LinkedIn engineers build and test products against WCAG 2.2 Level AA using internal accessibility tooling and third-party audits, iterate on Disability Answer Desk feedback, and triage accessibility bugs alongside functional ones.
ENABLE Model location
What it is
LinkedIn is a Microsoft-owned professional networking platform with over 1 billion members worldwide. As part of its accessibility program, LinkedIn has adopted WCAG 2.2 Level AA as its standard for product design and development. The company continuously audits its products -- internally and through third-party testers -- using assistive technologies including JAWS, NVDA, VoiceOver, and TalkBack.1
LinkedIn's accessibility infrastructure is notable for scaling builder-side accessibility across a massive engineering organization. In development, engineers build features against WCAG standards using automated accessibility testing integrated into the continuous integration pipeline -- using Deque's axe-core for web, Google's GTXiLib for iOS, and Google's Accessibility Test Framework for Android -- so that accessibility violations block code merges before reaching users.2 This approach, documented in LinkedIn's engineering blog, was among the first at enterprise scale to treat accessibility failures as build-blocking errors, not just post-launch bugs.2 In QA testing, products are audited with screen readers and assistive technology, and LinkedIn publishes conformance reports for Section 508, WCAG, and EN 301 549 standards -- a transparency practice rare among major platforms.1 In triage, accessibility bugs are identified and prioritized alongside functional bugs by engineering teams, using internal audits, automated test results, and code reviews. The Disability Answer Desk, available via Be My Eyes, acts as a builder-side support channel: it gives blind and low-vision users a live route to LinkedIn specialists after launch and feeds barrier reports into later triage and iteration. It does not replace internal audits, QA testing, or accessibility-focused development as the main ways barriers should be caught. In iteration, the company integrates feedback from all these channels into subsequent product releases.
LinkedIn's contribution to the ENABLE Model is its demonstration that accessibility can be embedded into engineering culture and infrastructure at scale as a build-blocking requirement, a reporting practice, and a distributed expertise model (via the A11y Champions Program), instead of as a compliance afterthought.2 This shifts the burden upstream, reducing navigator-side compensations and making accessibility a default part of product development.
To scale accessibility expertise beyond its core A11y Engineering team, LinkedIn launched an in-house A11y Champions Program in July 2017 -- a quarter-long training program where engineers commit 25% of their time to accessibility work, complete Deque University certification, receive mentorship from the A11y Engineering team, and return to their product teams as local accessibility experts. As of 2019, the program had graduated over 40 Champions across the engineering organization.4
LinkedIn also publishes Section 508, WCAG, and EN 301 549 conformance reports through its Accessibility Report Center, and has committed to meeting the European Accessibility Act requirements for LinkedIn Learning.1 The company has introduced features including auto-generated captions for video, alt-text prompts for image uploads, and the addition of "Dyslexic Thinking" as a recognized LinkedIn skill.1
Why it matters
Professional networking platforms are gatekeepers to employment opportunity. When LinkedIn's product is inaccessible, disabled job seekers, employees, and professionals are locked out of the primary platform through which many jobs are found, professional connections are made, and career advancement happens. This is not a convenience issue -- it is a barrier to economic participation.
LinkedIn's approach matters because it embeds accessibility across the product lifecycle rather than treating it as a post-launch compliance exercise. The automated testing pipeline catches common violations at build time -- the engineering team notes that "automated testing can identify issues faster than the amount of time it takes a human manual tester to file a single bug" -- while the A11y Champions Program distributes accessibility expertise into product teams so that engineers catch issues during development, not after release.2 The Disability Answer Desk creates a direct feedback loop between disabled users and the product teams who build LinkedIn's features, allowing accessibility issues to be triaged and iterated on continuously. The publication of conformance reports provides transparency about the current state of accessibility, rather than making vague claims of compliance.
Real-world example
In May 2021, LinkedIn's Disability Answer Desk became a Specialized Help provider on Be My Eyes, available 24/7 to assist blind and low-vision users with product issues, accessibility questions, and assistive technology support across LinkedIn, LinkedIn Premium, and LinkedIn Learning.3 Rather than relying solely on automated accessibility testing or internal auditors, this partnership offers users the ability to connect directly with LinkedIn's accessibility specialists through a live video call. This acknowledges that even well-tested products will have gaps -- and that disabled users need support channels that go beyond filing a bug report and waiting.
The A11y Champions Program has changed how product teams operate. Graduate Kirsten reported: "I now have the knowledge and voice to advocate for a11y in our product holistically, from product specs, to design, to engineering... it has been a huge change that I feel every team absolutely needs." Graduate Erin described catching "many accessibility bugs before they became public" through code reviews.4 Meanwhile, the company's announcement of "Dyslexic Thinking" as a recognized LinkedIn skill -- promoted in a blog post -- represents a different kind of accessibility intervention: reframing disability-associated cognitive styles as professional assets rather than deficits, changing how employers and recruiters encounter neurodivergence on the platform. However, it is unclear if "Dyslexic Thinking" is available as a standard selectable skill for all users in the LinkedIn UI.1
What care sounds like
- "Every product release is tested with JAWS, NVDA, VoiceOver, and TalkBack before shipping."
- "Our Disability Answer Desk gives users a direct channel to report accessibility barriers -- and we route those reports to the teams who can fix them."
- "We publish our conformance reports publicly so anyone can see where we stand."
What neglect sounds like
- "We tested with automated tools -- that covers accessibility."
- "Users can email support if they have an issue."
- "Accessibility conformance is an internal document -- we don't need to publish it."
What compensation sounds like
- "I had to find a sighted colleague to help me navigate LinkedIn's job application flow."
- "The Disability Answer Desk helped, but the bug I reported six months ago still isn't fixed."
- "I switched to email networking because the platform's screen reader experience was too frustrating."