Inclusive Web Design: Why Anchor & Alpine Prefers a Ground-Up Approach
Beyond Quick Fixes: Why True Accessibility Requires More Than Overlays
September 12, 2022
12 min
Amber Sawaya
Accessibility Overlays—like accessiBe and UserWay—are hot right now. While they can provide some increases in accessibility, we ultimately don’t recommend these tools. It’s best to do website accessibility right, starting with the brand colors and continuing through design and development.
We are big on accessibility here at Anchor & Alpine. We know that sites and products that are built with accessibility in mind not only serve a larger population but that the time and care spent on accessibility also garners better SEO and an overall increase in user experience.
We prefer to make websites accessible from the ground up; this starts in design with colors that have a good contrast, and type sizes that are large enough to read comfortably. We make sure that our HTML is semantic and that our tab order is correct. This helps us take advantage of the accessibility features already available in modern browsers, like the ability to use the keyboard to move through a site without a mouse, and to resize text with a key command. Still, as ADA lawsuits increase, our clients are asking about accessibility overlays—products like accessiBe and UserWay.
Accessibility overlays are third-party products installed on a website to detect and repair accessibility problems. Overlays promise that they only require a developer to add one line of code, while leaving the rest of the underlying source code untouched. Let’s look at the accessibility overlays on the market, what they do, what they cost, and why we don’t recommend them as a singular attempt to comply with accessibility standards.
Get to Know the Most Popular Accessibility Overlays
accessiBe and UserWay are two of the top accessibility overlays, and are collectively used by over a million websites. Their automated solutions make them a faster and more affordable alternative to 100% manual solutions (in the short term); however, their true effectiveness is heavily debated.
With AI technology, automated technology, and machine learning, both overlays can:
Stop animations for people with epilepsy.
Change color and contrast settings for the vision-impaired.
Activate reading mode for people with cognitive disorders.
Create quick navigation options for motor-impaired people.
Give some control over text size, text spacing, and highlight links.
Add accessible fonts, like dyslexia-friendly fonts.
Scan and evaluate all images with OCR (optical character recognition).
Use contextual understanding and AI technology to scan a site and analyze the elements from the collected data.
accessiBe and UserWay have similar features and pricing—the differences come in the speed of assessment and the manual options available.
accessiBe
UserWay
accessiBe audits 100% automatically using AI and API to scan a site.
UserWay uses automatic and manual auditing methods.
accessiBe works with Fortune 500 companies and small businesses including: Energizer and Fiverr among over 100,000 others
UserWay works with Fortune 500 companies and small businesses including: eBay, Fedex, and Coca Cola among over 1 million others
accessiBe claims to be GDPR and CCPA compliant and says they do not collect any personal data.
UserWay claims to be CCPA, FERPA, HIPAA, and GDPR compliant.
accessiBe offers a small litigation support package at no additional charge which gives minor support to their clients whose website’s accessibility is challenged.
UserWay offers an AI powered content monitor which scans the text on a site in order to identify potentially offensive language or discriminatory, biased, or racist language.
Pricing is based on total number of pages on a website and ranges from $49–249 per month. Additional solutions (all based) are available.
Pricing is based on page views and ranges from $49-229 per month. Additional solutions (fee based) are available.
Comparison of accessiBe and UserWay.
Additional solutions and fees are available in the two products:
Making a standard pdf form accessible – $12 per standard page, $40 per complex pdf page, and $120 per pdf page with interactive forms.
Real-time scanning service – $49/month for 10 pages, $299/month for 100 pages, and $999/month for 1000 pages.
Manually auditing a site for WCAG compliance – $2,500 for 10 pages, $7,500 for 30 pages, and $12,500 for 100 pages with the option to add a VPAT report for an additional $1,450.
Media remediation – $5-$15 per video or audio minute based on order quantity.
User testing with disabled users – $200 per tester hour.
An expert audit of your site – $450 per template, an optional set-up of the widget claiming increased functionality after that – $250 per website template.
A Voluntary Product Accessibility (VPAT) evaluation – $450 per product template.
The main difference between the tools is that accessiBe is fully automated, and UserWay offers manual scans. Besides that, both tools are similar in price and features. Both tools claim ADA and WCAG compliance which is strongly contested, and various lawsuits have been brought against sites utilizing these tools.
The Problems with Overlays
According to many accessibility experts, there is no fast and easy way to achieve full, long-term accessibility and WCAG compliance on a website. While these tools can be an interim solution, they are not considered sufficient or technologically capable of achieving WCAG compliance.
Legal Issues
A site with overlays is not immune to lawsuits, and it is deceptive marketing to claim otherwise. Although accessiBe and UserWay guarantee WCAG and ADA compliance, it has been determined that overlays can’t meet all the requirements of the WCAG Success Criteria.
In 2020, there were over 250 lawsuits against websites with overlays and 203 lawsuits against inaccessible mobile apps, according to UsableNet’s Accessibility lawsuit report. This report stated: “[accessibility overlays] do nothing to make a website more accessible for the blind.” Richard Hunt, one of the first attorneys to specialize in digital ADA cases, wrote that the only way to make a site accessible is to thoroughly test the site with disabled users to ensure they can understand and perform all functions on the site.
“[accessibility overlays] do nothing to make a website more accessible for the blind.” Richard Hunt, one of the first attorneys to specialize in digital ADA cases.
Richard Hunt
He said overlays could not fix accessibility errors or help you avoid an accessibility lawsuit. Additionally, a site containing an overlay could be specifically targeted by lawsuits.
Legal Precedents Against Overlays:
In 2018, The U.S. Department of Transportation sued the Scandinavian Airline System (SAS) for $200,000 for violating federal law and failing to serve their air travelers with disabilities appropriately. SAS created a separate website for users with disabilities which violated WCAG and the ADA. The “separate but equal” approach is not acceptable by the ADA and is considered non-inclusive. Often, a separate site made for people with disabilities is difficult to locate, has higher maintenance costs, isn’t updated alongside the original site, and is not received well by people with disabilities.
In 2021, a lawsuit was filed against the reading-glass company Eyebobs, which had installed accessiBe as their sole method of achieving accessibility. Karl Groves, an expert witness for the case, analyzed 50 websites that used accessiBe and found thousands of problems with the pages. His report called out the complexity of WCAG standards and said that achieving accessibility is impossible with machine learning alone. More conclusions from Karl Groves’ report follow.
Accessibility Auditor Turned Expert Witness
Karl Groves is an accessibility auditor and software developer and was called on as an expert witness in the Eyebobs accessibility case about the accessiBe overlay. He ran an audit of 50 websites using accessiBe, and examined their use of overlays as well as the overall effect of accessiBe on the sites’ accessibility. He concluded that WCAG is too complex and subjective for a site to be tested or repaired using automatic methods.
Interestingly, Groves found that “accessiBe customer sites are no better or worse than the broader Web as a whole.” From his analysis of overlays, accessibility, and WCAG standards, he deduced that there are some factors of accessibility that overlays cannot fix with automatic methods. An example he cites is that it is impossible to be 100% certain that the text alternatives of an image provided by an overlay are accurate. Additionally, content provided in Javascript or PDF1 cannot be assessed using an overlay, and would require additional services.
Among the most common errors that Groves found in accessiBe’s sites were with the accuracy of alternate descriptions (also called alt tags) of images and non-text content. accessiBe had issues recognizing non-text content in the first place, didn’t pick up on SVG images and therefore didn’t provide an alt-text for the image. He found that another common error was accessiBe’s inability to correct forms, labels, buttons, and menus to provide a good name for them. Although accessiBe claims to be able to solve issues with these features, Graves found that claim to be incorrect. He also found a third category of common errors with headings on accessiBe’s pages: pages on accessiBe’s sites with illogical heading structure, missing headings, and a disorganized page structure.
Groves concluded that without being able to service Javascript, PDF, video, audio, non-text content, document structure, forms, keyboard accessibility and focus control, and content within iframes accurately or at all, accessiBe (and other tools) are categorically incapable of making a site accessible on their own.
Public Reception of Overlays
According to a WebAIM survey, “A strong majority (67%) of respondents rate web accessibility overlays 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.”
72% of respondents with disabilities rated accessibility overlays as not at all or not very effective. Only 2.4% of respondents rated accessibility overlays as very effective.
WebAIM survey
There have been numerous protests and movements by people with disabilities calling on companies to remove these overlays from their sites entirely. In 2021, over 400 blind people filed an open letter calling on all companies who use automated accessibility overlays to stop. Also, the hashtag #AcessiBe was overrun by people who disagree that these tools make the internet more accessible. Overall, overlays are angering users and drawing business away from the companies who use them.
The primary problems with accessibility overlays: They often interfere with the tools that disabled people already use, they only detect 30% of accessibility errors on a site, and they are not effective at making a site accessible.
One of the primary problems with accessibility overlays is that they often interfere with the tools that disabled people already have.
Many disabled users (especially visually impaired users) already have screen readers and tools installed on their machines that they are comfortable using. AI overlays often interrupt and conflict with the tools already in use, making many sites unnavigable for disabled users. Many users opt to use their preset style sheets and feel overlays are redundant with current technologies. For them, overlays make the web browsing experience overcomplicated and frustrating.
Overlays have problems communicating visual elements such as graphs, tables, lists, infographics, and charts. They will simply lump all of these elements into the term “image,” which isn’t helpful for a user who is utilizing a screen reader to understand the information on the page. Learn more at https://overlayfactsheet.com/.
Many times, the AI will inaccurately guess what is in the image, or will catch some secondary information that isn’t the main subject of the image. For example, when identifying a picture of a woman drinking Starbucks coffee, the AI will call the image “a woman with long hair,” but neglect to include that she is drinking a beverage in the image, which may be the main reason that it was displayed on the webpage. Even though overlays can automatically identify and tag elements on a page, there is no way to understand if they are tagged correctly or not, and many times, they are not.
Additionally, “modern, component-based user interfaces, such as those using ReactJS, Angular, or Vue may change the state of all or some of the underlying page independently of the overlay, rendering it unable to fix those JavaScript-driven changes to content.”2 Overlays also don’t work well on mobile devices. Because 55% of email is opened first on mobile devices, and because mobile devices are responsible for 40% of global Internet traffic, businesses must prioritize their company’s mobile experience. Lawsuits increasingly target mobile sites and apps for their ADA and WCAG violations. Besides the legal implications, failure to create an accessible and responsive mobile experience will inevitably drive away a lot of traffic from a site.
Creating truly accessible websites is not only ethical but also beneficial for business. Households with a disability spend 86% more on retail than households without people with disabilities do. 91% of UK shoppers with disabilities said they had bailed out of inaccessible experiences in favor of a competitor.
Security Issues
Overlays are often hosted on the vendor’s server, so your company will not have control over the speed or security of the overlays. If their server is hacked, your website is at risk too. Additionally, overlays can gather information about a User using a screen-reader, such as the nature of their disability, age, ethnicity, and gender without obtaining the user’s permission, and without an opt-out setting. This is a major privacy concern for users.
Financial Concerns
Accessible overlays are subscription-based, so a company must continue paying for the service to access the features. It can often be more expensive for your site to mask the issue with overlays instead of adjusting the original code. Even UserWay, which offers manual testing, does not fix the original code but only the overlay layer, which doesn’t get to the root of the problem. The additional features and testing can also run up the annual bill on these services. Although overlays may appear to be the less expensive option in the short term, they are more expensive over time, and riskier than manually fixing the code of a site to be more accessible.
Accessibility Solutions
Unfortunately, overlays have not proven effective at solving critical accessibility issues on a site and are not recommended to be used as a long-term solution.
Accessibility is inherently complicated and interpretive; it is unrealistic to depend solely on fast-natured, automated solutions for compliance. The most efficient way to create a truly accessible site is to hire a web firm familiar with accessibility, and manually fix underlying accessibility issues instead of covering them up. If making your site accessible is time-sensitive, perhaps because a lawsuit was already filed against you, accessibility overlays can be a short-term solution while you are fixing the underlying code.
Accessibility Overlay Statistics:
According to a WebAIM survey: “A strong majority (67%) of respondents rate accessibility overlays 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.”
Households with a person with a disability spend 86% more on retail than households without people with disabilities.
91% of UK shoppers with disabilities said they had bailed out of inaccessible experiences for a competitor.
52% of customers are less likely to engage with a company because of a bad mobile experience. That’s ALL customers, not just customers with disabilities.
72.6% of internet users worldwide will only use their smartphones to get online by 2025.
61% of people won’t return to an inaccessible mobile site.
Want to try these out for yourself? Here are a couple of websites we found in the wild:
Special thank you to Hannah Brewster, our UX Researcher, for her work on the article.
Accessibility Overlays—like accessiBe and UserWay—are hot right now. While they can provide some increases in accessibility, we ultimately don’t recommend these tools. It’s best to do website accessibility right, starting with the brand colors and continuing through design and development.
We are big on accessibility here at Anchor & Alpine. We know that sites and products that are built with accessibility in mind not only serve a larger population but that the time and care spent on accessibility also garners better SEO and an overall increase in user experience.
We prefer to make websites accessible from the ground up; this starts in design with colors that have a good contrast, and type sizes that are large enough to read comfortably. We make sure that our HTML is semantic and that our tab order is correct. This helps us take advantage of the accessibility features already available in modern browsers, like the ability to use the keyboard to move through a site without a mouse, and to resize text with a key command. Still, as ADA lawsuits increase, our clients are asking about accessibility overlays—products like accessiBe and UserWay.
Accessibility overlays are third-party products installed on a website to detect and repair accessibility problems. Overlays promise that they only require a developer to add one line of code, while leaving the rest of the underlying source code untouched. Let’s look at the accessibility overlays on the market, what they do, what they cost, and why we don’t recommend them as a singular attempt to comply with accessibility standards.
Get to Know the Most Popular Accessibility Overlays
accessiBe and UserWay are two of the top accessibility overlays, and are collectively used by over a million websites. Their automated solutions make them a faster and more affordable alternative to 100% manual solutions (in the short term); however, their true effectiveness is heavily debated.
With AI technology, automated technology, and machine learning, both overlays can:
Stop animations for people with epilepsy.
Change color and contrast settings for the vision-impaired.
Activate reading mode for people with cognitive disorders.
Create quick navigation options for motor-impaired people.
Give some control over text size, text spacing, and highlight links.
Add accessible fonts, like dyslexia-friendly fonts.
Scan and evaluate all images with OCR (optical character recognition).
Use contextual understanding and AI technology to scan a site and analyze the elements from the collected data.
accessiBe and UserWay have similar features and pricing—the differences come in the speed of assessment and the manual options available.
accessiBe
UserWay
accessiBe audits 100% automatically using AI and API to scan a site.
UserWay uses automatic and manual auditing methods.
accessiBe works with Fortune 500 companies and small businesses including: Energizer and Fiverr among over 100,000 others
UserWay works with Fortune 500 companies and small businesses including: eBay, Fedex, and Coca Cola among over 1 million others
accessiBe claims to be GDPR and CCPA compliant and says they do not collect any personal data.
UserWay claims to be CCPA, FERPA, HIPAA, and GDPR compliant.
accessiBe offers a small litigation support package at no additional charge which gives minor support to their clients whose website’s accessibility is challenged.
UserWay offers an AI powered content monitor which scans the text on a site in order to identify potentially offensive language or discriminatory, biased, or racist language.
Pricing is based on total number of pages on a website and ranges from $49–249 per month. Additional solutions (all based) are available.
Pricing is based on page views and ranges from $49-229 per month. Additional solutions (fee based) are available.
Comparison of accessiBe and UserWay.
Additional solutions and fees are available in the two products:
Making a standard pdf form accessible – $12 per standard page, $40 per complex pdf page, and $120 per pdf page with interactive forms.
Real-time scanning service – $49/month for 10 pages, $299/month for 100 pages, and $999/month for 1000 pages.
Manually auditing a site for WCAG compliance – $2,500 for 10 pages, $7,500 for 30 pages, and $12,500 for 100 pages with the option to add a VPAT report for an additional $1,450.
Media remediation – $5-$15 per video or audio minute based on order quantity.
User testing with disabled users – $200 per tester hour.
An expert audit of your site – $450 per template, an optional set-up of the widget claiming increased functionality after that – $250 per website template.
A Voluntary Product Accessibility (VPAT) evaluation – $450 per product template.
The main difference between the tools is that accessiBe is fully automated, and UserWay offers manual scans. Besides that, both tools are similar in price and features. Both tools claim ADA and WCAG compliance which is strongly contested, and various lawsuits have been brought against sites utilizing these tools.
The Problems with Overlays
According to many accessibility experts, there is no fast and easy way to achieve full, long-term accessibility and WCAG compliance on a website. While these tools can be an interim solution, they are not considered sufficient or technologically capable of achieving WCAG compliance.
Legal Issues
A site with overlays is not immune to lawsuits, and it is deceptive marketing to claim otherwise. Although accessiBe and UserWay guarantee WCAG and ADA compliance, it has been determined that overlays can’t meet all the requirements of the WCAG Success Criteria.
In 2020, there were over 250 lawsuits against websites with overlays and 203 lawsuits against inaccessible mobile apps, according to UsableNet’s Accessibility lawsuit report. This report stated: “[accessibility overlays] do nothing to make a website more accessible for the blind.” Richard Hunt, one of the first attorneys to specialize in digital ADA cases, wrote that the only way to make a site accessible is to thoroughly test the site with disabled users to ensure they can understand and perform all functions on the site.
“[accessibility overlays] do nothing to make a website more accessible for the blind.” Richard Hunt, one of the first attorneys to specialize in digital ADA cases.
Richard Hunt
He said overlays could not fix accessibility errors or help you avoid an accessibility lawsuit. Additionally, a site containing an overlay could be specifically targeted by lawsuits.
Legal Precedents Against Overlays:
In 2018, The U.S. Department of Transportation sued the Scandinavian Airline System (SAS) for $200,000 for violating federal law and failing to serve their air travelers with disabilities appropriately. SAS created a separate website for users with disabilities which violated WCAG and the ADA. The “separate but equal” approach is not acceptable by the ADA and is considered non-inclusive. Often, a separate site made for people with disabilities is difficult to locate, has higher maintenance costs, isn’t updated alongside the original site, and is not received well by people with disabilities.
In 2021, a lawsuit was filed against the reading-glass company Eyebobs, which had installed accessiBe as their sole method of achieving accessibility. Karl Groves, an expert witness for the case, analyzed 50 websites that used accessiBe and found thousands of problems with the pages. His report called out the complexity of WCAG standards and said that achieving accessibility is impossible with machine learning alone. More conclusions from Karl Groves’ report follow.
Accessibility Auditor Turned Expert Witness
Karl Groves is an accessibility auditor and software developer and was called on as an expert witness in the Eyebobs accessibility case about the accessiBe overlay. He ran an audit of 50 websites using accessiBe, and examined their use of overlays as well as the overall effect of accessiBe on the sites’ accessibility. He concluded that WCAG is too complex and subjective for a site to be tested or repaired using automatic methods.
Interestingly, Groves found that “accessiBe customer sites are no better or worse than the broader Web as a whole.” From his analysis of overlays, accessibility, and WCAG standards, he deduced that there are some factors of accessibility that overlays cannot fix with automatic methods. An example he cites is that it is impossible to be 100% certain that the text alternatives of an image provided by an overlay are accurate. Additionally, content provided in Javascript or PDF1 cannot be assessed using an overlay, and would require additional services.
Among the most common errors that Groves found in accessiBe’s sites were with the accuracy of alternate descriptions (also called alt tags) of images and non-text content. accessiBe had issues recognizing non-text content in the first place, didn’t pick up on SVG images and therefore didn’t provide an alt-text for the image. He found that another common error was accessiBe’s inability to correct forms, labels, buttons, and menus to provide a good name for them. Although accessiBe claims to be able to solve issues with these features, Graves found that claim to be incorrect. He also found a third category of common errors with headings on accessiBe’s pages: pages on accessiBe’s sites with illogical heading structure, missing headings, and a disorganized page structure.
Groves concluded that without being able to service Javascript, PDF, video, audio, non-text content, document structure, forms, keyboard accessibility and focus control, and content within iframes accurately or at all, accessiBe (and other tools) are categorically incapable of making a site accessible on their own.
Public Reception of Overlays
According to a WebAIM survey, “A strong majority (67%) of respondents rate web accessibility overlays 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.”
72% of respondents with disabilities rated accessibility overlays as not at all or not very effective. Only 2.4% of respondents rated accessibility overlays as very effective.
WebAIM survey
There have been numerous protests and movements by people with disabilities calling on companies to remove these overlays from their sites entirely. In 2021, over 400 blind people filed an open letter calling on all companies who use automated accessibility overlays to stop. Also, the hashtag #AcessiBe was overrun by people who disagree that these tools make the internet more accessible. Overall, overlays are angering users and drawing business away from the companies who use them.
The primary problems with accessibility overlays: They often interfere with the tools that disabled people already use, they only detect 30% of accessibility errors on a site, and they are not effective at making a site accessible.
One of the primary problems with accessibility overlays is that they often interfere with the tools that disabled people already have.
Many disabled users (especially visually impaired users) already have screen readers and tools installed on their machines that they are comfortable using. AI overlays often interrupt and conflict with the tools already in use, making many sites unnavigable for disabled users. Many users opt to use their preset style sheets and feel overlays are redundant with current technologies. For them, overlays make the web browsing experience overcomplicated and frustrating.
Overlays have problems communicating visual elements such as graphs, tables, lists, infographics, and charts. They will simply lump all of these elements into the term “image,” which isn’t helpful for a user who is utilizing a screen reader to understand the information on the page. Learn more at https://overlayfactsheet.com/.
Many times, the AI will inaccurately guess what is in the image, or will catch some secondary information that isn’t the main subject of the image. For example, when identifying a picture of a woman drinking Starbucks coffee, the AI will call the image “a woman with long hair,” but neglect to include that she is drinking a beverage in the image, which may be the main reason that it was displayed on the webpage. Even though overlays can automatically identify and tag elements on a page, there is no way to understand if they are tagged correctly or not, and many times, they are not.
Additionally, “modern, component-based user interfaces, such as those using ReactJS, Angular, or Vue may change the state of all or some of the underlying page independently of the overlay, rendering it unable to fix those JavaScript-driven changes to content.”2 Overlays also don’t work well on mobile devices. Because 55% of email is opened first on mobile devices, and because mobile devices are responsible for 40% of global Internet traffic, businesses must prioritize their company’s mobile experience. Lawsuits increasingly target mobile sites and apps for their ADA and WCAG violations. Besides the legal implications, failure to create an accessible and responsive mobile experience will inevitably drive away a lot of traffic from a site.
Creating truly accessible websites is not only ethical but also beneficial for business. Households with a disability spend 86% more on retail than households without people with disabilities do. 91% of UK shoppers with disabilities said they had bailed out of inaccessible experiences in favor of a competitor.
Security Issues
Overlays are often hosted on the vendor’s server, so your company will not have control over the speed or security of the overlays. If their server is hacked, your website is at risk too. Additionally, overlays can gather information about a User using a screen-reader, such as the nature of their disability, age, ethnicity, and gender without obtaining the user’s permission, and without an opt-out setting. This is a major privacy concern for users.
Financial Concerns
Accessible overlays are subscription-based, so a company must continue paying for the service to access the features. It can often be more expensive for your site to mask the issue with overlays instead of adjusting the original code. Even UserWay, which offers manual testing, does not fix the original code but only the overlay layer, which doesn’t get to the root of the problem. The additional features and testing can also run up the annual bill on these services. Although overlays may appear to be the less expensive option in the short term, they are more expensive over time, and riskier than manually fixing the code of a site to be more accessible.
Accessibility Solutions
Unfortunately, overlays have not proven effective at solving critical accessibility issues on a site and are not recommended to be used as a long-term solution.
Accessibility is inherently complicated and interpretive; it is unrealistic to depend solely on fast-natured, automated solutions for compliance. The most efficient way to create a truly accessible site is to hire a web firm familiar with accessibility, and manually fix underlying accessibility issues instead of covering them up. If making your site accessible is time-sensitive, perhaps because a lawsuit was already filed against you, accessibility overlays can be a short-term solution while you are fixing the underlying code.
Accessibility Overlay Statistics:
According to a WebAIM survey: “A strong majority (67%) of respondents rate accessibility overlays 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.”
Households with a person with a disability spend 86% more on retail than households without people with disabilities.
91% of UK shoppers with disabilities said they had bailed out of inaccessible experiences for a competitor.
52% of customers are less likely to engage with a company because of a bad mobile experience. That’s ALL customers, not just customers with disabilities.
72.6% of internet users worldwide will only use their smartphones to get online by 2025.
61% of people won’t return to an inaccessible mobile site.
Want to try these out for yourself? Here are a couple of websites we found in the wild:
Karl Groves calls out Flash, Javascript, Silverlight, and PDFs as having inaccessible content. We removed Flash and Silverlight from the article body because those technologies are out of date and no longer used. [↩]
Accessibility Overlays—like accessiBe and UserWay—are hot right now. While they can provide some increases in accessibility, we ultimately don’t recommend these tools. It’s best to do website accessibility right, starting with the brand colors and continuing through design and development.
We are big on accessibility here at Anchor & Alpine. We know that sites and products that are built with accessibility in mind not only serve a larger population but that the time and care spent on accessibility also garners better SEO and an overall increase in user experience.
We prefer to make websites accessible from the ground up; this starts in design with colors that have a good contrast, and type sizes that are large enough to read comfortably. We make sure that our HTML is semantic and that our tab order is correct. This helps us take advantage of the accessibility features already available in modern browsers, like the ability to use the keyboard to move through a site without a mouse, and to resize text with a key command. Still, as ADA lawsuits increase, our clients are asking about accessibility overlays—products like accessiBe and UserWay.
Accessibility overlays are third-party products installed on a website to detect and repair accessibility problems. Overlays promise that they only require a developer to add one line of code, while leaving the rest of the underlying source code untouched. Let’s look at the accessibility overlays on the market, what they do, what they cost, and why we don’t recommend them as a singular attempt to comply with accessibility standards.
Get to Know the Most Popular Accessibility Overlays
accessiBe and UserWay are two of the top accessibility overlays, and are collectively used by over a million websites. Their automated solutions make them a faster and more affordable alternative to 100% manual solutions (in the short term); however, their true effectiveness is heavily debated.
With AI technology, automated technology, and machine learning, both overlays can:
Stop animations for people with epilepsy.
Change color and contrast settings for the vision-impaired.
Activate reading mode for people with cognitive disorders.
Create quick navigation options for motor-impaired people.
Give some control over text size, text spacing, and highlight links.
Add accessible fonts, like dyslexia-friendly fonts.
Scan and evaluate all images with OCR (optical character recognition).
Use contextual understanding and AI technology to scan a site and analyze the elements from the collected data.
accessiBe and UserWay have similar features and pricing—the differences come in the speed of assessment and the manual options available.
accessiBe
UserWay
accessiBe audits 100% automatically using AI and API to scan a site.
UserWay uses automatic and manual auditing methods.
accessiBe works with Fortune 500 companies and small businesses including: Energizer and Fiverr among over 100,000 others
UserWay works with Fortune 500 companies and small businesses including: eBay, Fedex, and Coca Cola among over 1 million others
accessiBe claims to be GDPR and CCPA compliant and says they do not collect any personal data.
UserWay claims to be CCPA, FERPA, HIPAA, and GDPR compliant.
accessiBe offers a small litigation support package at no additional charge which gives minor support to their clients whose website’s accessibility is challenged.
UserWay offers an AI powered content monitor which scans the text on a site in order to identify potentially offensive language or discriminatory, biased, or racist language.
Pricing is based on total number of pages on a website and ranges from $49–249 per month. Additional solutions (all based) are available.
Pricing is based on page views and ranges from $49-229 per month. Additional solutions (fee based) are available.
Comparison of accessiBe and UserWay.
Additional solutions and fees are available in the two products:
Making a standard pdf form accessible – $12 per standard page, $40 per complex pdf page, and $120 per pdf page with interactive forms.
Real-time scanning service – $49/month for 10 pages, $299/month for 100 pages, and $999/month for 1000 pages.
Manually auditing a site for WCAG compliance – $2,500 for 10 pages, $7,500 for 30 pages, and $12,500 for 100 pages with the option to add a VPAT report for an additional $1,450.
Media remediation – $5-$15 per video or audio minute based on order quantity.
User testing with disabled users – $200 per tester hour.
An expert audit of your site – $450 per template, an optional set-up of the widget claiming increased functionality after that – $250 per website template.
A Voluntary Product Accessibility (VPAT) evaluation – $450 per product template.
The main difference between the tools is that accessiBe is fully automated, and UserWay offers manual scans. Besides that, both tools are similar in price and features. Both tools claim ADA and WCAG compliance which is strongly contested, and various lawsuits have been brought against sites utilizing these tools.
The Problems with Overlays
According to many accessibility experts, there is no fast and easy way to achieve full, long-term accessibility and WCAG compliance on a website. While these tools can be an interim solution, they are not considered sufficient or technologically capable of achieving WCAG compliance.
Legal Issues
A site with overlays is not immune to lawsuits, and it is deceptive marketing to claim otherwise. Although accessiBe and UserWay guarantee WCAG and ADA compliance, it has been determined that overlays can’t meet all the requirements of the WCAG Success Criteria.
In 2020, there were over 250 lawsuits against websites with overlays and 203 lawsuits against inaccessible mobile apps, according to UsableNet’s Accessibility lawsuit report. This report stated: “[accessibility overlays] do nothing to make a website more accessible for the blind.” Richard Hunt, one of the first attorneys to specialize in digital ADA cases, wrote that the only way to make a site accessible is to thoroughly test the site with disabled users to ensure they can understand and perform all functions on the site.
“[accessibility overlays] do nothing to make a website more accessible for the blind.” Richard Hunt, one of the first attorneys to specialize in digital ADA cases.
Richard Hunt
He said overlays could not fix accessibility errors or help you avoid an accessibility lawsuit. Additionally, a site containing an overlay could be specifically targeted by lawsuits.
Legal Precedents Against Overlays:
In 2018, The U.S. Department of Transportation sued the Scandinavian Airline System (SAS) for $200,000 for violating federal law and failing to serve their air travelers with disabilities appropriately. SAS created a separate website for users with disabilities which violated WCAG and the ADA. The “separate but equal” approach is not acceptable by the ADA and is considered non-inclusive. Often, a separate site made for people with disabilities is difficult to locate, has higher maintenance costs, isn’t updated alongside the original site, and is not received well by people with disabilities.
In 2021, a lawsuit was filed against the reading-glass company Eyebobs, which had installed accessiBe as their sole method of achieving accessibility. Karl Groves, an expert witness for the case, analyzed 50 websites that used accessiBe and found thousands of problems with the pages. His report called out the complexity of WCAG standards and said that achieving accessibility is impossible with machine learning alone. More conclusions from Karl Groves’ report follow.
Accessibility Auditor Turned Expert Witness
Karl Groves is an accessibility auditor and software developer and was called on as an expert witness in the Eyebobs accessibility case about the accessiBe overlay. He ran an audit of 50 websites using accessiBe, and examined their use of overlays as well as the overall effect of accessiBe on the sites’ accessibility. He concluded that WCAG is too complex and subjective for a site to be tested or repaired using automatic methods.
Interestingly, Groves found that “accessiBe customer sites are no better or worse than the broader Web as a whole.” From his analysis of overlays, accessibility, and WCAG standards, he deduced that there are some factors of accessibility that overlays cannot fix with automatic methods. An example he cites is that it is impossible to be 100% certain that the text alternatives of an image provided by an overlay are accurate. Additionally, content provided in Javascript or PDF1 cannot be assessed using an overlay, and would require additional services.
Among the most common errors that Groves found in accessiBe’s sites were with the accuracy of alternate descriptions (also called alt tags) of images and non-text content. accessiBe had issues recognizing non-text content in the first place, didn’t pick up on SVG images and therefore didn’t provide an alt-text for the image. He found that another common error was accessiBe’s inability to correct forms, labels, buttons, and menus to provide a good name for them. Although accessiBe claims to be able to solve issues with these features, Graves found that claim to be incorrect. He also found a third category of common errors with headings on accessiBe’s pages: pages on accessiBe’s sites with illogical heading structure, missing headings, and a disorganized page structure.
Groves concluded that without being able to service Javascript, PDF, video, audio, non-text content, document structure, forms, keyboard accessibility and focus control, and content within iframes accurately or at all, accessiBe (and other tools) are categorically incapable of making a site accessible on their own.
Public Reception of Overlays
According to a WebAIM survey, “A strong majority (67%) of respondents rate web accessibility overlays 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.”
72% of respondents with disabilities rated accessibility overlays as not at all or not very effective. Only 2.4% of respondents rated accessibility overlays as very effective.
WebAIM survey
There have been numerous protests and movements by people with disabilities calling on companies to remove these overlays from their sites entirely. In 2021, over 400 blind people filed an open letter calling on all companies who use automated accessibility overlays to stop. Also, the hashtag #AcessiBe was overrun by people who disagree that these tools make the internet more accessible. Overall, overlays are angering users and drawing business away from the companies who use them.
The primary problems with accessibility overlays: They often interfere with the tools that disabled people already use, they only detect 30% of accessibility errors on a site, and they are not effective at making a site accessible.
One of the primary problems with accessibility overlays is that they often interfere with the tools that disabled people already have.
Many disabled users (especially visually impaired users) already have screen readers and tools installed on their machines that they are comfortable using. AI overlays often interrupt and conflict with the tools already in use, making many sites unnavigable for disabled users. Many users opt to use their preset style sheets and feel overlays are redundant with current technologies. For them, overlays make the web browsing experience overcomplicated and frustrating.
Overlays have problems communicating visual elements such as graphs, tables, lists, infographics, and charts. They will simply lump all of these elements into the term “image,” which isn’t helpful for a user who is utilizing a screen reader to understand the information on the page. Learn more at https://overlayfactsheet.com/.
Many times, the AI will inaccurately guess what is in the image, or will catch some secondary information that isn’t the main subject of the image. For example, when identifying a picture of a woman drinking Starbucks coffee, the AI will call the image “a woman with long hair,” but neglect to include that she is drinking a beverage in the image, which may be the main reason that it was displayed on the webpage. Even though overlays can automatically identify and tag elements on a page, there is no way to understand if they are tagged correctly or not, and many times, they are not.
Additionally, “modern, component-based user interfaces, such as those using ReactJS, Angular, or Vue may change the state of all or some of the underlying page independently of the overlay, rendering it unable to fix those JavaScript-driven changes to content.”2 Overlays also don’t work well on mobile devices. Because 55% of email is opened first on mobile devices, and because mobile devices are responsible for 40% of global Internet traffic, businesses must prioritize their company’s mobile experience. Lawsuits increasingly target mobile sites and apps for their ADA and WCAG violations. Besides the legal implications, failure to create an accessible and responsive mobile experience will inevitably drive away a lot of traffic from a site.
Creating truly accessible websites is not only ethical but also beneficial for business. Households with a disability spend 86% more on retail than households without people with disabilities do. 91% of UK shoppers with disabilities said they had bailed out of inaccessible experiences in favor of a competitor.
Security Issues
Overlays are often hosted on the vendor’s server, so your company will not have control over the speed or security of the overlays. If their server is hacked, your website is at risk too. Additionally, overlays can gather information about a User using a screen-reader, such as the nature of their disability, age, ethnicity, and gender without obtaining the user’s permission, and without an opt-out setting. This is a major privacy concern for users.
Financial Concerns
Accessible overlays are subscription-based, so a company must continue paying for the service to access the features. It can often be more expensive for your site to mask the issue with overlays instead of adjusting the original code. Even UserWay, which offers manual testing, does not fix the original code but only the overlay layer, which doesn’t get to the root of the problem. The additional features and testing can also run up the annual bill on these services. Although overlays may appear to be the less expensive option in the short term, they are more expensive over time, and riskier than manually fixing the code of a site to be more accessible.
Accessibility Solutions
Unfortunately, overlays have not proven effective at solving critical accessibility issues on a site and are not recommended to be used as a long-term solution.
Accessibility is inherently complicated and interpretive; it is unrealistic to depend solely on fast-natured, automated solutions for compliance. The most efficient way to create a truly accessible site is to hire a web firm familiar with accessibility, and manually fix underlying accessibility issues instead of covering them up. If making your site accessible is time-sensitive, perhaps because a lawsuit was already filed against you, accessibility overlays can be a short-term solution while you are fixing the underlying code.
Accessibility Overlay Statistics:
According to a WebAIM survey: “A strong majority (67%) of respondents rate accessibility overlays 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.”
Households with a person with a disability spend 86% more on retail than households without people with disabilities.
91% of UK shoppers with disabilities said they had bailed out of inaccessible experiences for a competitor.
52% of customers are less likely to engage with a company because of a bad mobile experience. That’s ALL customers, not just customers with disabilities.
72.6% of internet users worldwide will only use their smartphones to get online by 2025.
61% of people won’t return to an inaccessible mobile site.
Want to try these out for yourself? Here are a couple of websites we found in the wild:
Karl Groves calls out Flash, Javascript, Silverlight, and PDFs as having inaccessible content. We removed Flash and Silverlight from the article body because those technologies are out of date and no longer used. [↩]
Karl Groves calls out Flash, Javascript, Silverlight, and PDFs as having inaccessible content. We removed Flash and Silverlight from the article body because those technologies are out of date and no longer used. [↩]
