After doing many full-blown accessibility audits of large public and private sector websites and native mobile applications, I tried to reflect over the owner/stakeholder perspective. To audit accessibility of larger products takes time. If we round the WCAG success criteria to 50, and define that we need a minute for every success criteria we quickly calculate that auditing hundreds of URLs / components / views takes hundred hours.
Changes to the site / app done while we do the auditing may be very dangerous as they potentially introduce new accessibility issues that can be missed, so they must ideally wait for the audit to be finished or at least synchronize scope all the time, so that we get changes only in parts that still need to be audited. Sure, design systems help, but at the end of the day they are only a part of the puzzle as we need to check the content as well.
Agile world requires agile accessibility testing as well
It’s obvious that before-mentioned process can slow things down and also make everything more expensive. Especially when we try to work agile – in sprints or similar project methods, continuously delivering value. There we can perhaps choose multiple paths, for example a path that looks at the components, a path that checks scope based on WCAG Evaluation Methodology and a more agile path that follows the recent deliveries. For sure agile alone is not enough, but in some cases it can be a good and effective start, especially if our activities reach out to adding new components to a mature design system.
Accessibility maturity of organization is crucial, but if we are just starting with accessibility program, then it’s probably smart to divide the efforts and make sure we prevent new accessibility issues and detect and fix existing ones in parallell. Based on most crucial end user journeys, most popular pages and views, most valuable paths, we can define our scope and act on it, synchronized with change management.
Adding accessibility to definition of done, acceptance criteria, and other mechanisms and even rituals to make accessibility integral – and shifted to left – is best practice.
Often we do not need full-blown accessibility audits
Sure, ideally we would need to do audit of everything. But often that’s not viable and economic. Therefore we have the WCAG-EM scope process in the first place – admitting that it’s often not possible to measure accessibility of all parts and that we need to find representative samples, critical paths and add some randomness. Just to be more confident. But still – another audit could reveal different, previously undetected accessibility issues.
But we can also start slowly and make it incrementally. First check for the most serious accessibility barriers, then proceed with others. Save time and provide results at speed.
It requires a good partner on the owner / stakeholder part – to pinpoint most critical parts, but that goes also for WCAG-EM anyway. But often we can quickly do an audit of main elements and incrementally do other audits when previous findings are fixed. Starting with navigation, structure, main templates, forms and proceeding with design system can achieve optimized results and expand our knowledge of the product, if we also get dialog from owners / stakeholders that know the most important and valuable parts.
This saves resources and money, and works quite well on the long run. Surely, we can’t have accessibility conformance report done quickly (unfortunately those are often the required output), but we can make the product more accessible faster and less expensive in most cases.
Incrementally doing audits of smaller parts and identifying key components, getting into design systems and independent parts, concentrating on templates and content at the same time and properly describing and identifying the problems and solutions for the problems will, ideally, also provide knowledge spreading for people that will act on reports. Adding proper training to the situation, so that people will be more independent and can also fix older things, would then spread through organization, when management would prepare the conditions for it.
Once again we can conclude that accessibility is a team effort. That auditors can’t really work isolated if we want to make them most effective, that auditing at the end is a costly affair and that we can also shift accessibility auditing to the left to make it more effective.
