174: Beyond the Scan: 7 Manual Accessibility Tests to do Right Now, Meridian Hive Peach Ginger Wine Cocktail

By /

Automated accessibility testing is important, but it cannot tell you whether real people can successfully use your website. In this episode of Accessibility Craft, Amber, Chris, and Steve explain how manual accessibility testing fills the gaps left by automated tools and share practical tests that anyone can begin performing today.

The conversation covers keyboard navigation, browser zoom, screen reader testing, dynamic components, focus management, heading structure, and the accessibility problems that automated scanners frequently miss. The hosts also examine several developments from Converge Accessibility’s June 2026 legal update, including a case that demonstrates why documenting ongoing accessibility work can matter during litigation.

Before talking about manual accessibility testing, the team tests Meridian Hive’s Peach Ginger Wine Cocktail, a lightly carbonated mead cocktail that produces surprisingly divided opinions about sweetness, peach flavor, and for one host, a mysterious lingering funk.

Episode Outline

  • The team reviews Meridian Hive Peach Ginger Hard Honey
  • Why maintaining audits, scans, and remediation records may strengthen a legal defense
  • An inaccessible clickwrap agreement and the legal implications of asking someone else to accept it
  • Proposed accessibility-lawsuit reporting requirements in California
  • The return of the federal Online Accessibility Act
  • Why automated tools cannot identify every accessibility barrier
  • Where to begin when manually testing a large website
  • A practical 15-minute homepage accessibility test
  • Keyboard commands every manual tester should know
  • Testing websites at 200% and 400% zoom
  • Why mobile navigation might still be relevant for desktop users
  • Getting started with VoiceOver and NVDA
  • Common problems that only manual testing reveals
  • Testing modals, carousels, search filters, headings, focus order, and dynamic content
  • A reminder of what accessibility is really about

Links & Resources Mentioned

Download Accessibility Checker and use coupon code AccessibilityCraft to save 10% on any plan.

Subscribe for more practical conversations about digital accessibility, WordPress, emerging legal developments, and beverages that inspire unusually serious debate.

Accessibility Craft is hosted by Amber Hinds, Chris Hinds, and Steve Jones. They are experts in digital accessibility and creators of software, courses, and specialized services that have made millions of websites more accessible through their work.

To learn more about us, you can visit our website.

Listen

Watch

Youtube video

Transcript

​Chris Hinds: Welcome to Accessibility Craft, where we explore the complex challenges and emerging trends that are shaping digital accessibility, while sipping on unique craft beverages. This show is proudly produced by Equalize Digital, The most trusted name in WordPress accessibility. Join us every week as we break down accessibility news and share the expert strategies we’ve used to help make millions of websites more accessible.

Grab a drink, the show starts now!

Amber: Hey everybody, it’s Amber, and I’m here today with Chris.

Chris: Hello everyone.

Amber: And Steve.

Steve : Hey, everybody.

Amber: And this is episode number 174 of the Accessibility Craft podcast, which means you can find show notes and a full transcript at accessibilitycraft.com/174. Before we dive in, I want to ask you, yes you, people who are listening right now or watching on YouTube, please do us a huge favor and hit that subscribe button if you haven’t.

We have 724 YouTube subscribers as I’m recording. Let’s see if we can make that number get a little closer to 1,000. We would love you if you would do it. Or wherever you are listening, another great way to help us out is to give us a review. Hopefully, it’s five stars. If it’s not, don’t give us a review. No. We’ll take your four-star reviews, too, or any feedback.

So, anyway, now that I have done all the requisite I am a YouTuber/podcaster begging and pleading, I feel like I need a drink.

Today’s Beverage

Chris: Yeah. Well, I’ve got just the drink for you, Amber. We are having a total mom drink today. We are trying out Meridian Hive’s Peach Ginger carbonated wine cocktail with added peach, and it is from a meadery. So that’s not a meatery, like a place you would go to eat meat. It’s a meadery or a mead brewery. They make mead.

Amber: So first of all, I wanna ask right now if I should be offended that you said the mom wine cocktail is a drink for me because, mean, I think we all have memories back to the ’90s and our mom drinking those wine cooler things in the…

Steve : Yeah.

Amber: … glass bottle. I don’t know what brand they were. And I’m not sure that’s a cool

Chris: Well, this isn’t Chablis or whatever it was, right? That comes in a box. This is a little bit… This is for the sophisticated moms, the moms with good taste, right?

Amber: Oh, my mom was fancy. She didn’t drink out of a box, but it was glass bottles, but it was still a, a wine cooler, whatever that is. I have no idea.

Steve : Straight from the bottle.

Chris: On ice too, if I remember correctly, the stories you told me, right?

Amber: Well that, I mean, that’s my stepmom. Hopefully she doesn’t listen to this. She still does. She likes to drink very sweet wine with ice. Kinda like a sangria, I guess, only it’s not a sangria. She just puts ice in her wine. Yeah.

Steve : I always feel like that waters it down.

I was at a family thing and somebody had some wine and they’re like, “You want some?” I was like, “Sure.” And they got these big wine glasses and filled it all the way to the top of the glass, and then put ice in it and handed it to me.

And I’m like, “Why’d you put ice in it? You ruined it.” And then I took a drink. I was like, “Wait, this doesn’t even taste like wine. It’s sh- it’s like sugar water,” you know?

Amber: Yeah. So, so that’s the question. Is that what this is going to be like?

Steve : Oh, let’s see.

Amber: Meridian Hive.

Chris: Yeah. Is it gonna be sweet or is it gonna be a dry… ‘Cause you can do dry meads, right? They can just ferment all the sugar out of the honey. And it’s 5% ABV. We’ll see. I’m gonna crack this open.

Amber: I’m gonna admit that I actually don’t drink much mead and I don’t know much about it. So I was reading the can, and it says, “Ingredients: water, honey, peach concentrate, ginger, potassium metabisulfite,” in parentheses, “for freshness.” And I was like, what makes this alcoholic? I have no idea. You can ferment… the peach or it’s the honey?

Chris: The honey, yeah, ’cause it’s sugary. Just you know, when you press grape juice and put it in a vat, you know, the bacteria, the yeast get on it, and they ferment the sugar into alcohol…

Amber: Yeah.

Chris: …and they off-gas.

Amber: I guess if you read the ingredients of a wine bottle, it would say grapes.

Chris: It’s just grape must.

Amber: It’s the same thing. Yeah. I bought a… I wanted to see what color it was because the can is such a pretty pink, and it says peach, and I was like, “Maybe it’ll have a really nice rosé kind of color,” but it might not. Have a pint glass, though. I don’t know if this is the right glass for a mead, but that’s what I’m pouring it into, is a pint glass.

Steve : So definitely not a wine cooler.

Chris: No. No, it’s definitely, it’s got a different flavor to it, and it has that characteristic mead finish on it. But I’m gonna let Amber taste. She’s showing off the color right now, which is like a light straw yellow. Little bit of an orange tinge to it.

Amber: A lot of bubbles.

Steve : Yeah.

Amber: They’re, so they’re not running in a stream like champagne, but it’s really bubbly, but no head.

Steve : None at all.

Amber: Yeah, very interesting. Right.

Chris: The nose for me is kind of muted. I didn’t smell a lot of peach. I expected to smell more peach than I did.

Amber: Oh, I smell peach.

Chris: I smell a bit of it, but it’s not… But I’m also trying to get a smell out of a little tiny hole in the top of a can, which is not ideal.

Amber: Or this is…

Chris: So this can be forgiven for an unimpressive nose.

But the flavor on the other hand…

Amber: Or this is gonna be like the time we had strawberry beer, which I picked out, and I was like, “It tastes like strawberries,” and you guys were like, “No.” You couldn’t get it at all. I’m like, I get the peach.

Steve : I’m getting honey wine, not peach.

Chris: Yeah. Yeah, so it’s it’s got a, a bit of a tartness to it. There’s definitely a sweet undertone. It is sweet.

Steve : Yeah.

Chris: I do get peach, but peach feels secondary, and I get the thing I always get with mead, which I went out of my normal zone picking this one out, ’cause I don’t normally go for meads, as evidenced by the fact that I think we have literally never done a mead on Accessibility Craft before. Because they all have this kind of long, mushroomy kind of aftertaste, like a bit of a funk that just doesn’t well with me personally.

But I think it’s a personal taste thing, ’cause this, if you like mead, you would probably like this. But for me personally, I’m not super into it.

It kind of tastes like peach. What do you think, Steve?

Steve : You’re saying like it leaves like a film on your tongue? Is that kind of like…

Amber: I don’t get that.

Steve : It’s…

Amber: I wouldn’t say mushroomy is a word I would describe this as.

Steve : It ling- it lingers.

Chris: S- it’s funky.

Steve : It’s very sweet. I like the bubbles. I mean, I don’t mind it. I can drink it. It it’s not offensive to me. But it does linger, that whatever you’re trying to describe with the mead, it does linger for me quite a bit, but…

Amber: So I have had other meads. It’s been a while. I feel like I have them when we go down to our nerdy board game nights because one of the guys that Chris is friends with, he and his wife always bring mead. This is…

Chris: He makes it.

Steve : So…

Amber: Yeah, he makes it. This is an easier drink than his mead. Which I’m always like I don’t know about that.” I feel like this has a very friendly flavor to me. I don’t think it’s over sweet. I have definitely had way sweeter beverages. I get the honey. I get the peach. I’m not a huge like honey person. I don’t… When I have a biscuit, I don’t like put honey all over it like some people do. I’m like, “Nah, no, too sweet. I don’t like that flavor.” But I don’t mind it in this drink. It’s not overwhelming. And yeah, I like the bubbles, like. All right, well fine. Okay, it is a drink for me.

Steve : It’s a mom drink.

Amber: And I was sitting here being like, “I won’t like it because Chris called it that.” And now here I am, I like it.

Steve : Yeah.

Amber: I might actually give it two thumbs up reluctantly. I was, like, debating do I only give it one? But like I wouldn’t drink it all the time, but I see scenarios in which this would be great, and it feels lighter than some of the other things that I frequently drink.

There’s nothing bitter about it. I don’t get any funk like Chris says. I don’t know. But…

Chris: That’s, to be clear, like when I say that, it’s only on the finish and after I’ve swallowed the drink, right? It’s like just kind of a, a funkiness that lingers that’s characteristic of meads that I personally taste. Where the beverage leads, like the second it hits your lips, like it’s sweet, it’s peach, it’s bubbles, right? And that’s great, you know? I can see people liking that. Also, the level of sweetness in this one has me… I don’t know if anyone else is familiar with this feeling, but when you have one of those sweet alcoholic beverages where I feel like if I keep drinking this, I’m gonna get a headache and I’m gonna feel bad in about two hours. And this is not, I’m just kind of getting that feeling from this. And again, this is all personal taste. So when I give my rating, I’m gonna try to divorce personal taste and feeling from how it actually tastes, because this is a serious beverage rating podcast, right? But I think…

Amber: What is your rating?

Chris: Like me, just the human, I’m a thumbs down on it. But as a recommendation to other people, if you like mead or are mead agnostic, I would give it a thumbs up, a single thumbs up.

Amber: Okay, I’m forcing you to pick because whenever we put ratings on our brand new website, which is gonna allow us to put ratings, you gotta choose.

Chris: That, then that means this, thumbs in the middle, yeah..

Amber: You’re in the middle.

Steve : I like it. It’s a easy … It feels like a, a honey wine to me. It’s a easy drink. It doesn’t seem to offend. It’s only 5% alcohol, so I feel like I can do this and keep going on working with my day. So I’ll give it, I’ll give it one thumb in the middle and one thumb up. Now, I don’t know.

Amber: Higher than Chris.

Steve : Yeah.

Chris: So in our system, that’d be like a .5 something.

Amber: No, it’s all whole numbers because let’s be easy for the developers.

Steve : There you…

Chris: I think…

Steve : …go.

Chris: We should have a rating system so complicated we need ChatGPT’s help to figure it out.

Steve : That’s right. It’s gonna be algorithmic, different weights.

Amber: The extra fun is when I probably make Paola, I’m sure she’ll hate this task, go back through all the old episodes and figure out what numbers to give.

Chris: Oh. Oh, no.

Amber: Or we’ll be-

Chris: That’s actually probably a perfect task for AI.

Amber: Yeah, actually, ’cause you could take the transcript, right?

Chris: The transcript can tell you.

Steve : Yeah.

Amber: It could be like, “Here’s the ratings for all these people.”

Chris: Because we actually transcribe our stuff properly, right?

Steve : There you go.

Amber: Accessibility helps make lots of things better.

Converge Accesibility’s Hefty June 2026 Legal Update

Chris: Perfect segue into our warm-up topic today, which I came across an article. Converge Accessibility, which provides a lot of legal updates around court cases and laws put out a June 2026 legal update.

I think we might have also shared this in Focus State recently. But this, the June update in particular was longer and had a lot more going on, and I thought we could just talk through some of the highlights or things we found interesting. I will start. Th- they reference one particular case, which I’m going to…

I’m probably gonna butcher this name, but Patrykus v. Dormco Corp was a case, and this Dormco Corp. basically got sued for allegedly having an inaccessible website that someone tried to use. And the plaintiff didn’t reference much in their initial complaint, but Dormco Corp. had a history of audits, of monitoring, of regularly scanning their website. They had this whole paper trail that they were able to, through counsel, submit and show and prove, “Hey, we’re measuring and managing accessibility over time.”

And that pushed all the pressure back on the plaintiff and their lawyer to prove that they actually encountered a barrier, and they could not do it, and so the case was dismissed as moot. I’m highlighting this just for our audience because I think it is, you know, one in a sea of cases where the best defense is a good offense, right? It’s getting in there, scanning, auditing, and attacking accessibility and solving issues at their source and documenting your progress. That one really stuck out to me, and I thought it was cool that Converge highlighted it.

Amber: Yeah. So one of the ones that I think was interesting is Herrera v. Victoria’s Secret, this is a case that was about a clickwrap agreement. So I don’t know if it’s specifically about you know, I agree to cookies or I’ve read and agree to the privacy policy, but something like that.

There was some sort of, opt-in or legal-esque agreement, and a blind plaintiff couldn’t read it asked their sighted sister… I’m gonna say her sighted sister. I don’t actually know. But, Oh yeah, her. So the plaintiff was a woman.

Asked her sighted sister to close it for her so she could use the website. So the sighted sister then clicked it, and then they ended up saying, “Hey, this doesn’t apply to me because I never read the terms, and therefore, I never agreed to them.” What the court found, the California court, was that, nope, your sister acted as your agent when she clicked secure sign-in for you. And so therefore, it did legally apply. And they sent the case back to… And they also were like, “Nope, can’t be in California. You gotta be back in Ohio,” which is, I guess, maybe where they live.

Chris: And I think it was arbitration…

Amber: So…

Chris: …too, right? Like they had agreed to an arbitration clause or something like that, or am I confusing that with a different case?

Amber: They sent it back to Ohio, and then the Ohio court said that you have to do individual arbitration, and there’s not gonna be any class action on this. It only is you. So I do think that’s interesting. You know, maybe if we have any blind listeners, you should be aware of this. If you ask somebody to click something, then you are basically giving them the right to give you an agreement, is what this sounds like. So before you ask them to just close the thing that’s blocking you from using the website or agree to check the box to agree to the terms, you might wanna make them read you the terms make sure you actually agree to them, or make sure that they’re at least doing it on your behalf and summarizing it, not just clicking the thing.

Steve : Yeah, I thought one that stood out a little bit too was the California ABV 649, which would re-require attorney filings on website accessibility lawsuits to report them and their outcomes to the California Commission on Disability Access. So that would give like actual numbers of lawsuits and settlement rates and how cases are resolved. And which actually kind of makes this whole litigation thing, you know, go full circle so we can actually get real data on how these things turn out and if, you know, if it actually is a net benefit to the user that was put out from something being inaccessible.

So I think that’s an interesting one to actually get you know, hard data on the outcomes of these lawsuits.

Amber: Yeah.

Chris: Yeah. Yeah, totally. And before we kind of preview our main topic, the, the final one that I wanted to mention is the Online Accessibility Act, which is a law that’s kind of been poking its head up every once in a while in the US, and it had sponsors before, now it doesn’t. Well, it’s been… Basically, the entire text of that law has been copied and pasted, and now it has two new sponsors, bipartisan sponsors.

I don’t have the names in front of me right now. But that is back in circulation and something that we’ll continue to keep an eye on. But by my reading of, what this law would do in the summary on Converge, it kind of sounds like it would, and I’m painting this in very broad strokes, but it sounds like it would be something like the EAA but for the US. But more focused on not just e-commerce, right? Like much more broad. So that’s kind of interesting and something for everybody to keep an eye on.

Amber: It’s an amendment to the Americans with Disabilities Act to specifically say that it covers online accessibility for-profit entities.

Steve : Right, because WCAG isn’t necessarily like the compliance standard or benchmark right now, right?

Amber: Yeah, right now the only thing we have that says WCAG under the ADA is under Title II, which only applies to government.

And so if this passes at the federal level, then it would apply to private entities as well, which would be, as I said, for pro- businesses, but also nonprofits. So any private non-government website within whatever standards that the ADA requires.

So very interesting. They have a whole roundup of a bunch of different bills at the state level as well, so it’s worth visiting our show notes and grabbing that link and going to look and see what might be happening in your state there’s a lot of states here that are doing different things.

Wait, we’ve never talked about this before?

Amber: So was thinking about what are we gonna get together and chat about on this Friday, and I’ll admit we’ve been, like, buried working on a whole bunch of things, and I was sick earlier in the week. And so I was just like, “Okay, wait. What have we not talked about?” And did you guys know that we have never actually talked about how to manually test for accessibility? Does that shock you?

Chris: Today I learned.

Amber: I know. I was like, “Wait a minute. We’ve never…” And this is a question that always comes up, and I know we’ve done some meetups that demo that. We have a lot of great content. We have a really helpful documentation article in Accessibility Checker documentation we’ll include in the show notes that talk about how to do manual testing. But that’s what I wanna talk about. Should we take a quick commercial break first and then come back and chat about you actually do this on your website?

Chris: Let’s do it.

Brought to you by Accessibility Checker

Steve Jones: This episode of Accessibility Craft is sponsored by Equalize Digital Accessibility Checker, the WordPress plugin that helps you find accessibility problems before you hit publish. Thousands of businesses, nonprofits, universities, and government agencies around the world trust Accessibility Checker to help their teams find, fix, and prevent accessibility problems on an ongoing basis.

New to accessibility? Equalize Digital Accessibility Checker is here to teach you every step of the way, whether you’re a content creator or a developer, our detailed documentation guides you through fixing accessibility issues. Never lose track of accessibility again with real time scans each time you save, powerful reports inside the WordPress dashboard, and a front end view to help you track down hard to find issues.

Scan unlimited posts and pages with Accessibility Checker Free. Upgrade to Accessibility Checker Pro to scan your website in bulk, whether it has 10 pages or 10,000. Download Accessibility Checker today at EqualizeDigital.com/Accessibility-Checker. Use coupon code AccessibilityCraft to save 10% on any plan.

Amber: All right, we’re back. Hey Steve, why do people actually have to manually test when they have Accessibility Checker installed?

Why Automated Tests Can’t Catch Everything

Chris: Yeah, you built the the perfect software product, man. Surely it just finds everything.

Steve : Yeah, well, you know, unfortunately, automated testing can only go so far, right? Automated tests typically only cover thirty to fifty percent of WCAG issues because many require a human to review the context of what’s being added or what doesn’t exist on the page. So, in the plugin we differentiate these things, right?

And we will use automation to surface things that require manual review. The terminology that we currently use is, you know, we have problems, which are accessibility problems that we define with a good amount of certainty that are actually problems.

And then we have issues that are needs review, which we can say, “Hey, this could potentially be an issue. We’re not a hundred percent sure. We need you to review it as a human and see if it actually is an accessibility issue.” If it is, it needs to be remediated. If it’s not, you can actually dismiss it with a reason that it’s not an accessibility issue.

In the plugin, you can even put a descriptor in there you to add more context to why you don’t think it is an accessibility issue, so that if it comes up again later, that you have a little bit of history and a little bit of context to why you dismissed that. So there’s many examples of these things that require human review.

It’s alt text descriptions, right? If it has alt text, does it actually describe the image correctly

Amber: Or the purpose of the…

Steve : Or the purpose.

Amber: ( Cross talk ) … linked.

Steve : Yeah. Yeah.

I think it will check like if it’s like numbers only, things like that. I’d have to go through the actual rule and look at all the things that we check. But it is a whole, it’s a whole list of, you know, low quality alt texts that we check against.

We’re always adding more, right?

Amber: But I think that’s the perfect example, right? So if you have a alt set on an image is linked and it’s not super long and it doesn’t include like weird words or strings of numbers, Accessibility Checker might not say anything to you about it.

Steve : Yeah.

Amber: It is entirely possible that, I mean, we just saw this on a site we were remediating.

It didn’t really flag very many links that were missing an accessible name. but what happened was all the linked images for blog posts were just using the alternative text from the image and there wasn’t anything saying, “Hey, this is the link that goes to the title of the post.” If the alternative text for the image, the content creator was doing a great job, they were describing the image, which makes sense in the context of the post where you want to describe the image, but does not make sense as the name. Like that thing, that is definitely hard for I think an automated tool to find.

And so I think that’s a really good example of you just got to check that and make sure, well, is or not?

Chris: And any automated tool that tries to claim 100% coverage, they’re gonna have to have so much needs review stuff highlighted on the page that they basically might as well just put a big box around your entire website and say, “Needs review,” which defeats the purpose, ’cause it’s just a fire hose of useless information. Which we have literally used checkers that are from third parties who are like, “No, we want you to use our systems while you’re auditing or remediating” or whatever, where the information, a lot of it is next to useless because these teams who don’t understand signal-to-noise ratios and accessibility remediation just try to, you know, slam you with as much as possible and call it useful, which is not actually useful.

Unless you’re extremely experienced, like Amber and Steve and our team, at, parsing through all of that. I think on that note, I am curious about something. So we’ve established that we need to do manual accessibility testing. Does that fundamentally mean that if I’m in charge of a website, I should go through every URL on my website and listen to it on a screen reader and try to use it with a keyboard? Is that the level of work that someone has to do?

Amber: I mean, eventually maybe, but I don’t want anyone listening to this to think you have to do that in order to fix things. I would say the first best start is to test everything on your homepage, and you do it in this order. You do automated scans first, find those obvious things, then you do manual.

But I would do top to bottom, ’cause that’s gonna get your navigation menu. It’s gonna get you know, like maybe some components, if you’ve got tabs or sliders or something on your homepage, ’cause a lot of people put kind of fancier elements on their homepage. And, I would do that first, and then I would fix a bunch of things instead of trying to audit, you know, 500 pages on your website. Because how useful is it to have a list of, here’s the thousands and thousands of issues I found across 500 pages.

Steve : Yeah.

Amber: Yeah.

Steve : It’s a lot easier as you go along. I think I would jump back a little bit to when we were talking about images and like decorative images or images that actually need alt text. And the reason why I’m kind of jumping back and talking about that a little bit too is that in Accessibility Checker, we’ve talked about adding a feature for you to actually tag images as decorative, meaning that there is an actual data point stored in the database on the image that it has been marked as decorative.

But instead of adding that into Accessibility Checker, we went a step further, and we committed it to WordPress Core. So now in, in Gutenberg, in WordPress Core there’s a toggle with the new version that probably might be out by the time this episode goes out, but the betas are out now. WordPress 7.1, there will be a checkbox to check it as decorative, which will tell Accessibility Checker that it’s definitively been marked as decorative.

Now, you can still mark it even if it isn’t decorative, but at least allows the software to know that, hey, somebody has said this is decorative. And it’s just a step further that we’re trying to go on the automated side to assist the manual review side.

If You Have 15 Minutes to Manually Test, Here’s What We’d Do

Amber: Yeah. so it’d be interesting, let’s say someone only has 15 minutes, and of course I have opinions on this, but I’m gonna let you guys go first. What would you say are the manual tests that they should perform, and maybe describe how they would do that? Don’t know who wants to go first, but like what would you say if someone only has 15 minutes and they wanna test their homepage manually right now.

Steve : Yeah.

Amber: Listening to this episode.

Chris: I can describe what I do in our sales accessibility evaluation process, because that is intended to be a very quick check, but that actually tries to get to the root of what is the overall accessibility health of this website and are there major barriers?

I am doing keyboard testing. So I’m going through the website with a keyboard using tab, arrow, enter, space to try to get through the interface, and I will make note of any places where I get trapped. So if a modal comes up, can I– do I stay in the modal? Do I exit the modal while the modal’s still blocking my view? That sort of thing.

And I will try to go through the page. You mentioned top to bottom for the homepage. I will start in the primary navigation menu, starting with checking for skip links, and I will work through, and I will see if I work through the entire interface in a top to bottom, left to right order.

Amber: How do you do that?

Chris: Time I reach the bottom of the page, I will then scroll back up and I will see did my keyboard order skip anything, right? Did it skip an entire section, an entire button, an entire form because maybe something wasn’t marked as focusable? And that will be my first test that I do.

Amber: If somebody doesn’t know how to manual test, how do you move from the top to the… literally what key are you pressing on your keyboard?

Chris: Tab.

Amber: Okay. And if you wanna go backwards? Okay, so you hit tab and it does what?

Chris: Shift tab if I need to go backwards.

Amber: Okay.

Chris: If I end up in some side of– some sort of dropdown or some kind of unique interface component, like a slider maybe or a dropdown menu, I will try to use the arrow keys if my tab doesn’t get me through. And that is generally what I’ll do to see if I can get through the interface and use or at least focus on most of the elements.

The other thing that I will do is zoom my browser in to two hundred percent and go through the page top to bottom again. What’s getting cut off? Do I have a sideways scroll that shouldn’t be there? Am I losing entire bits of content, right? Maybe I’ll have a reference version and then a two hundred percent version side by side, and then I will do it again at around four hundred percent. And I will do an entire top to bot-bottom scroll again. Again, can I still see all the information? Can I still get through it? And if I’m spending extra time, I might, at those two zoom levels, also try to use the keyboard again in the same way I described previously to see if any interface components are being cut off.

That is not complete auditing, but if I had fifteen minutes, those are the specific manual components that I do without mentioning any automated components, which I’m not allowed to talk about

All right. So, so you talked about you use the tab key to move forward and shift tab to move backwards, and sometimes you’ll check if arrow keys work, like if you’re in a nav menu. Steve, what other keys do you have to press when you’re doing manual testing?

Steve : You have to press the space. And check boxes, you can u- use the space bar key as well. You need to use the enter key when you’re dealing with buttons that are interactive components, or if you’re running a screen reader, like you can use control, option, space bar, which the voiceover application will tell you to do that when you enter one of those interactive components like that.

I mean, sometimes you gotta use arrow keys when you get to tabs, right? If tabs are set up properly, you should get to the first tab, and then you can use your left and right arrow keys to go between the tabs. Yep. Yep. ‘Cause it’s, a lot of times what you’ll see is that the tab key will work for all of those tabs, but that’s not necessarily the correct structure for it

Amber: Yeah. Yeah, so I mean, I think what I always tell people if you only have 15 minutes is pretend you don’t have a mouse.

Steve : Yep.

Amber: And then use those keys that Chris and Steve mentioned, see if you can do everything that you want someone to do on your website. So let’s say it’s an e-commerce website. That means being able to search for a product, your search results to find the right color and size or whatever.

Product page, select the right size and color or whatever variation it is, add it to a cart, get to the checkout and complete a checkout, all those things, and never use a mouse ever.

Steve : Yep. And I’ll add onto that. If there’s a login, make sure your login experience works. And Chris touched on this a little bit, too. Navigation. Make sure the navigation works with a keyboard. You would be surprised how often we find out that you cannot get to fly out menus on navigations with a keyboard.

Chris: Yeah.

Amber: And this is why…

Chris: Super common.

Amber: What Chris mentioned with the zoom, like you have to zoom in and keyboard test again. Because a lot of times you might think, “Okay, well the desktop menu is fine and I can open and tab through all the drop-downs on desktop.” I think we’ve seen a lot where if a low vision user is zoomed in, they get the mobile version of the website on their desktop computer, then they get into the hamburger menu? And a lot of times you can’t. So you gotta check that zoom like Chris mentioned.

Don’t Forget the Screen Readers

Chris: So Amber, you’ve you’ve taught literal courses on this. I think you’re the most appropriate person to answer this, what advice do you have for getting started with screen reader testing?

Amber: Yeah. I will say mentioned this in our accessibility check-checker documentation article about manually testing. You have not completed manual testing if you have not turned on a screen reader. There are some really interesting things that might look like they work fine when you’re keyboard testing, so accordions, tabs. We see this a lot where they do function with a keyboard, they can open and close, but they’re missing the ARIA attributes that might tell someone, you know, is this expanded or collapsed?

Or it might be a div that somebody gave a tab index to and created the JavaScript so it could be opened and closed, but it doesn’t ever say that it’s a button. So when they get to it, they’ll just hear div and the name of the text, and they won’t even know that it’s a functional component that they could use a space bar and return key on if they wanted to. so, so I will say…

Chris: My… wait. My personal favorite that I feel like we have to call out is noisy sliders that autoplay.

Amber: Yes.

Chris: Announce what they’re displaying even if the user is no longer focused on it or…

Steve : Steve Yep.

Chris: … using it.

Steve : But at least if they’re doing it, it’s ambiguous, right? It’s just like…

Amber: Yeah, “Slide.”

Steve : “Slide.”

Amber: We’ve had users or I’ve done testing where it’s like you’re somewhere else on the page and you just hear, “Slide. Slide.”

Steve : “Slide two.”

Amber: Slide two. Or if they’re really noisy and it’s tracking changes on the whole container, it’ll read the image alt and the heading and the paragraph and the button, and it’s super distracting if you’re somewhere else.

And then we’ve… We’re just like, “Oh my gosh, how do I find this, the pause button and turn this thing off?” Now, that is very screen reader specific. I have noticed in my own testing and with users that VoiceOver on Mac announce slides a lot more, than NVDA or JAWS will. So sometimes if you’re looking at a website with just NVDA, you might not realize that all three carousels on your homepage are saying, “Slide. Slide.”

Steve : And I’ll add a, like a little sanity kind of check on using Voiceover or you know, NVDA for the first time is learn where your mute key is. Like for a visual user and somebody that’s always been a mouse user and a visual user, the first time hearing that readout can be like, oh, it’s so much. I like… And visual users don’t interpret that nearly as fast as people that use it all the time, and especially blind people who have adapted their hearing to be heightened, right?

So that mute key control helps me a lot. Gets me out of trouble.

Amber: So we have two courses, which Chris mentioned, and I’ll be totally honest. What advice do I have for getting started with screen reader testing? You should take one of our courses.

Steve : Yeah.

Amber: Whatever. This is our podcast. We can pitch if we want, but that, that is the reality. But I will say the very first lesson we have is how to turn it on and how to turn it off, and I teach that first, and then I’m like, “Okay, now go try turn it on and then turn it off” right away because I’ve seen people’s face the first time they turn it on, and so I would say, you know, if you really wanna get into manual testing and you really wanna get into screen reader testing and feel more comfortable with it we have two courses.

You pick the one that matches your machine, or you can get both. And I walk you through all the keyboard shortcuts, and then also we have, I think it’s 15 different components. So it’s like here’s a carousel, here’s accordions, here’s a navigation menu, here’s a form, here’s a tab panel component.

All these different examples, and I show bad and good, and a lot of them, the bad examples were coded so that if you were just keyboard testing, you would think they worked fine. So we worked really hard to create examples that if you did not turn on a screen reader, you wouldn’t know there were problems. And so I walk you through those and explain them and then show the good example and all that kind of stuff. You know, you just have to kinda start doing it. It is definitely scary in the beginning, but that is key to manual testing and can’t be skipped.

Common Accessibility Problems That Only Manual Testing Idetifies

Steve : So what are the most common accessibility problems discovered during manual testing that automated tools generally miss?

Amber: Well, so we talked a little bit about the slides that make issues or things with s- with screen reader ARIA. I think a lot of times automated tools can’t tell. I’ve thought a lot about this, and I don’t know if you have thoughts, Steve, on would we ever be able to create a rule in Accessibility Checker that’s “This might be a button”? Right? If we see a div or a span with just text inside it, like you’re like, “Oh, maybe they’re trying to make that into a close button for a modal.”

Steve : Yeah.

Amber: And I do think maybe in the future we’ll be able to add more rules like that, but I think there, it is really hard for any automated tool, not just us, like all of the browser extensions, like they frequently will miss div that has been made functional with JavaScript.

It’s hard for an automated tool to find that, and unless you’re using your keyboard.

Steve : Yeah. I think there’s a cross-section here too between the manual and the a- automatic scans and stuff. And we talked about it a little bit before with, the low quality alt text and, you know, click here and read more are ambiguous. And we actually will use the automated tool to assist in the manual test of that, where if you write click here or learn more or read more on a button, the tool will pop up and say, “Hey, this needs your review.”

That pops up, like I know that the click here buttons are huge, and re- remediations w- yeah, learn more all the time. And there’s a little bit of a pattern to make those accessible, right? It’s, you know, it’s learn more about post title, right? In, in, in some cases, even the post title is not always the best solution there.

But we can we can alert a, a manual tester, tester to say, “Hey, this is a problem.” And I see that’s one of the, the biggest ones I see is the ambiguous text.

Amber: Yeah. I mean, I would say in general another thing that I always tell people needs a lot of manual testing is any time you’re doing search and filtering with Ajax. Any time you’re dynamically changing content on a page.

Opening modals or nav menus is another one we mentioned, the mobile ones with the hamburgers. You always wanna make sure you manually test those components with both a keyboard and a screen reader.

Steve : For order and focus trap, we see that a lot with modals. Yeah.

Amber: Yeah, I think like tab order is difficult for an automated tool to catch. And then when it’s dynamic changes like that too, sometimes depending on how the website is built, like the modal content might exist in the footer and could be checked, but other times if it’s literally only being loaded on click, there’s no way an automated tool, unless you wrote like some sort of script to say, “Okay, click this, then scan,” right?

So you might have to make sure you’re manually clicking the thing and then scanning, or you might have to like manually click and then manually test to, in order to actually get the contents for that dynamic component.

Steve : Yeah. So we see color, using color alone to convey meaning a lot.

Amber: Yeah, that’s hard for an automated check too.

Steve : That definitely requires a manual.

Amber: Do you have an example of that, Steve? Like something where…

Steve : Yeah…

Amber: Might have had that failure on a website?

Steve : Yeah, a map key, calendar, you know, flags, little flags on calendars that…

Amber: It’s oh, if it’s in category A, it’s blue. If it’s in category B, it’s green.

Steve : Right, right. And, a way to remedy that is, you know, i- interject a icon, a shape , so it’s not color alone.

Amber: Yeah, like it could be a blue square and a green triangle or…

Steve : Right. Right.

Amber: …icon that matches your category name.

Steve : I mean, you could even get in trouble with that with links, right?

Amber: The state? Not having an underline at all?

Steve : Yep.

Amber: I wonder, I’ve wondered about this because we have an automated fix in Accessibility Checker that will allow you to add link underlines. I wonder if we could create a rule that is “You disabled underlines on your body,” like in your CSS.

Steve : Yeah.

Amber: All right, gotta go open a GitHub issue.

Chris: One that I remember coming up sometimes particularly for interface components that interrupt the user maybe or it’s a modal or something, if the user leaves that for any reason, I know we’ve encountered situations where it takes their focus back to some other unrelated area on the page.

They lose their place, which depending on the type of interface, like if they’re browsing stuff in a list and trying to move back and forth quickly and parse information that can be extremely inconvenient ’cause they have to you know, they get to item number 30 and they also want to review item 31. They open it up, they look at it, and then it takes them back to number one. They have to go to 31

30 things to get there, right? And then it’s this horrible process. I think we’ve seen that on transportation websites before and some other places in our user testing.

Amber: Yeah, that reminded me of smooth scroll. If your website uses smooth scroll, that’s another thing you really need to manually check. And I think I know for sure we have a YouTube video. I’ll throw a link in the show notes, but we might also have a blog post that talks about and shows what… How smooth scroll can break. But it will visually look like it’s moved focus and it scrolls the page down, but it leaves the focus exactly where it was , which could break your skip link, but it could also break any jump links that you have anywhere on your site. So if you have that, that definitely requires manual testing.

Steve : And I think an easy one that we see all the time that actually has a automated and a manual component to it is heading hierarchy, right? So Accessibility Checker will flag if you get your heading hierarchy wrong.

Amber: Well, just if you skip a level, right?

Steve : Yeah, if you skip a level, but it doesn’t check if the visual hierarchy is correct.

Amber: Well, I think the other thing too is unless we were to try and get AI to read the text of all the headings and be like, “Did they do this right?”

I guess possible it might not flag any problems if let’s say every heading on your page was an H2.

Steve : Yeah. Yeah.

Amber: Because all H2s can follow H2s. You haven’t skipped… So it wouldn’t flag any heading issues. But then if you’re like, “Okay, but this H2 is blog posts, and then we have all the blog post titles in H2s.” Well, no…

Steve : Yeah.

Amber: …not actually correct.

Heading order, is something I think that will always have to be some sort of manual testing, which is just look at your headings in outline format and make sure they’re nested appropriately. Even if the automated tool isn’t flagging that anything has been skipped, you should still be looking at your headings and making sure that, you know, if something is a s- a child item of a heading above it, it shouldn’t be the exact same as its parent heading.

Whoever You Are, You’ve Got This

Chris: Well, I think we’ve given people something to think about. And here’s the thing. I know we just spent the better part of 10 minutes rattling off all kinds of problems that you could check for, and that could maybe produce a sense of overwhelm in some listeners. But here’s the thing to remember. Remember what Amber said step one was. Fire up your homepage, start hitting tab, zooming in and zooming out.

Amber: I think you said that was step one.

Chris: Well…

Amber: I’ll take the credit though. I mean, I don’t think we wanted this episode to be overwhelming. Is a very ordered that outlines all of this. I think piece on manual testing is just starting to do it and trying to build it into your process. I mean, I actually had a conversation this week with one of our customers because they have a really old website that we’ve been helping them remediate for a couple of years, but it’s, you know, a certain amount, and it has this block library on it, and I haven’t looked at all the blocks in the block library. If there’s one they haven’t used before who knows? We didn’t put it on there, right? It was there when it came to us.

And they were like, “Hey, we rebuilt this one page and it’s got these flip cards. Are they good?” And they’re like the things where it just shows like a heading, and when you hover your mouse, it flips over and it shows more content. And I just said to her, I was like, “Okay, let’s talk a minute about before you build something new with a block that’s never been used on any other page of your website, what should this look like?” And that’s what I think this is just don’t use your mouse. Can you still get the information? Because with those, they didn’t flip over when you tabbed.

They don’t receive focus. There’s no way for someone without a mouse to see or hear that content. And that’s where manual testing comes in. It helps give you a broader perspective, and it doesn’t have to be scary.

Steve : Yeah, and I’ll just lean in a little bit as the technical guy making the software to say that, you know, generating a report or even passing tests is not really the main goal of accessibility. It’s, you know, it’s about whether real people can do real tasks that they need to do on the web, and I think manual testing is kind of what bridges that gap.

Amber: Well, thanks guys. Thanks to everybody who’s listening and hitting the subscribe button and the review button. will see you in two weeks. Cheers.

Chris: Bye everybody!

Steve : Cheers.