Beyond EAA compliance: Accessibility becomes the benchmark for digital quality

Over a year has passed since the European Accessibility Act (EAA) took effect, yet many companies are still uncovering gaps in the accessibility of their digital services. While some digital platforms may comply with recognized accessibility standards, technical conformity is not always sufficient to ensure an accessible user experience.

People who rely on assistive technologies may still face obstacles such as unclear workflows, error messages, and content that is hard to interpret with screen readers. Accessibility is now seen as a key factor in digital quality rather than merely a compliance requirement. Nevertheless, creating accessible digital experiences demands more than good intentions.

It necessitates thorough assessments, audits, and real-world testing to validate accessibility in practice. This challenge is particularly significant for tech firms, where accessibility must be addressed across an array of products, development teams, and release cycles. To achieve this, a continuous approach to accessibility is required, built on a robust accessibility ecosystem that integrates assessments, audits, automated testing, and feedback from people with disabilities throughout the software development lifecycle.

Assessments serve as the initial step in accessibility improvement. They offer a broad overview of accessibility risks, helping organizations pinpoint priority issues early and identify areas for targeted investigation. These assessments typically employ automated tools alongside focused manual reviews of designs and/or products to guide organizations on where to concentrate their accessibility efforts for the greatest quality improvements.

Audits represent the next phase of this process. They conduct a thorough evaluation of digital products in line with recognized standards like the Web Content Accessibility Guidelines (WCAG 2.2) and the EAA. Audits document accessibility issues, assess their severity, and provide remediation recommendations. While assessments cover the broad strokes, audits delve deeper by systematically examining entire products or services to confirm compliance and outline a clear remediation plan.

Audits may also extend beyond websites and applications to encompass customer-facing digital content such as PDFs, presentations, and other documents subject to accessibility legislation. While assessments and audits establish a solid foundation for an effective accessibility program, they do not guarantee a positive user experience.

Individuals using assistive technologies might still encounter barriers that automated checks and standards-based testing cannot identify. Therefore, audits should be part of a broader accessibility strategy, supported by expert evaluations and input from people with disabilities. While automated testing tools have become crucial in modern software development, they have limitations.

These tools rapidly pinpoint common coding issues like missing alt text, insufficient color contrast, and improper heading structures, helping teams resolve straightforward problems early in the development process. They also integrate seamlessly into CI/CD pipelines, effectively catching basic accessibility issues before software is released.

AI is enhancing the capabilities of these tools, enabling them to recognize patterns, cluster similar issues, and suggest potential fixes, thereby reducing repetitive work for development teams. However, AI cannot substitute for human expertise. Machine-generated fixes may be inaccurate, tackle superficial issues, or provide a misleading sense of accessibility if the overall user experience has not been adequately evaluated.

Even with these advancements, automated testing has its limitations. As the W3C points out, “web accessibility evaluation tools cannot determine accessibility; they can only assist in doing so.” They cannot ascertain if link text is meaningful in context, whether keyboard focus follows a logical sequence, or how someone using a screen reader perceives a user journey.

Automated tools highlight symptoms but lack the human context to fully grasp the impact on usability. Hence, effective accessibility programs combine automated testing with manual expert reviews and testing by people with disabilities who daily use assistive technologies. By merging these approaches, organizations gain a more comprehensive understanding of how people interact with digital products, helping them identify accessibility barriers that automated testing alone might overlook.

Building an accessibility ecosystem Organizations should regard accessibility as a continuous capability rather than a one-off project with a fixed end point. Instead of relying solely on periodic audits, they need to establish an accessibility ecosystem that incorporates inclusive practices throughout the development process. An effective ecosystem unites developers, designers, accessibility specialists, automated testing tools, and people with disabilities in a continuous feedback loop.

By integrating accessibility considerations into planning, design, development, testing, and release, organizations can identify and resolve barriers throughout the development process, reducing the effort required to address accessibility issues and preventing delays that can arise when problems surface during final compliance checks.

Microsoft's journey Microsoft is one example of a tech company that has embraced these principles across its Cloud & AI portfolio, which comprises over 1,000 products. Rather than viewing accessibility as a compliance exercise, Microsoft has cultivated an accessibility ecosystem that permeates all aspects of its product development.