WebAIM's Million 2024 is out!

I was very happy to hear the planned release time for this year's report. I couldn't wait to get my hands on the data.

The most common errors are still the same

Over 96% of all errors detected fall into these categories. No surprise there. The following are still the most common:

• Low contrast text
• Missing alternative text
• Missing form labels
• Empty links
• Empty buttons
• Missing document language

While the percentages have been going down, what I find quite alarming is that the number of empty buttons is not going down. I have a really hard time understanding why it is so hard to add descriptive text to all links and buttons. It shouldn't be that hard.

Low contrast text is still the most common error

The error is still found on over 80% of home pages.

Sometimes these can happen due to a font using a separate font instance for bold instead of using font-weight. I know from personal experience But I doubt very much that the issue is all that common.

A lot of pages have low contrast text. How can it be? Pretty much all automated tools will report on this, even the free ones. So if anyone has ever taken a look at the front pages with a monitoring tool, these issues should have been fixed already.

Not only are the issues easy to locate, they are also easy to fix on the code side. Well, unless of course, the problematic color is a brand color. But then I'd like to ask the question who picked an inaccessible color for the brand? Well, the question I'd more like to ask is why was it chosen and why hasn't it been changed already. Brands change their look all the time. There are so many shades of colors that it really isn't too hard to find several colors that have enough contrast.

I can understand that not every brand can support AAA-level requirements for their colors. But AA is doable. There are no excuses.

One out of every four linked images was missing alternative text

This is a real problem. When this happens, the links themselves are not descriptive. So this also fails two WCAG criteria instead of "just" one.

All images that act as links need to have a descriptive alternative text that describes the link target.

One third of images don't have good alternative text

The data shows that one third of images on popular home pages may have missing, questionable, or repetitive alternative text. One third!

This includes images that have image alternative text such as alt="image", "graphic", "blank". Or the file name. Or an alternative text that is identical to adjacent text or the alternative text of an adjacent image. We clearly still have much to learn about the alternative texts.

Heading levels are still getting skipped

According to the data, skipped heading levels were present on 37.9% of pages. The number has gone down a bit.

What I found interesting is also that the number of pages with no headings present at all went up. Not great.

Over 30 % of form inputs don't have a proper label

Every form input should have a label. You can label a form field with either

• <label>,
• aria-label,
• aria-labelledby, or
• title.

ARIA use has increased by a lot

According to the data, ARIA code usage increased 15% compared to last year.

Would be interesting to know how much of this ARIA was needed.

Also, sites with ARIA usage usually had a higher number of detected errors. But this might not mean anything ARIA-related, the sites might just be more complex. Complexity usually means more errors.

Links continue to have ambiguous link text

13.2% of pages had ambiguous link text. It is a bit less than last year, which is good.

You should generally avoid ambiguous link text, such as "click here", "more", or "continue". They can be only understood within the context they are in. Well, assuming the context tells the user anything more.

The optimal situation would be to have each link be understandable on its own. But for AA-level WCAG compliance, having links that are understood within their context, is ok. So you might, for example, have a link inside a sentence. I would still avoid the examples mentioned above.

Skip links are problematic

Less than 20% of home pages had a skip link present. The number has gone down, which is not great.

What makes the situation worse, one out of every 10 skip links was broken. Either they were inaccessible or the targets didn't exist.

It doesn't help much if we add broken skip links, now does it? Remember to always test these when you make them.

What is the conclusion of the report?

There are more errors. But the results are not all bad.

Sites that had fewer errors before have been improving. This would suggest that increased focus on accessibility has an impact. And I believe it does.

I see a big difference in my work if a client really cares about accessibility. Clients who care and because of that focus more on the topic, have been able to reduce the number of errors present on their site. Some have had quite big numbers even. But the clients who don't care, haven't seen such improvements. I hope I can make more clients of mine care about this important topic during this year!

Regulation in general seems to work. Sectors like government and education are doing much better than others. This does make sense because money is a motivator. It's a shame that it's needed, but I'm still glad to see that it has an effect.