Thursday as my colleague, and I sat at her desk, staring at the screen. Monica and I were both Product Designers who had joined a company around the same time. We had just finished a round of user testing with some prototype screens and received some great feedback.

As designers, it was our job to design user experiences for digital platforms. Sometimes that meant designing web pages, other times it meant designing web applications or maybe native mobile applications. We would theorize on what we could create to empower users to do what they needed while also driving business outcomes, then test those ideas to verify their effectiveness. Both of us were exceptionally passionate about our careers, even though we offered different perspectives on the best way to do our jobs.

Monica was savvy at understanding complex business cases, while I was an accessibility specialist. I would stack in accessibility considerations with my designs to ensure they were accessible to disabled users. I could use the Web Content Accessibility Guidelines (WCAG) to assess our work and ensure we weren’t unintentionally excluding users. My specialty was considered peculiar for a designer, but Monica greeted it with enthusiasm.

As we sat at her desk, Monica pulled up a web page. She beamed at me with curious eyes as she said, “You’re the accessibility specialist. What do you think?”

I furrowed my brow staring long and hard into the pixels of the monitor. Monica had pulled up a website that I had never heard of before, but my attention was caught by the circle stuck to the bottom right-hand corner of the screen. It had an icon of a little person with their arms outstretched, symbolizing “accessibility.”

“What is it?” I asked her, squinting my eyes on the screen as she selected the circle, expanding a list of options.

“It’s called ‘accessiBe.’ It’s supposed to make websites fully accessible, isn’t that awesome?”

“Yes…” I mumbled pensively. My gaze was still fixed on the object overlaid on the website. The list of options Monica had expanded from the overlay showed a set of what the product called “Accessibility Adjustments.” Monica turned on the “Blind-users” accommodation toggle on the overlay, and a small ding emitted from the computer in eager confirmation.

The gears in my head began to wind faster. I shot my eyes back towards Monica and promptly asked, “So this makes this website accessible for blind users now? What does it do?”

“Yeah! Now you don’t have to do all of that extra work,” Monica smiled warmly.

“How do you know it works?” I asked slowly, careful not to emphasize my doubts too prominently.

“Well, they couldn’t sell it if it didn’t. Besides, a bunch of huge brands use it. They couldn’t be wrong.”

I nodded curtly. I didn’t have the heart to tell Monica, who seemed well-intentioned enough, that I was very skeptical of the overlay she was showing me. It seemed too good to be true.

“Wow, that sounds amazing, I’d be interested to know if it works.”

A scene from the movie the Wizard of Oz with the bright red slippers on Dorothy, as the witch’s green hands reach towards them.
Vidor, King, et al. The Wizard of Oz. Metro-Goldwyn-Mayer (MGM), 1939.

Bad products, good intentions.

It’s been a few years since Monica and I had that conversation.

However, I’ve had countless nearly identical conversations about that same product since then with many other people. Over and over again, well-intentioned people come to me to share products like accessiBe, Userway, EqualWeb, and more. They are always so eager to tell me the entire web can be made effortlessly accessible from a single overlay, using “technology powered by AI.”

The difference between then and now is that I know that these accessibility overlays don’t work. They can’t work, actually. Fully automated accessibility solutions are the magic slippers of accessibility. They sound nice, but they don’t exist and can’t exist.

Some part of me wanted to believe Monica. I wanted it to be true, even though I knew it couldn’t be. The idea of the entire internet suddenly becoming accessible to disabled people, with just a cute little icon here and a toggle button there seemed terrific. What a wonderful world it would be to access our healthcare seamlessly, get election results in real-time, or order as much pizza as we wanted.

But then no one would need to remember that disabled people exist or account for our varying needs and preferences.

No one would have to remember that humans don’t have an “average” to design for. That about 26% of the population in the United States is disabled. That many of us will become disabled in this lifetime, temporarily or permanently.

The product these companies try to sell you is fantasy. Websites cannot be made fully accessible with automated solutions.

I’ve emphasized it time and time again: accessibility is not “on” or “off.” It is more or less. A product is more or less accessible.

There is no magical solution that achieves perfect accessibility. AccessiBe, Userway, and other overlays promise “ADA & WCAG Compliance” with just a little JavaScript. As though we might tap our red slippers together and make a wish.

But accessibility doesn’t work like that. The reality is that digital accessibility requires us to ensure that fundamental elements of our products are accessible.

An overlay cannot fix accessibility issues baked in HTML. It cannot add meaningful alternative text to images or add captions to videos. It cannot fix your forms, your buttons, or your links. Even though these overlays may offer some seemingly nice-to-have functionality, they do not fix inherent accessibility issues that are baked into designs and code.

Perhaps what is more concerning about these overlays is that they forget that accessibility benefits everyone.

I can’t count the number of people in my life who have come to me to share accessible designs that they just thought of as “usable.” Some of those people are disabled, and others are not. Many of us that are benefitting from inclusive design are not disabled or don’t realize we are disabled. We often don’t even think about it in terms of accessibility.

Great design is not at odds with accessibility. Great design is accessible. AccessiBe, and other products pitching fully automated accessibility, would have you believe otherwise.

Even more, these products are offering inherently segregated experiences. They force users to select their disability to receive an “accessible” experience. Is a website truly inclusive if it asks users to disclose their disability and then be presented with a non-functional solution, segregated from “normal” users?

Many users don’t even know what audience they should select when they’re asked to self-identify. Some people don’t always know that they benefit from accessibility and others aren’t diagnosed with a disability even though they have one. The opt-in, self-identifying experience of these products requires a level of self-awareness that users tend not to have, or don’t want to think about when trying to quickly navigate a webpage

Then, what happens to the users who use this overlay?

Let’s try thinking about it in terms in terms of physical accessibility. Imagine if you ran a grocery store and hired one of these overlays to make it accessible. Instead of installing ramps, putting groceries within reach, and the like, they stood at the front door, asking each customer if they were disabled. If a customer said yes, your accessibility overlay would send them to the back of the building and make them pick out groceries from a different room with limited options.

That sounds awful. So why is a comparable experience okay on a website?

That’s the version of web accessibility these products are selling. Not an inclusive, comprehensive approach, a lie, the “othering” of people. Default accessibility considerations are what make an experience accessible, not add-ons. Defaults have the power to save lives, change lives, and empower people.

But you don’t have to take my word for it. You can reference the voices of over 650 people who advocate for accessibility, have disabilities, and know far more about this than I do alone. We have all signed the Overlay Fact Sheet to help spread the word about what these products are doing.

Haben Girma, one of the signers of the Overlay Factsheet, is a human rights lawyer, speaker, author, and a Deafblind woman who graduated from Harvard Law. She speaks to her experience using the accessiBe website/accessibility overlays, and cautions against it.

Many disabled people have signed that fact sheet and shared how accessiBe and other accessibility overlays have worsened their experiences online. In a survey conducted by WebAIM, 72% of respondents with disabilities rated these tools as not at all or not very effective.

A strong majority (67%) of respondents rate these tools as not at all or not very effective. Respondents with disabilities were even less favorable with 72% rating them not at all or not very effective, and only 2.4% rating them as very effective.

These products don’t even have a neutral impact, they are actively making the internet less accessible. Countless websites and applications make their experiences less accessible by using them because they don’t know better. I’m writing this because I want you to know better.

Most recently, the National Federation of the Blind even disavowed accessiBe due to their harmful practices.

If that’s not enough, perhaps you’d be interested in knowing that websites using accessiBe are being sued for inaccessibility. A web accessibility lawsuit can cost quite a lot of money, but even more, it can damage your reputation for years to come.

A scene from the movie Wizard of Oz where Dorothy, the Scarecrow, the Tin Man, and the Lion discover that a man pretending to be the wizard is a fake, using machines behind a curtain.
Vidor, King, et al. The Wizard of Oz. Metro-Goldwyn-Mayer (MGM), 1939.

Look behind the curtain

I’m asking you to look behind the curtain and see what fully automated accessibility solutions are — a lie.

I realize I may come across strongly in this article. I do so because I’ve watched accessiBe (and others) ignore these issues for years. But even worse, they are now claiming that anyone sharing this information is a “competitor.” Not only are they making web accessibility worse, but they are working to disempower the voices of countless people, disabled and abled, putting in the work to improve accessibility across the web.

I can’t sit back and let this company gaslight the people who need accessibility most for the sake of their own profits.

If fully automated web accessibility were actually possible, I would happily quit my job tomorrow. I would love to live in a world where things online could be accessed so easily, but right now, that’s not possible. We signed the fact sheet because we need organizations to understand that so people can use their products and services. I do this work because I believe the internet should be a place of equal access, not because I want job security. If I wanted that, I would not have chosen this path.

My work means having conversations like the one I had with Monica every single day. It means dismantling ableism both within myself and within vast systems and communities. My work means I have to write articles like this because I have to do what’s right, even though I’d much rather be playing 3 hours of Stardew Valley and going to bed early. I do this job because we need to be accountable for creating an equal and inclusive web.

Don’t waste your money on products pitching fully automated accessibility. Focus those efforts on having your internal teams learn about the Web Content Accessibility Guidelines and making damn good experiences instead.

Need more proof?

Learn more about how products claiming to be fully automated web accessibility solutions are failing users.

Resources are listed categorically and alphabetically.

In this article

In the news

Regarding legislation and lawsuits

From organizations

From accessibility specialists

Disclaimer: This article is written by myself, representing myself, and does not represent any employer I’ve worked with, currently or previously.

The UX Collective donates US$1 for each article we publish. This story contributed to World-Class Designer School: a college-level, tuition-free design school focused on preparing young and talented African designers for the local and international digital product market. Build the design community you believe in.
Design
Accessibility
A11y
UX
Ts

--

--

Anna E. Cook, M.S.
UX Collective
Follow

Written by Anna E. Cook, M.S.

976 Followers
Writer for

She/they. Senior Inclusive Designer.

Follow

Help

Status

About

Careers

Press

Blog

Privacy

Terms

Text to speech

Teams