GAAD 2024 Slovenia – state of e-commerce accessibility – a year before the new legislation

13th Global Accessibility Awareness Day was second official GAAD in Slovenia and I am once again honored to be a part of it. My idea for this time was to examine the state of accessibility of e-commerce in Slovenia (as new legislation starts in 2025 and we do not have any other data available).

This time I decided to leave automatic accessibility testing out and check half or WCAG 2.1 on A and AA levels manually. As manual audits take time, I decided to limit our analysis to half of it and to only check 20 e-commerce websites that we are quite confident is also made in Slovenia (so we dropped the multinationals operating in Slovenia). This project was made by a team of accessibility auditors, some senior and some junior (with verification and help), part of Institute for digital accessibility A11y.si (opens in new window, in Slovenian).

Motivation

As mentioned – there are not a lot of similar studies that use only manual testing. Automatic accessibility testing only gets you so far, so I decided that e-commerce deserves better quality.

Some WCAG success criteria is also very specific and is often not found on home pages, so we didn’t limit us on those alone. We tried to analyze real world scenarios:

• search for products or services,
• add to cart flow,
• registration,
• contact,
• finding general information

So your typical user journeys that are crucial for all users.

We did not expose domains and any data that would identify e-commerces, our motivation is to present overall accessibility and have the data to compare the trends in the future (especially when European Accessibility Act will be enforced in 2025).

Methodology

Scope was 20 e-commerce websites which we are confident are really made in Slovenia and have more than 2 million Euro turnaround and more than 10 employees.

There are thousands of smaller e-commerce sites, but those will be exempt from the new legislation (mostly micro business), so we targeted the bigger ones that are not founded (and developed) by large international players. Such players will probably have their products fixed centrally, and, I suspect, faster due to larger pan-European marked risks.

Selected WCAG success criteria

As mentioned, we only evaluated 26 WCAG 2.1 success criteria. Those were:

1. 1.1.1 Non-text Content – WCAG 2.1 A
2. 1.4.5 Images of Text – WCAG 2.1 AA
3. 1.2.2 Captions (Prerecorded) – WCAG 2.1 A
4. 1.3.1 Info and Relationships – WCAG 2.1 A
5. 2.4.6 Headings and Labels – WCAG 2.1 AA
6. 2.4.2 Page Titled – WCAG 2.1 A
7. 3.1.1 Language of Page – WCAG 2.1 A
8. 1.4.1 Use of Color – WCAG 2.1 A
9. 1.4.3 Contrast (Minimum) – WCAG 2.1 AA
10. 1.4.11 Non-text Contrast – WCAG 2.1 AA
11. 1.4.4 Resize Text – WCAG 2.1 AA
12. 1.4.10 Reflow – WCAG 2.1 AA
13. 1.3.4 Orientation – WCAG 2.1 AA
14. 2.1.1 Keyboard – WCAG 2.1 A
15. 2.4.1 Bypass Blocks – WCAG 2.1 A
16. 2.1.2 No Keyboard Trap – WCAG 2.1 A
17. 2.4.3 Focus Order – WCAG 2.1 A
18. 2.4.7 Focus Visible – WCAG 2.1 AA
19. 2.4.5 Multiple Ways – WCAG 2.1 AA
20. 2.2.2 Pause, Stop, Hide – WCAG 2.1 A
21. 2.4.4 Link Purpose (In Context) – WCAG 2.1 A
22. 1.3.5 Identify Input Purpose – WCAG 2.1 AA
23. 3.3.1 Error Identification – WCAG 2.1 A
24. 3.3.2 Labels or Instructions – WCAG 2.1 A
25. 3.3.3 Error Suggestion – WCAG 2.1 AA
26. 4.1.3 Status Messages – WCAG 2.1 AA

Preliminary results

We will publish more findings in Slovenian language soon, but to provide few hints:

• Average website had 16.95 WCAG failures (out of 26 tested).
• Most accessible website had only 13 fails, least accessible had 20 fails.
• All sites failed 1.1.1, 1.3.1, 1.4.3 and 4.1.3
• Most sites failed 1.2.2, 2.4.1, 2.4.7, 2.4.4 and 3.3.2
• Least sites failed 1.3.4 and 2.1.2

We also tried to collect “interesting” findings and found the following:

• 350 tab key presses were required to reach to main content on a single online store (also missing skip to content and focus visible),
• modal windows were often invisible to screen-readers and unable to operate for keyboard users,
• CAPTCHA was still a problem in some cases, making it impossible for blind users to register,
• we also found a HTML table including another table that included another table – just for the purpose of designing a page.

Building awareness

Most of the issues were actually very simple to fix and my contribution to this GAAD was to make people aware how easy it is to fix – or even better prevent – them. A lot of them could have been prevented in minutes if designers and developers knew about them and planned accordingly.

I also pointed out that accessibility benefits more than just people with disabilities – it also benefits search engines and large language models. Semantics that is needed for assistive technologies and even color contrasts are beneficial for all of us, but they are also beneficial to crawlers and computer vision.

We still need more accessibility awareness, that’s vital. I am happy that more than 150 people attended (both on-site and on-line).

I need to work on my presentation skills, but I think I got the message out there. The problem is that it didn’t get any main stream media attention and I made some suggestions to the board on how to maybe change that.

It was nevertheless warming my hearth with a good discussion in the panel and when stakeholders approached me to build new relationships. Key to bringing the accessibility field to wider public.