Why accessible content is AI-ready content

[00:00:21.690] Joel Goodman: 

Hello, everyone! Welcome to our webinar today. Megan and I are excited to be back talking to y'all about accessibility.

Up front, wanted to, let you know about this AI visibility report that we offer for free. If you would like to, you can scan that QR code there. We'll also have a link towards the end, and we'll send out these videos, or this video for this presentation later on.

They'll give you a snapshot of where your website is right now in terms of how visible it is to AI search, as well as analysis of your content, what questions AI can actually answer, what it can't. And then we'll give you some actionable steps to improve.

what your content's doing in AI Search. So, feel free to scan that QR code, and like I said, we will follow up with you later. My name is Joel Goodman, I'm the Vice President of Growth Strategy here at Squiz.

With me, as, I will say always, because Megan, you and I do these all the time together, is Megan Andrews, our Director of Client Partnerships.

As I said, recordings will be available afterwards, watch your emails for that link. There's also a Q&A open, so if you have questions while we're running, please add those to the Q&A box. We'll be looking at them as we go through this, hopefully answering them.

We might, if there's enough questions, we'll answer them as we go through the content today, but we also can follow up at the end, and we'll try to answer all of them. If we don't get to them, we'll send an email out with those answers.

Afterwards, and then if you have any technical issues, please drop your questions or comments in the chat, and someone will jump in to help you out and make sure that you're able to participate.

So… Let's dive in. Who is Squiz? If you're not familiar with us, we're an AI-powered suite of products that kind of let marketers control their entire digital experience from start to end, and so that includes Squiz Content Intelligence, which we'll talk a little bit about today, and you’ve, if you've been to other these webinars, you've probably heard about it as well, but it's a full content health solution for accessibility and AI visibility.

I also have, funnel-back search with conversational search, which is the best site search on the market, and it has an entire AI conversational layer on top of it that helps.

And then we have a full digital experience platform, which covers your content management, personalization, A-B testing, optimization, everything you need to do, or all the tools that you need to do, all the things you're supposed to be doing with your website as a marketer. Learn more at that QR code, or you can go to squiz.net and cruise around the website that we recently relaunched, and it's beautiful, and I did a lot of work on it, so I hope you like it.

Here's a little bit of what we are going through today. So we're going to talk about the AI search visibility problem, why content isn't showing up, how people are making decisions and doing more research through AI search. You've heard about it, you're aware of it. We'll kind of talk about that a little bit.

But we really want to dig into why accessibility go hand-in-hand, and they really are the same challenge, so you kind of get, you know, double bang for your buck.

We'll look at what your content needs to make it accessible and AI-ready, give you a fixed list so that you know what to do to improve that, and then we'll talk about scaling this up in a way that's manageable for those of you that have thousands and thousands of pages on your website. And I know that a ton of you have that problem happening.

So, Megan, let's talk about the AI reckoning and what we've been hearing. You've been doing a lot of this work over the last couple of years, and I want to… I just want to hear which of these, phrases are standing out to you more than others.

[00:04:06.220] Megan Andrews: 

Oh, totally. Well, maybe the comforting part of this is that no matter what kind of organization or company you are right now, you're all facing this challenge. AI is not only here, but as you know, changing so quickly, and even accelerating in its pace. So, everyone has this sense of what are we supposed to be doing?

How is it different from what we've been doing, right? So this, you know, the top right I'll look at, we have no visibility of how or if our content is being cited in AI tools. You know, what are the best practices we need to follow now to make sure of that? That's really top of mind.

And I know you've… you've been hearing about this too, and you study this as well.

[00:04:48.640] Joel Goodman: 

I do nerd out and read a lot of things around this. Like, these are actual quotes that we're getting from other customers that we have of our products.

You know, a lot of people know that this is a problem, but they don't know how to actually solve it. You know, how do we get our content to show up in those answers that an AI search engine is giving when the people are out there asking those questions? And what I… what I keep coming back to is.

It doesn't really matter what questions that people are actually asking in ChatGPT or in Perplexity or in Google. It matters whether or not the AI search platform can read your content in the first place, because if it can't, you're never going to be in those answers.

And honestly, we’ve talked about this in previous webinars, but when that question is asked by a specific person, the AI doesn't answer that question. It generates 50 or 60 or sometimes 100 additional questions underneath the surface and goes out and does its own research before coming back and answering that question.

So if you're not showing up for as many of those answers as possible, you're probably not going to be in that answer in the first place. And so that's really where we need to start. The content that's on your website needs to be in a position where it's even visible at all to AI that's likely to show up in any sort of AI research.

And just ignore the questions that people are actually asking on these AI search platforms, because it doesn't really matter if you can't show up in the first place. So we know that everyone has an AI disability problem, right?

[00:06:19.040] Megan Andrews: 

Yeah, it's very true, and I… this is a great problem statement that could make you panic a little bit, but you don't need to, because actually, as we'll go through, sure, everyone has this challenge, but there are so many principles you can follow that are gonna really ensure that you're gonna show up, that your content's ready to be read, and it helps you with all of your accessibility. So, these go hand in hand.

And that's the important part, I think.When it comes to… the easiest way to think about this role sometimes, I'm like, okay, well, we're humans, and how do humans access information? Well, turns out there are people who might be sight impaired or need accessible features on a website. We've been used to that, right?

Fortunately, AI are large language models, they are machines, and so they're gonna read content just like a screen reader. So those same best practices are just gonna come right back up, just gonna sort of double down on the idea that, you know, everything's gotta have that metadata, those descriptors, everything that a screen reader would need anyway.

[00:07:25.010] Joel Goodman: 

Yeah, it's this idea that, you know, accessibility isn't even just for people with, you know, with challenges to reading that content. It's really about anyone being able to consume your content. And as much as I hate to think of an AI bot as a person or as a someone, like, it's another entity that's having to come and figure out what's on your pages.

And so the same practices, as you said, that allow screen readers to understand the content on the page, that even allow SEO crawlers and bots to understand the content on your page and the structure on your page.

All that work you've done for accessibility to do… to fix those things is going to benefit you on the AI search side, because they read it in the same way, and they understand it in… well, they at least start to parse it in the same way. And so, starting at that beginning and making sure the structural side of it is complete is just super, super important.

So to kind of connect these dots a little bit further between how accessibility and AI works, like, every organization has accessibility targets, especially those with giant websites, and the thing is that the accessibility work that you're doing now is probably even more valuable, because it's affecting so many other parts of that website's reach, not just the part for those people that need accessibility affordances for them to be able to understand your content.

So everything that you've been doing to work towards that compliance, whether it's the legal versions, I mean, we know that the, you know, we know that the U.S. ADA Section 2, or Title II revisions have been pushed out to 2027, but we were all gearing up for April 2026.

A lot of higher education institutions were putting extra work into making sure their accessibility guidelines were tightened up, and that their websites were super accessible.

All that work is good to do. One, I'd say just from the side of being fellow humans that need to, you know, that need to provide that access to other humans. I think on the flip side of that, it's just good for business, because if you're not doing accessibility work, there's a whole swath of people that are not able to access your content, and they're not going to be potential customers, or potential, potential students, or potential partners, or whatever that is.

And then on the kind of third layer now, it's doing that work also helps you with that AI search visibility. And so it wasn't a waste, all the compliant work isn't a waste, nothing's changed on that front, and in fact, it gives you a little bit of a head start in being AI-ready, and increases your chances of being cited by LLMs.

I really like this chart, becauseit shows you… I was… I was joking, I was joking as we were putting this together, it's… it's that meme with the two hands clasped and the big biceps, you know? It's like, yeah, teamwork sort of thing. That's really what this is, right? You can see what screen readers need.

You can see what the fix is, and you can see how that affects it on the AI readiness side of it. And so that part of the middle is really where that works. So, Megan, I'm wondering, like, what have you seen as you've gone through some of the AI readiness audits that you've been doing for Squiz customers? and Are there any here that you think you're, I guess, are more common than not? Like, or any that kind of strike a chord with you and make you more interested in it?

[00:10:44.900] Megan Andrews: 

Yeah, there is one that always jumps out. Everyone's probably aware of those pesky PDFs that are sitting across your site. They're there, they happen! People love a downloadable PDF.

not great for accessibility, and we've all been aware of that as well. So, turns out we should think about that even more, getting that content into structured HTML on your pages. That's great for LLMs and accessibility. The other one that jumps out is this clear heading hierarchy, right? So, you'd be surprised, and I've been on a content team, right, where, look.

the H1 looks this way, the H2 looks this way. H3 looks, like, aesthetically a little bit more like what I want, so I'm gonna throw that in there in the wrong place. It happens, but…

truly, if you want your information to be organized the right way, go work on the styling on those, those elements, but don't, flip those orders. That's just not gonna help out with reading it, right?

[00:11:44.130] Joel Goodman: 

Yeah, and I'm sure our friends on the higher ed side of this have seen that. I know when I worked, when I worked at the two institutions I worked at, we saw this all the time, where someone will be editing content, like, oh, it just looks a little bit better, I just want the text to be smaller, and so the easy way to do that is just to change a heading, but if you skip those levels, it's bad for SEO, it's bad for accessibility, it's bad for, it's bad for AI visibility, because it can't understand what

What the hierarchy order is.

So a lot of these are kind of under-the-hood sort of things. I mean, talk about semantic HTML, that's just having modern HTML that's communicating what the different regions or landmarks within your page are. It's not something a, you know, a normal content person might be able to do, but it is something that should be built into your CMS or your DXP.

You know, the alt text issues, I think everyone's pretty on top of that. Everyone by this point, knows why that's important. I think the one for content folks really comes down to this kind of specific and explicit language, right?

It's making sure that you're not using flowery speak. In higher ed, we sometimes like to have very academic speak. In business, we like to have very marketing-speak sort of stuff, right? So how do you make sure that you're specific and clear and consistent with what you're saying so that the model understands it, the person that's there understands it, and that, you know, the… even, like, the level, the reading level is one of those things. The simplicity is always better.

Within those sorts of things.

And we know that from Google's research, like, AI agents view websites in three different ways. So they'll screenshot them, which is expensive and token-heavy, so it doesn't really happen all that often. That's more if someone's doing, like, extreme work with, say, like, Claude or one of those, you know, kind of frontier agents, and doing real work on that front. They definitely look at raw HTML, and then they look at an accessibility tree.

And the accessibility tree is this browser-native API that gives a semantic summary of that whole page. So it's the same thing that screen readers use, as we were saying.

And for an AI agent, it functions as this sort of, like, high-fidelity map of your website.

It ignores the visual styles, it ignores all of, all of the pretty stuff, and it's looking just at that structural layer. So it's really, it's really important that your accessibility tree ends up being clean, that's that kind of, like, technological side of it, so that your site's agent-ready, because otherwise it's not going to be seen, or it's not going to be parsed in the way that you want it to by those agents.

So that's the very technical side of it.

We want to run a poll, and so we want to understand where, where you are at in this journey. So let's, see if we can get some answers in here. Poll question is, when it comes to your content accessibility and AI readiness, do you even know where to start?

So we'll leave this up for a couple of minutes. I don't know, Megan, what are you seeing when you're… when you're talking to people? Do you… I feel like people generally don't know where to start, but, yeah.

[00:14:48.230] Megan Andrews: 

Yeah, people have certain inclinations and expertise around the world we've been in, right? So I think SEO and accessibility, the way we've understood it, is still… it's not worlds apart from where we are now. If anything, I think the people don't understand exactly what they… what more they can do to optimize. So a lot of it is learning, you know, we'll be talking with…

You know, customers we have have different levels of understanding around this, too. Teams have different understanding. I've… I love how much content and technical teams can work together on these principles. We even saw it with those, right? There's a lot about structure, and, you know, the tree that can be read in a certain way with all of our tags.

But, that specificity piece is really important for content teams. You know.

Large language models love specificity, so I really like to think of this as a combination of your technical and content teams working together, and, sometimes… sometimes we don't always know the best way to start with that.

Let's see what everyone said.

[00:15:58.510] Joel Goodman: 

Here we go. Alright, so we are… it's looking like, a good chunk of you havea plan in place. You're working on it, and that's awesome.

Some of you, know their issues, but don't know what to fix first. That's… that's a very common problem that we hear all the time.

Some of you don't even know how big the problem is yet. It's important to get a handle on it, I would say, and hopefully we can give you some good guidance today in this webinar on where to start. And the same with those of you that know what needs to be fixed, but you don't know how to act on it. I think we can give you some good tips. And no matter what it is, how you answer this question.

I… we just kind of want to tell, like, reassure you, like, all of this stuff works together, right? So, from an accessibility standpoint, these aren't competing priorities, they all reinforce each other.

You can see kind of the overlaps here between accessibility and SEO and the AEO, GEO stuff, the things that you know about, the things that you've been hearing all the time, this is the sort of thing that, is… is helpful, and the work that you do in one area is going to affect the other ones in a positive way.

[00:17:07.710] Megan Andrews: 

Yeah, this is, you know, don't worry about memorizing this Venn diagram, but it is… it is helpful to understand how these all interact, that they all are related. We actually were talking through this together and came up with this idea, like, you know, this could be in a Venn diagram of just understanding the layers that… that are here.

And I think that, you know, again, none of this is actually, like.

a lot of these are some of, like, the basics we've been thinking about. I think what I often look at now with content teams is that accuracy, completeness, and question matching, or question bank testing, that's a little bit more of a new world for us.

if… what types of questions are actually being asked? We've talked about this before, but users have been forced into the box of traditional keywords, and so when we sit down a lot of times, we're wondering, okay, what word do I need to come up with that'll get me what I need?

Whereas now you're seeing this real shift in, conversational natural language. So that intent is different. The, the way that large language models are interpreting the question is a little bit different. They ask themselves a question to answer. So I think that's a… that's a newer space to be aware of.

[00:18:24.960] Joel Goodman: 

Yeah, I agree. So we have one question in the Q&A that's come up, and it's very specific, asking about accordion widgets and AI readability.

all of… the answer is always going to be, it depends.

Our colleague, Pez, in the UK says this all the time, or has for years and years, but it's one of those things that it just… it just depends.

So, accordion widgets themselves are not bad. It all depends on how you have built them, or how your developer has built them.

If they are loading content in with JavaScript, and it's hiding it beyond, hiding it when not visible, literally hiding it from the page when not visible with JavaScript. 93

then that is a problem. If you're using semantic markup, if you're using the native ways to do that with an HTML page, if you're building it in a way that is accessible, then it's not going to cause a problem for an AI, for an AI bot or anything that's crawling your site. And that's really what you can think about with any of these specific things, right? It's like.

Video embeds can be inaccessible unless you do them the accessible way. Put a transcript in, put audio descriptions in, that sort of a thing.

Once you, again, look at this Venn diagram and everything we've been talking about, once you approach the build part of it in a way that is accessible and adheres to the WCAG standards.

then you're essentially making it visible to AI. So take that answer and apply it to almost anything, going forward, and you'll be set up for success.

I want to move on to the fix list, Megan. You mind talking through this a bit?

[00:19:57.450] Megan Andrews: 

Yeah, well, you know, print it out, put it somewhere you can see it. This is essentially what we just talked about, but in a checklist style. I love a checklist, so I would do something like this.

But it also helps to have a framework, right? I've been talking about this a lot with teams, where you say, hey, if we're all learning some of these elements together, at least have a common vocabulary around it and understanding of what you're doing. So, sometimes this is just something you could throw up in a meeting and say, hey, you know, teams, we're thinking in these terms now, so as you're going through your day-to-day and thinking of your best practices.

you know, come back to this and start making it your behavior around how you're creating content or looking through pages and auditing them. So, again, all of this is very much what we just talked about, but, you know, I think flagging some of these where, that vague language might come up, this is something you wouldn't always find if you're not doing question bank testing or having a large language model ask things. So that's one that… it'll test you. It'll test your marketing team, to pull that language and make it a little more specific.

I love this terminology, consistency one. So, consistency is another thing that large language models are great at finding patterns or consistent information. So if there's a theme you really want to get across, in the values or messaging of your company or your institution.

This one comes up, I was on a call yesterday with a customer about this, where they're like, hey, you know, actually, what we really want people to get from our site is that we're pretty small, and we want one-on-one interaction. We want people to reach out to us when they have questions. We don't want to actually answer everything on the site.

So, I said, okay, that's a great thing that you could… if you came up with two or three sentences that say that in a succinct way, and place them around your site, that's something a large language model could pick up on, too, and surface back to a user.

[00:21:47.950] Joel Goodman: 

Yeah, and that's… I think that's a… it's a shift in how we think, because a lot of times the… like, that approach in particular is… is sort of an old-school CRO approach, right? It's like, well, if they talk to us, they're going to get a, you know, they're going to get a better answer, they're going to feel.

More connected, and I think it's completely valid, but the reality is that most people are not starting at that point, and if you’re user, your website visitor, can't pick up on that directly without going through something, like, that AI bots don't feel, right? Like, LMs don't feel anything, they don't have emotion, they don't make decisions that way. They're not gonna say, oh, I didn't find it, I'm gonna go ask the person. They say, oh, I didn't find it, I'm gonna use someone else's content.

And so it's important to at least be clear about That intention, becausebots aren't gonna pick up on it, right? Machines can't pick up on what our intentions are unless we're telling them exactly what they are.

Yeah. I think one of the… one of the interesting on the dev and technical side, because I'm more of a technical person anyway, that… that I… I added to this list in particular is… is site speed, and I think, I've… I've… Built a lot of my reputation on site per site speed and web performance over the years.

Slow-loading sites, which is still a huge problem, in… especially in higher ed, but businesses too, we don't think about this. Slow-loading sites are inaccessible to humans.

Think about someone in a rural area, or someone that, you know, uses a phone only, and only has a data plan on their phone. If your website's slow and they can't load it, they're going somewhere else. It actually is the same thing for AI bots.

We've seen this, in research over the last few months. your website tells the… tells the AI crawler that, that it's too slow and it's loaded, they're moving on to the next one, because these companies don't want to waste money waiting for your website to load and to crawl it.

They're gonna… that's a… that's a negative trust signal. They're gonna look for a better, more positive trust signal. So it's an accessibility issue that improves what you're doing, and so… on the, on the AI side of things. So it's important to… to go through this and remember that all of the stuff that you're working toward, again, with accessibility ishelping AI search and AI visibility, with your website.

[00:24:05.640] Megan Andrews: 

We have a couple questions, Joel, which might be nice to just get into these.

[00:24:10.530] Joel Goodman: 

I would love to answer… I'd love to answer, this one about, it says, can you talk about how ARIA impacts AI readability? So ARIA labels are technical, and they get put on specific elements to, provide descriptive language to a screen reader, essentially.

One of the struggles with ARIA is that we tend to overuse them within our code, and so it's important to make sure that you are doing strong checks across how ARIA is being implemented on your website. Because if it's implemented incorrectly and it's confusing an AI reader, it's also going to be confusing a screen reader at the same time, vice versa, right?

So, ARIA itself can have a positive impact if it is implemented in the correct ways that support accessibility. Again, if you're doing it accessible, it's helping the AI visibility side of it, but if you're over-engineering your ARIA label.

and it becomes confusing to a screen reader, what's going on on the page, it's gonna be confusing to an AI bot as well.

You want to tackle one of these other ones, Megan?

[00:25:16.150] Megan Andrews: 

I think we'll kind of get to these as we, go through the next little bits of information here, because it's… Yeah, about, you know, what mistakes are organizations making with their visibility for AI, and then how frequently do we need to be looking at our content? So, this is, you know, this is a helpful process to think about this as an iterative process.

You know, auditing is never done, just like a website's never done. There's always room for continual management of those things, and the most important thing is to just audit regularly. Now, that could be different depending on your organization. You could do it once a year and just be glad you did that, but maybe quarterly, maybe monthly, depends on how much new content you're putting out. In a lot of higher ed organizations, you've got a lot of evergreen content that's not changing too much, but for those dynamic pieces, just understand the pace of your own output.

So, Auditing is first, then prioritizing, whether it's lining up with your strategic objectives of your institution or your company, understanding, wow, we really gotta focus on, you know, research or admissions, whatever those things are.

Or, you could prioritize by understanding traffic behavior on your site, and even just where the biggest problems exist. Sometimes it just matters to make big improvements across pages.

You know, where they have, a lot of errors or risks for visibility. Yeah.

And then the rest of it is just, you know, take that fixed checklist, understand those basic principles, go ahead and track what you're doing. You want to let everyone know what you're doing anyway, have a way to track it. And then, like I said, it's a repeatable thing. Embrace it, it should be happening regularly.

[00:26:58.410] Joel Goodman: 

Anyone remember this? Ugh. So, this is… this is actually… this is… this is something that we've anonymized, and this is from my previous work running an agency. When we would do accessibility audits, we would run this through something like DeQ's Axe Accessibility Auditor, and you just get back a giant CSV with all of these things to fix, right? And it's… it's descriptive.

But it's overwhelming. I mean, this is, like, this was hundreds of rows to try and fix, you know, across several pages. It's just a whole lot that you have to do. And, you know, we talk about this framework, and, like, the framework is really good, those steps that fix list, like.

It's… it's e… like, it's good to follow if you wanna, if you wanna, like, dive in and do it, you know, on your own, but, like.

You know, this works for 10, 25, maybe 50 pages if you really want to go through some pain, but it's really difficult when you're looking across an entire website. You've got10,000 pages, or, like, some of our… some of our higher ed customers in the UK are looking at hundreds and hundreds of thousands of pages across, you know, across their websites.

And that just becomes unsustainable, especially in higher ed, where, you know, well, I mean, honestly, any… any sector, like, we're all facing budget crunches, we're all facing, staffing crunches, like, all these different things.

There's so many pressures to get this stuff done, so you need sustainable ways to go through and make these fixes and be able to prioritize them, right?

[00:28:29.950] Megan Andrews: 

We did love the idea of having this visual, just to remind everyone. It can be painful, but it doesn't happen.

[00:28:39.440] Joel Goodman: 

But, like, a lot of the tools that we've seen, too, you know, there are a lot of accessibility tools out there, and a lot of people run them, and, you know, we've had accessibility tools built into FunnelBack for years and years and years, but, you know, a lot of times this is what you get. It's like, cool, you're compliant, checkmark on, you know, on all the different things.

You know, like, can everyone get your content? You know, can screen readers get in keyboards? All those sorts of things, like, and cool, if you got the green tick, you're happy, right? You're compliant.

But this isn't the stuff that these tools are saying, you know? It's not telling you what copy is vague, or, like, if there's conflicting information across multiple pages, or, you know, where your AI right is. Some of these metrics are starting to get in there, but definitely not the conflicting information side of it, getting really deep in the structural issues of where your content is at, and whether or not that's harming the visibility of your content. You know, most of those tools are assessing that structure part of it, but not many or any of them are really affecting or assessing the content that you actually have.

[00:29:42.900] Megan Andrews: 

One thing I would say about that, Joel, is that it's also really hard. A lot of teams will end up having subjective opinions managing these decisions.

[00:29:50.400] Joel Goodman: 

Is this clear?

[00:29:51.030] Megan Andrews: 

content good? And it's like, well, if so-and-so says it is, I guess it is. That's a great one. Yeah, it can be nice to have this sort of equalizer of, hey, this is a large language model.

[00:30:01.020] Joel Goodman: 

Yeah.

[00:30:01.570] Megan Andrews: 

Of our content, so there's not an opinion that needs to be formed here. This is just something that needs to be known about how machines are seeing our content.

[00:30:09.950] Joel Goodman: 

Interesting. Yeah, that's… that's a great point. Alright, let's… let's run one more, poll here. Are you using your accessibility tools to their full potential? For those of you that have a tool, if you don't, there's an option for you, but for those of you that have a tool.

Do you feel like you're using it 100%, or is it more like, I mean, I've worked with a lot of institutions over the years that have an accessibility tool in place, and they use it for, like, 5% of what it does, or 20% of what it does. They're paying a lot of money for it.

But yeah, I'd love to, I wanna see, I wanna see how people are doing.

[00:30:43.530] Megan Andrews:

I often see some super users out there where there's a little bit of a, well, you know, this guy does it for us, and he's the only one who knows the tool, and that can also be a challenge.

[00:30:55.830] Joel Goodman: 

Only one person that can run the tool and figure out what it says, and I mean, I get that. I've probably been that person in my career.

[00:31:04.080] Megan Andrews: 

Exactly. Yep. Small teams, you often end up with that, right? You know, you've… only so many people can learn so many tools.

[00:31:10.500] Joel Goodman: 

Yeah, no doubt.

Alright, let's see how y'all answer. Well, there you go. That's… that's what I thought. And this is… 44% of the respondents here said they've got tools to not gain full value from it, and that's… again, that's common. I'm not… that doesn't surprise… surprise me.

Proud of all of you that… that feel like you're getting all of that… that value out of them, because I love to extract value out of whatever it is I'm paying money for, so that's awesome.

And yeah, a couple of you out there rely on a super user, but it doesn't really scale, and some of you… a fairly good chunk, 26%, don't have a dedicated accessibility tool.

And I'm curious to know what y'all are doing. But anyway, whatever you answer here.

We've got, we've got an option. Of course, we have an option, right? This is part of what we're showing here. We recently launched, Squiz Content Intelligence, and if you have seen anything that I've posted on LinkedIn the last couple weeks, you've seen how this works.

Our accessibility auditor withinContent Intelligence looks across all of your content, and then assesses it up to WCAG 2.2 AAA standards, but it goes beyond that.

It tells you what type of fix it needs to be, it prioritizes those fixes based on how often it finds it on your site, if it can detect that it's in a, you know, a component or, you know, well, a reusable component on your website, then it can tell you, like, hey, this one fix is gonna fix a whole bunch of stuff and makes it really, really easy.

We did that on our site, actually. We found one error in a HubSpot embed form on our new website. We fixed it once, it fixed it on every single page.

That was awesome. But really, it just gives you a way to look across all of your content and figure out exactly what's wrong and prioritize what it is that you need to work on, right?

So we'll send a video along with that, you can check out… you can find me on LinkedIn, Joel Hitman, and you can see some of the videos that I posted as well.

But we tried to find a way to make these tools as usable as possible, and so if you want a demo of it, we can show you what this looks like. Request an AI visibility report, we can tell you what's happening on your site, but then we'd love to show you how the accessibility auditor and the AI readiness tool within Content Intelligence works together.

I think it's super cool. I don't know, actually using it made me way more, like, more excited than I was when I knew it was coming out, actually using it was, was, was really cool and helped us just get things done a lot faster and improve… improve our own website, which is… which is cool. You don't often get to use your own tools to improve your website, which is cool.

So, let's… let's recap, Megan.

[00:33:58.000] Megan Andrews: 

Yeah, well, we did get through a lot, hopefully not too much, right? Bite off what we can chew here, but really the bottom line is that accessible content is AI-ready content, so those good principles are going to serve you well. AI search platforms are changing a lot of where decisions are made.

A lot of initial research is happening off of your website. So if there's a lot of discovery happening there, knowing and being confident that the content that's being… not even just knowing it's getting read, but then what's being read and surfaced for users is really important. So, if your content can't even be read, you're not even going to be in the answer. You won't exist for the user's research.

So that's really important, and fortunately, accessibility, number two, accessibility and AI readiness are very similar, and that's not the same challenge. So, you know, think of these machines. What blocks a screen reader is probably blocking that AI search platform.

And then the last two, I mean, Joel, like, I think we saw that with the poll, right? Is your accessibility tool doing enough? Do you have the right tool? Can you leverage it, right?

[00:35:01.290] Joel Goodman: 

Right. And I think a lot of times when you have a tool and you feel you're not getting the most out of it, a lot of that comes down to how the tool itself is built. If it's not helping you use it to its full extent, then there's an issue.

And I think that's… That's what I really like about how Squiz Content Intelligence was designed. It's meant to take out any of that guesswork. It's meant to tell you exactly what's wrong and how to fix it, so that you can just move on, and you can scale that out to a whole team if you want. If you've got a bunch of content people that can go tackle the content ones, sweet, have them go content… tackle the content ones, right?

If you've got devs that need to make changes to your actual CMS or your templates you can send them that information, that… the actual fix, they can take a look at it and get it fixed.

That's how we operated on our own site when we fixedthe few issues that we found when we first ran it. It's really about getting everything that you actually need, and not having to overpay for tools that you're not gonna get, or, like, you know, overpay for help to figure out how to use the tool, right?

But then also being able to automate both the audit side of it, the monitoring side of it, and the fixes, so that it scales across a small team, or even a large team that's got a lot of stuff that they have to work on.

So we've got a few more questions. If you want to email us your email address, or our emails, you can also, take a look at the QR code, or follow that link, or just go to squeez.net and look for Content Intelligence, because it's right there on the page as well.

Kevin had a follow-up to our chat about ARIA labels, and he asked, or said ARIA can be checked in the accessibility tree.

Yes, it can, but… but also, I think, like, ARIA labels, there's a… there's a strategic knowledge that you have to know about how to employ them and deploy them.

Aria labels are not to be, as I said earlier, you don't just put them on everything. There are certain… there are certain semantic HTML landmark tags that have their own semantics built into them, and putting an ARIA label on top of themactually makes it less accessible, so it's better to have some sort of a tool that can check it, unless you feel extremely confident with WCAG and how it should be deployed.

If you're already doing accessibility remediation and you already know this stuff, sure, look at the accessibility tree and figure out what's wrong and make the fixes.

If you don't consider yourself an accessibility expert, it's better to rely on a tool or a person or some structure that actually does have the expertise built into it, especially with RA labels, because they can be tricky in terms of where and how to deploy them across your code.

That was a real nerdy technical question, but one that I have… one that I have, seen in the past. What other one do you want to look at, Megan? What are some of these other.

[00:37:57.130] Megan Andrews: 

Well, you know I love a content question. This one, how do you balance plain language requirements with maintaining an organization's tone of voice?

I love that. I am a huge fan of using… leveraging AI in ways that, can maybe enhance what a human's already created. So, what I really don't want to see is that all of us end up sounding the same on the internet. Just because AI recommended this, this is a great way to say this.

Well…Keep the human in the loop. Tone of voice is so important, especially if you want to stand out.

How are you going to be any different from all of your competitors if you can't have a distinct tone of voice?

So that, that is still important. Start with your own people creating things, and then check that against large language models. Don't lose… don't lose your values or yourself in the AI revolution.

[00:38:53.910] Joel Goodman: 

Totally. We had a question about, which pages to prioritize to edit for accessibility and AI first. I mean, that's what content intelligence kind of helps you figure out, but I would say, like, if you're running through sort of the framework that we presented earlier.

The best thing to do is actually look at your SEO data. Like, look at what pages are getting the most traffic and which ones are the most valuable for whatever your business goal is, right? If it's, you know, if it's higher ed and you're looking to increase enrollment, you might want to look at your program pages and make sure that they are as readable and accessible as possible.

You might want to look at your financial aid and tuition pages, that sort of thing. If you're a business and you're doing, you know, if you're a law office, right, you might want to look at your practice areas and make sure that they're very clear and that they state where you are at in the country, and how well, you know, where your expertise lies, and where your partner's expertise lies, all of those sorts of things.

So it really does come down to who your audience is, and looking at your SEO data can help you figure out what pages get the most amount of traffic, but also which pages, people are dropping off of, because perhaps the content isn't good enough for them.

Content Intelligence helps solve that problem without you having to decipher all the data in your SEO, in your Google Analytics, or whatever it is that you're taking a look at.

[00:40:18.510] Megan Andrews: 

I think we kind of got to a lot of these. I mean, the… you know, if there's… if there's one takeaway piece of advice, Joel, that just… people are gonna go on their way from this webinar, what should they be thinking about in the next week or two?

[00:40:30.900] Joel Goodman: 

You know, I think that most people are already thinking about accessibility and have been, and everyone that's on this call, obviously, is thinking about accessibility, right? And it's been, in a lot of sectors, top of mind for us for years and years and years, and more urgently in the last year as we've been working through the Title II deadline that's now been extended.

Like, that's all top of mind. Lawsuits, you know, whatever else, there's all kinds of stuff that comes along with accessibility.

If you're… if you're already working towards that, the number one takeaway is, like, you're doing a good job, keep at it. Like, you need to, like, it's good to have accessible content, it's good for so many different reasons.

On top of that, I think, you know, we keep coming back to this across all of our, kind of content intelligence-focused webinars.

Make sure that your content itself is consistent, is clear, and is understandable by people, because if it's understandable and clear and consistent and readable for humans it's gonna be the same thing for the AI bot, and there isn't really a way to game this system yet. It really comes down to great content and great structure supporting that content, so that it's accessible and readable.

[00:41:46.500] Megan Andrews: 

I like that. The games will probably come later.

[00:41:49.800] Joel Goodman: 

No doubt. Once it's pay-to-play, I'm sure there's going to be something that happens, but we're not quite there yet, right? So, yeah, to wrap that up, just want to say thank you all for being here with us. Again, our emails were previous. We're going to send these slides out, this video out.

So that you can read it, or read the slides, watch the video, all that sort of stuff. If you would like a free AI visibility report, we would love to create one for you and kind of show you where your website is at. Then we can also give you a demo of the Accessibility Auditor and the AI readiness tools within Content Intelligence.

I think it's pretty cool. If I'm on the call, I'll show you the actual website that we ran it on, on squizz.net, and kind of where our gains have been, because I think that's kind of fun.

Otherwise, feel free to email, ask at schools.net if you've got additional questions, and we can answer them in the follow-up. Otherwise, thanks for being here. It's been… it's been great. Megan, always a pleasure to have a webinar with you.

[00:42:47.250] Megan Andrews: 

Yeah, can't wait for the next one, Joel.

[00:42:49.490] Joel Goodman: 

It'll happen, I'm sure it'll happen. Thanks, everybody.

[00:42:52.610] Megan Andrews: 

Thanks, y'all.