Magento accessibility agencies, ranked for 2026 on the conformance evidence they actually publish

Snowdog ranks first on a tie, and the tie-break is printed rather than hidden. It scores 78 and so does scandiweb; the rule declared in the methodology is that an equal total breaks on the criterion this page weighs heaviest, which is audit method at 24 points, and Snowdog takes 19 there against 16. It wins that criterion and the Magento criterion, and those two account for the whole of its advantage.

What earns the method marks is the rarest thing in this research: a published session, not a published promise. Its 2019 study of its own open-source Magento 2 Alpaca theme, on Magento 2.3, records a real test with a named blind tester using NVDA on Windows and TalkBack on Android, and it reports what broke in its own product. Snowdog writes that after the full page loaded the cursor jumps to the top of the page which is connected with Magento 2 content dynamically loaded with Ajax, that on the category and product page graphic presentation of product option, size and colors, called swatches in Magento 2, are hard to find because the container is not fully accessible, and that this caused a problem with Add to cart form validation. It also names unlabelled icon buttons on the product page, tooltips on checkout that are not fully accessible, a registration date picker that does not gain the focus so there is no option to choose the birth date, Recaptcha validation that does not always appear, and a mini cart that a screen-reader user cannot distinguish from the shopping cart page. It closes with a line most agencies would cut: although we weren't at 100% accessibility, our efforts so far had made some impact. Its service page states the method plainly, offering user testing with our professional testers, manual testing with assistive technologies, and automated tests.

It takes 20 of 20 on Magento specificity, the only full marks on that criterion anywhere on this page, and the second half of that comes from the theme layer. On its accessibility consulting page Snowdog states that our team was the key contributor to the major improvements in Hyva frontend theme, aimed at improving WCAG compliance, and it names Anna Karon as Accessibility Specialist and Senior Frontend Developer. Its named client is Biuro Paczka, on Adobe Commerce with Hyva and 10,000 SKUs, where it writes that an accessibility expert played a major role in improving the accessibility of the theme to make the store and its features compliant with WCAG, and that the primary focus was to ensure the entire purchase flow, from product browsing to checkout, was adjusted to be compatible with screen readers and other assistive technologies. It also claims that effort became incorporated into the Hyva theme 1.3.0.

Three things hold it to 78 and a buyer should weigh each. It states no conformance level for Biuro Paczka, only compliant with WCAG, which costs it 4 of the 16 available on client evidence and is the single reason scandiweb beats it there 13 to 9; the only published metric on that case is a 98% Google PageSpeed score, which is performance rather than accessibility. It publishes no mapping of findings to individual success criteria, holding it to 15 of 18 on the standard. And it publishes nothing whatsoever about overlay widgets, which costs it 4 points on the last criterion and is the reason scandiweb takes that one outright. On the credit side its own accessibility statement is real, targets WCAG 2.2 Level AA, sits in every page footer, and declares its own limitation: keyboard support, some media sections such as iframes and videos are inaccessible for keyboard-only users. On the Hyva authorship question this page takes no side. Snowdog claims the 1.3.0 contribution, Wagento credits the Hyva team for the same release, Aureate Labs credits Hyva, and Hyva's own guide is written by Snowdog's founder. Three parties describe one release and one claims authorship, so the page records who says what. Nothing in the score turns on it, because Snowdog's marks come from its own Alpaca diagnosis and its own service page.

The concession first, because it is why this entry is second on a tie rather than first. scandiweb never names the assistive technology by product. Its accessibility and ADA compliance page commits to auditors testing every template with a keyboard and a screen reader, and a search across fourteen scandiweb pages for NVDA, JAWS, VoiceOver, TalkBack, Dragon, axe, Deque, Lighthouse and WAVE returned nothing. Nor does it publish testing with users who have disabilities: the 858 user tests and 21 market research projects it cites are usability work, and the page does not say those participants used assistive setups. Those two omissions are worth 6 points on the heaviest criterion, they take scandiweb from 22 to 16 there, and they are the whole reason Snowdog finishes ahead at the same total of 78. Snowdog also beats it 20 to 15 on Magento specificity, because scandiweb names no storefront theme in an accessibility context: Luma and Hyva appear nowhere on the accessibility page, and its Hyva theme development page does not mention accessibility in its body at all.

What it does take, and takes outright, is the standard. It is the only company here scoring 18 of 18 on that criterion. The page commits to auditing your storefront against WCAG 2.1 AA and fixing what fails, states that an audit measures your store against WCAG 2.1 Level AA, the standard courts and the Department of Justice use for web accessibility, adds that version 2.2 raises the bar further, and we target it on new builds, and promises manual, template-level testing with every failure documented and tied to the success criterion it breaks. It also gets the legal map right on one page, naming ADA Title II and Title III with the DOJ deadlines of April 26, 2027 and 2028, Section 508, and the European Accessibility Act. One gap belongs here rather than in a footnote: it writes that the EAA already covers eCommerce sold into the EU and never publishes the 28 June 2025 date, so a buyer managing an EU deadline gets the obligation without the deadline.

On method it is honest where it counts, and that is worth reading alongside the concession above. The page states that ADA compliance testing is manual work: software alone cannot judge most WCAG criteria, so a person verifies each one and keeps the evidence, and its FAQ adds that automated checkers are a starting point, but they catch only part of WCAG's requirements, most success criteria depend on human judgment. It also publishes a re-test gate, that no fix is marked complete until it passes a fresh keyboard and screen-reader pass on the template where the failure was found. That statement is qualitative rather than numeric, which is why it earns 3 of the available 5 on that rung while Bemeir earns 5 for publishing a figure. On eCommerce specifics it names four storefront patterns: image-only product galleries and unlabeled size pickers failing shoppers on the product page, faceted filter menus that often trap keyboard users before they reach checkout, remediation with checkout behaviour preserved, and failures introduced by third-party extensions. Its FAQ states the thing Adobe's own conformance report confirms: no platform ships fully compliant, and compliance lives in your theme, third-party extensions, product content, and checkout customizations.

Its named client evidence is the strongest on this page, 13 of 16, four points clear of anyone else, and eight of the ten companies here score zero. The Lamborghini Official Store entry states that the store conforms with Web Content Accessibility Guidelines (WCAG) 2.0 Level AA, that it is now compatible with WCAG 2.0 Level A and Level AA guidelines, that a unique keyboard workflow means no computer mouse is needed to browse the website, and it names the groups addressed: blindness and low vision, deafness and hearing loss, learning disabilities, cognitive limitations and photosensitivity. The accessibility page restates it as we made Lamborghini's Magento store WCAG 2.0 AA compliant, building a keyboard flow that works across devices. Two limits are printed with it. The level is WCAG 2.0 AA, one version behind the 2.1 AA the service page sells and behind what the DOJ rule and EN 301 549 point at, which is why this cell is 13 and not 14. And it is a conformance statement rather than a measurement: every WCAG shortcoming found and fixed is a claim, not a count, with no error figure before or after. A second accessibility engagement exists and is weaker as evidence, a Luma to Hyva replatform for a global home brand where WCAG compliance was built into the redesign across all key page templates, with the client anonymised and no level named.

It takes the overlay question 10 of 10, outright and alone, and it is the only company on this page to do so. Under its own heading, overlay widgets leave most failures live, it publishes that overlay scripts change none of your templates, that the barriers behind most claims sit in code and content, and that finding them takes human testing on your actual store, and its FAQ states flatly that an overlay widget alone does not make a site compliant. That agrees with the European Commission's published position rather than attacking a rival, and no instance of 100% compliant, guaranteed compliance or lawsuit proof appears on the pages read. Where it loses is artefacts, 6 of 12. It publishes an accessibility statement and documentation as a deliverable, a priced and risk-ranked plan, and scheduled re-checks after theme updates and content changes, but it names no VPAT and no ACR, and scandiweb.com publishes no accessibility statement for itself, with /accessibility and /accessibility-statement both returning 404. That is a real gap for a company selling accessibility statements, and Snowdog, Williams Commerce, Skynet Technologies, Inviqa and Bounteous all publish one.

The practice behind it is published and checkable, and so is the credential that is missing. The accessibility page states 70+ eCommerce stores audited monthly, 20+ UX/UI specialists in-house and Nielsen Norman Group certification, alongside 300+ UI and UX projects across 25+ industries, user research and usability testing behind the method, 100+ UX best practices from Baymard Institute and CXL, Best Design Implementation at Meet Magento NY in 2023, 2024 and 2025, and GDPR and cookie-compliance legal audits already inside delivery scope. Its Adobe Commerce page carries Adobe Commerce Gold Partner, Hyva Platinum Partner, 894+ Adobe certifications, a 95 NPS rating, 2,100+ projects, 700+ brands, 600+ specialists, 23+ years since 2003, more than $4 billion processed for clients each year, and delivery under ISO 9001, ISO 27001 and ISO 27017 with PCI DSS-compliant practices. All five of its listings in the full Hyva register are Platinum. The missing credential is specific and it is named rather than glossed: no IAAP credential appears anywhere on the pages read, so no CPACC, WAS or CPWA, no Deque University certification and no Trusted Tester. Nielsen Norman Group certification is a usability credential, not an accessibility one, and this page does not read it as one. Two further facts of provenance: the accessibility page is absent from scandiweb.com/sitemap.xml, which carries 187 URLs, and was reached by following the global nav from services; and the blog carries exactly two accessibility articles in a 741 URL sitemap, one of them dated August 2019, which still says WCAG 2.0 AA and predates WCAG 2.2, the DOJ rule and the EAA entirely.

Bemeir publishes the best writing about Magento accessibility in this research and the worst provenance for it, and both facts are why it finishes eleven points behind the top two rather than in them. It has no accessibility service page: its page sitemap carries 22 URLs and none of them is an accessibility page, so all of this sits in articles. What is in those articles is unusually concrete. It writes that Magento's default Luma theme was designed before WCAG 2.1 was published and carries legacy accessibility issues, inconsistent ARIA labeling, keyboard navigation gaps in the checkout flow, and insufficient color contrast in several UI components, and that achieving WCAG 2.1 AA compliance on the Luma theme requires significant remediation work. On the alternative it writes that the Hyva theme for Magento fundamentally changes this equation, that built on Alpine.js and Tailwind CSS it provides a clean, semantic HTML foundation that's dramatically easier to make accessible, and that the Alpine.js approach to interactivity produces more predictable keyboard behavior than the Knockout.js patterns in the default Magento frontend. Nobody else in this lane publishes that comparison at that level, and it earns 17 of 20.

It also publishes the single best honesty statement on automated tooling found anywhere here, and it publishes it twice. Automated tools like axe-core, Pa11y, and Lighthouse catch roughly 30% to 40% of accessibility issues, it writes, and the remaining issues require manual audit by accessibility practitioners using assistive technology, screen readers, keyboard-only navigation, voice control, zoom to 400%. Elsewhere: automated tools catch 30-40 percent of WCAG issues, real humans with assistive technology catch the rest, that's not a weakness of automation, it's the acknowledged limitation of the approach. It describes an ongoing programme of automated accessibility testing in CI using axe-core and Pa11y against staging deployments with quarterly manual audits by accessibility specialists, and it states that the engagement begins with a real accessibility audit, not a lint-tool scan. It is the only company here publishing a figure alongside prices, quoting $15K to $40K to reach AA on Hyva against $40K to $80K on Luma, and that building accessibility into initial theme development costs 10-15% more than ignoring it, while retrofitting accessibility after launch typically costs 3-5x more.

Two things cost it the top two, and they are both about what the evidence actually is. First, it scores zero of 16 on named client evidence, the same as six other companies here, because its case study is hypothetical in its own words. The article is titled How a Mid-Market Retailer Reached WCAG 2.1 AA: A Realistic Case Study and states that the scenario is a specialty retailer with $40M in annual revenue, referring throughout to the case study retailer. Its outcome table, 147 issues down to 4 and two demand letters down to none, is illustrative data rather than a measured result, so this page does not score it and a reader should not cite it. Second, the contradiction. Under its own heading, the category that doesn't work: accessibility overlays, Bemeir writes that courts have repeatedly found that overlays do not produce actual compliance and that sites relying on them remain vulnerable to ADA claims, that for enterprise compliance programs, the clear recommendation is to avoid this category entirely, that overlays are a compliance anti-pattern, and in another article that the overlay tool category has been thoroughly discredited by both accessibility advocates and courts, do not buy these tools. On the same domain it operates an accessiBe partner page stating that we have partnered with accessiBe to bring you a unified approach of service and support, reproducing accessiBe's own line, the #1 Web Accessibility Solutions for WCAG and ADA Compliance, and closing with an offer to help you get started on a project with accessiBe. Both statements are genuinely published on bemeir.com, so both are printed here. It costs Bemeir the two-point no-contradiction rung and nothing more, because the anti-overlay statement is real and it still earns its 4 points.

Skynet Technologies is two businesses on one domain and the score reflects both, so read the split before reading the number. The audit side is genuinely strong and takes 12 of 12 on conformance artefacts, the only full marks on that criterion here. Its manual accessibility audit page offers WCAG 2.0, WCAG 2.1 and WCAG 2.2 Level AA conformance testing and states that each issue finding is mapped to WCAG 2.1, 2.2 success criteria, ADA, EAA, Section 508, and other relevant standards, which is the only explicit criterion-level mapping commitment in this research and earns it 18 of 18 on the standard. It runs a dedicated VPAT and accessibility conformance report page, publishes its own accessibility statement, offers an optional re-audit service to validate the fixes, promises a compliance roadmap, and prices the work: manual accessibility audit just starts at $500.

Its method statement is careful and it names more assistive technology than anyone else here. It writes that its experts perform manual, automated, semi-automated, and assistive technology audit for digital assets, that for automated WCAG testing it uses Skynet Accessibility Checker, WAVE, Google Lighthouse and a11y, and that while it's quick, easy, and cost-effective, these tools may generate false positives or negatives, so it suggests automated audit as a good starting point. On the manual side it writes that the audit is performed by experts for all templates, web pages, PDFs, videos, and other digital assets using assistive technologies such as screen readers, keyboard-only navigation, and voice tools, and that unlike automated scans, it identifies usability barriers and complex WCAG failures that tools often miss. The named tools are JAWS, NVDA, ZoomText and Dragon Naturally Speaking, with TalkBack for Android and VoiceOver for iOS. That is 19 of 24, level with Snowdog, held back only by the absence of a figure and of published testing with disabled users.

What sinks it to fourth is the other half of the business and the shape of its Magento content. Its Magento presence is a widget, not a diagnosis: the Magento accessibility extension page promises Magento website accessibility improvements in just two minutes, states that it will take 5 minutes to install and that the widget is designed to work seamlessly with all major Magento themes, and answers whether it supports Magento multisite setup. That is installation and theme compatibility rather than storefront analysis, and no Luma, Hyva, swatch, faceted navigation or Magento checkout code-level finding appears on the pages read, which holds it to 9 of 20 on the criterion the lane is named after. On legal posture it takes 4 of 10. The widget page claims that All in One Accessibility is an AI based accessibility tool available with 90 plus features and supported in 190+ languages and that it enhances website WCAG compliance up to 90%, depending on the website's structure and platform and additionally purchased add-ons, alongside the line, is your site at lawsuit Risk, and a benefit bullet reading reduce the risk of legal lawsuits. A quantified compliance figure attached to a two-minute install costs it half that rung, it publishes nothing saying an overlay cannot achieve conformance, and the Overlay Fact Sheet lists All in One Accessibility by name among overlay products. One credential needs its exact wording preserved: Skynet writes that we are an organizational member of W3C and IAAP. That is corporate membership. No individual CPACC, WAS or CPWA holder is named, and no Deque University or Trusted Tester credential was found on the pages read.

InteractOne publishes the second most useful description of what actually breaks on a large Magento catalogue and the weakest standard statement of any company here that has a service page, which is an odd combination and it costs it heavily. Its ADA remediation article on a 60,000 SKU B2B catalogue names faceted navigation filters implemented as unlabeled links or divs, with keyboard focus thrown back to the top of the page on every refresh, spec and compatibility tables built without header cells or scope, which turns a fitment chart into noise, quantity steppers and bulk reorder grids that cannot be reached or operated by keyboard alone, part-number search that returns results without announcing them, and product images whose alt text just repeats the SKU. It adds a genuinely platform-specific warning, that on Adobe Commerce, a well-meant global change to a grouped product grid can break how customer-specific pricing displays, and no scanner will tell you that it did. That is 12 of 20.

The problem is where none of it lives. Its accessibility service page describes work tailored to your unique platform, like BigCommerce, Shopify, and Magento is not in that list; on that page Magento appears only inside client testimonials and in the footer navigation. This page does not score a testimonial as a company's published commitment, so the Magento evidence counts from the articles and loses the 3 points for sitting on a service page. Worse, the service page names no WCAG version and no conformance level at all, using WCAG standards and web accessibility guidelines, such as the WCAG throughout; the version and levels appear only in the articles, which is why it takes 15 of 18 rather than nothing. On method it is reasonable and unquantified, writing that we spot issues that automated tools can miss, from color contrast problems to missing alt text and broken navigation, and in its FAQ that automated tools can catch surface-level issues, but they often miss more complex problems like poor screen reader behavior or improper keyboard navigation, and that a manual ADA compliance solution, backed by human experts, offers deeper testing.

It takes 2 of 10 on legal posture, the joint lowest here, for three reasons that a buyer can check in a minute. Its own Magento accessibility guide recommends an overlay, listing UserWay as providing a comprehensive accessibility widget among tools to maintain compliance, while the same article warns to beware of one-size-fits-all tools, which is the contradiction that costs it the no-contradiction rung. The same article claims that achieving Level AA compliance automatically ensures compliance with other country-specific regulations, such as the ADA or the European Accessibility Act, which is a near absolute compliance claim. And its service page carries the heading Certified Website Accessibility Expertise over a body line reading, our team is trained in WCAG accessibility standards and the latest ADA requirements. Trained is not certified, no IAAP, Deque University or Trusted Tester credential was found on the pages read, and the gap between the heading and the sentence under it is the kind of thing a plaintiff's lawyer reads carefully. It scores zero on client evidence: the 60,000 SKU catalogue is a generic scenario with no client named, and the seven named testimonials on the accessibility page are all about development and Magento migrations rather than accessibility.

accessiBe is ranked here because it is in lane on the page's own test, holding a dedicated Adobe Commerce page, and because leaving out the vendor a buyer will actually encounter would make this list less useful rather than more honest. What it publishes is a compliance claim for an automated product: that accessiBe's AI-driven, automated solution helps make existing websites ADA-compliant by remediating the website's code in adherence with WCAG 2.2 Level AA, and that the solution is session-based, meaning only those with disabilities will see adjusted design elements, leaving your design completely intact. The same page names WCAG 2.1 AA as the standard a website needs to be remediated to in an earlier paragraph, so two different versions describe the same job on one page, and its global navigation repeats the line, the #1 cited in eComm ADA lawsuits.

One claim on that page is checkably wrong about the Magento ecosystem and it matters, because it is the premise of the pitch. accessiBe writes that there are currently no extensions, plugins, or preset themes and templates available that can help you make your website effectively accessible and that accessibility is completely a DIY affair for Adobe Commerce users. Hyva publishes an accessibility guide describing the work done on its theme and reporting that an audit identified 159 accessibility violations within Hyva; Skynet Technologies sells a Magento accessibility extension listed on the Adobe Commerce Marketplace; UserWay publishes a Magento plugin install guide. Three of those are competitors of accessiBe, which is exactly why the claim is worth checking rather than repeating.

What the independent sources say is kept separate from what accessiBe publishes, because that distinction is the whole point. The European Commission states on its own web accessibility page that overlays, or any other tools which do not ensure the website itself meets the detailed criteria of the standard, are not an appropriate solution, and that it is best to fix accessibility issues at their source. The Overlay Fact Sheet, signed by contributors and editors for the WCAG, ARIA and HTML specifications including Matt King, Chair of the W3C ARIA Authoring Practices Task Force, states that while the use of an overlay may improve compliance with a handful of provisions in major accessibility standards, full compliance cannot be achieved with an overlay, and that these products' documented inability to repair all possible issues means that they cannot bring a website into compliance. It names accessiBe among overlay products. accessiBe publishes no statement that an overlay cannot achieve conformance, which costs it the 4 point rung.

In fairness, it sells human work too, and that is where its 48 points mostly come from. Its expert audit service offers a detailed line-by-line audit and prioritized action plan, states that audits evaluate your site based on the most current WCAG standards and criteria including user interface components, forms, contrast, page structure, keyboard operability, and, ultimately, usability with assistive technologies, and includes a post-audit review within 90 days. It runs a VPAT service, litigation support and user testing. The heading over that audit offer reads, address accessibility needs that require manual remediation, which is an implicit admission that the widget does not cover everything, and it is also the contradiction that costs the last rung: the same site says the automated product makes websites ADA-compliant. One finding is worth stating plainly because of who it is about. accessiBe's own accessibility statement URL redirects to a PNG image file, so its statement is served as a picture rather than as text, which earns nothing on that rung. No named client with a measured accessibility result was found on the pages read.

Klizer, formerly DCKAP, is the only company here that publishes both a price and a genuinely candid limitation, and it also publishes an overclaim three paragraphs away from it. The price first, because almost nobody in this lane gives one: it states that the ADA complaint process at Klizer will start at $11K and it includes overall AUDIT and fixes for A and AA guidelines, with their typo for compliant preserved, and that the final price will depend on the size, current degree of complaint, and other potential issues. The candid part is a FAQ answer: asked whether a website can become 100% ADA-compliant, it answers that achieving 100% compliance is challenging with the developing technologies and current regulations, and that Klizer ensures your website meets the latest ADA standards, significantly improving accessibility and minimizing the risk of non-compliance. On the same page it also writes that you can get full compliance with WCAG success criteria and ADA, which is the near absolute claim that halves its legal posture rung to 2 of 4.

Its standard statement is current and complete enough to take 15 of 18, naming both a version and a level: it offers detailed reports that break down your website's compliance with Web Content Accessibility Guidelines 2.2 and ADA standards and states that we use assistive technology to address sensory guidance issues and meet the Level AA of the Web Content Accessibility Guidelines. What it does not publish is a mapping to individual success criteria; the report contents are listed as categories, color vision, alt text, deficiencies, keyboard navigation, rather than criteria. Its named assistive technology is broad but dated, listing screen reader, JAWS, NVDA, Windows Eyes and a text-zoom tool. Window-Eyes was discontinued in 2017, so that list has not been reviewed in some years, which is the sort of detail worth raising on a call.

Where it falls to seventh is Magento and process. Its accessibility service page contains no Magento content at all, and its Magento accessibility article repeats Magento in every heading, what is Magento accessibility compliance, Magento ADA compliance key requirements, strategies to get Magento accessibility compliance, without naming Luma, Hyva, swatches, layered or faceted navigation, the product gallery, Magento checkout or the Magento admin anywhere in it. That is keyword-level coverage and it earns 6 of 20. The article also claims that Magento encourages store owners to make their websites ADA-compliant and provides clear guidelines to help achieve this, with no citation to any Adobe source, which is unfortunate given that Adobe does publish a conformance report and it says the storefront partially supports twenty-nine criteria. Its published process step is assessment through automated accessibility tools, and it writes that our automated tool improves overall web accessibility and helps you avoid demand letters and discrimination claims, with no statement anywhere about what automated tools cannot find. No VPAT or ACR, no accessibility statement for its own site, no scheduled re-testing and no named client accessibility result were found on the pages read.

Williams Commerce finishes eighth and publishes the single best artefact in this entire research, which is a useful demonstration that a good score and a good document are not the same thing. Its accessibility statement states outright that currently, this website is not conformant with WCAG 2.2 level AA. It then enumerates roughly twenty of its own automated findings, each mapped to the success criterion it breaks, for example color-contrast, ensures the contrast between foreground and background colors meets WCAG 2 AA minimum contrast ratio thresholds, mapped to WCAG 2.2 AA 1.4.3 Contrast (Minimum), and focus-modal-none, when the modal dialog is activated, keyboard focus is not placed on or in it, mapped to 2.4.3 Focus Order and 2.4.7 Focus Visible. It then publishes a separate keyboard testing list, including skip to content doesn't work against 2.4.1 Bypass Blocks and a search layer keyboard-trap missing against 2.1.2 No Keyboard Trap, and a screen-reader testing list including logo, alt missing. It commits to responding to accessibility feedback within 30 business days and names the Equality Advisory and Support Service and the Equality and Human Rights Commission as escalation routes. A company publishing its own failures, mapped to criteria, with a response commitment and a regulator named, is the artefact every buyer should be asking a supplier for. It takes 18 of 18 on the standard largely because of it.

That same statement also documents a proper mixed method, writing that we employed a multi-faceted assessment approach encompassing automated testing with AxeDev Pro, manual testing focused on keyboard navigation, screenreader compatibility evaluations, and thorough image testing. Notably its accessibility service page does not: there the only named tooling is automated, that we use leading testing tools like Google Lighthouse to review your new website against all the important web accessibility standards and markers, covering multiple browsers and devices, with auditing and reporting available as standalone services. No screen reader or keyboard-only testing is described on the service page at all, and no statement about what automated tools cannot find appears anywhere on the pages read. Its standard statement differs between the two pages too, WCAG 2.1 AA on the service page and WCAG 2.2 AA in the statement.

What holds it to 43 is that it scores zero of 20 on the criterion this page is named after. Magento appears twice in the accessibility service page HTML, both times in navigation or footer context, and the build-phase copy is content-management-system generic, offering to review the code itself, as well as the editor's workflow within the content management system, to ensure accessibility is prioritised, with alt tags for images and caption fields. No Magento or Adobe Commerce accessibility content was found on the pages read. Its legal posture is clean and unexaggerated, writing that in our view, web accessibility should be something achieved in a complete and uncompromising way and that accessibility should not be a tick-box exercise, with having strong accessibility reduces legal risks rather than any guarantee. It publishes nothing about overlay widgets, no VPAT or ACR, no scheduled re-testing, no dated remediation roadmap and no named client accessibility result on the pages read.

Aureate Labs ties Williams Commerce at 43 and ranks ninth on the same declared tie-break, the heaviest criterion, where Williams takes 13 and Aureate takes 5. It has no accessibility service page: its page sitemap contains none, and everything here is two blog posts. Those posts are better than their tier suggests. On the theme layer it writes that Hyva Theme version 1.3.0 and later implements accessibility features according to WCAG, Web Content Accessibility Guidelines, 2.1 level AA requirements, that the 1.3.0 update brought in significant advancements by aligning the theme with WCAG 2.1 Level AA standards, and, importantly, that prior to this release, while the theme prioritized performance and modern design, it lacked specific accessibility enhancements. It names EN 301 549 as the European standard for digital accessibility and cites WCAG 2.2 as the latest version in its UX post. That earns 17 of 20 on Magento specificity, third highest on this page. The accessibility work on Hyva 1.3.0 is credited to Hyva, not to Aureate Labs, and this page does not read it as Aureate's own delivery.

Its storefront advice is concrete where most companies are vague, and it is aimed exactly where Adobe's own conformance report puts the exceptions. It writes that you should start with form labeling fields, that every input should have a visible, persistent label, not just placeholder text that disappears once someone starts typing, and that every part of the checkout process, from adding details to confirming the order, should follow a logical tab flow. It carries sections on ensuring accessible forms and checkout and on additional tips to maintain accessibility for a Magento store, and it illustrates alt text specificity, preferring a sky blue leather satchel with perforated texture over a blue handbag.

It also publishes a figure for what automated testing misses, which only Bemeir otherwise does, writing under a Manual versus Automated Testing Tools heading that automated tools can quickly identify common issues like missing alt text or color contrast violations, yet typically catch only about half of accessibility problems. That about half is more generous than Bemeir's 30 to 40 percent and than the ceiling the Overlay Fact Sheet cites, and a reader should notice that the three published figures in this market do not agree with each other. What Aureate Labs does not publish is any method of its own: no manual testing process, no assistive technology, no audit scope, which is why it takes only 5 of 24 on the heaviest criterion, all of it from that one figure. No VPAT or ACR, no accessibility statement for its own site, no scheduled re-testing, no named client accessibility result and no position on overlay widgets were found on the pages read.

Wagento finishes last of ten and it finishes last for one reason: it publishes no accessibility method of any kind, taking zero of 24 on the heaviest criterion. There is no accessibility service page either. The path a buyer would guess, its ADA compliance URL, redirects to the homepage, and its page sitemap, which runs to 139 KB of URLs, contains no accessibility page. Everything here is two blog posts.

What those posts do contain is the most granular published list of Hyva accessibility changes in this research, and it is component-level rather than abstract. Reporting on Hyva Themes 1.3.0, Wagento lists enhanced top-menu sub-menus to support both hover and keyboard navigation, improved focus management for keyboard navigation when the mini cart is open, enhanced product swatches with accessible labels, product names added to Add to Cart and Add to Wishlist button aria labels on the product grid, expanded category and search results page filters made accessible for assistive technology, improved color contrast on list and grid mode selection in the PLP toolbar, and optimized product rating form radio buttons for accessibility. It names the standard, writing that the update ensures websites are not only user-friendly but also inclusive and compliant with the WCAG 2.1 level AA requirements. That is 17 of 20 on Magento specificity, the same as Aureate Labs, and it is the whole of Wagento's score alongside the standard and a clean legal posture.

The attribution deserves credit rather than a deduction, because it is the correct kind. Wagento does not claim this work. It writes that after months of dedicated effort, the Hyva team have released Hyva Themes 1.3.0, and describes itself as a proud partner of Hyva that believes this release sets a new benchmark. Reporting a partner's release accurately is better practice than absorbing it, and it is worth contrasting with Snowdog, which claims authorship of accessibility work incorporated into the same version. This page records both statements and adjudicates neither. Nothing else was found on the pages read: no audit method, no VPAT or ACR, no accessibility statement for its own site, no named client accessibility result, no certifications and no published position on overlay widgets.