Master Digital Accessibility to Make Sure Your Tech Venture Grows with Inclusivity with Michael Nicastro

Did you know 1.3 billion people worldwide face barriers when using digital products? And that companies ignoring accessibility are losing out on an estimated $1.2 trillion annually? Brands like Apple, Microsoft, and Nike are investing heavily in digital accessibility, not just for compliance, but because it drives growth. If your company isn't making accessibility a priority, you're not just missing out on revenue, you're losing customers. How do you ensure your tech venture is accessible to all while maintaining high quality assurance standards and scaling these efforts through automation? That's what we'll dive into today. Welcome back to Mastering Tech Growth, where we interview industry leaders to bring you actionable insights and inspiring stories. I'm your host, Mike, and today we're covering a very overlooked aspect of scaling tech ventures: digital accessibility. It's not just a compliance checkbox; it's a growth strategy that unlocks new markets, drives revenue, and helps your product reach its full potential. Joining us today is Michael Nicastro, VP Group Director of Technology and Head of Accessibility at Digitas. Michael is passionate about ensuring that digital products are accessible to everyone and uses quality assurance and automation to drive growth without sacrificing inclusivity. Michael, thank you for joining us today.

Thank you for having me, Mike. Uh, love your platform. Uh, yeah, great intro. What's interesting is I was hired at Digitas to run quality assurance and QA engineering. Accessibility wasn't a function at Digitas at all. Um, you know, they I don't want to— it's not badmouthing them. It just— it wasn't a focus, especially on the client side. They had internal practices, but I sort of came at the nascent of when they really needed someone with my skill set. So I quickly pivoted what we were doing there and built an accessibility center of excellence, which was greatly needed. So there's two facets of that. There's the internal preparedness and readiness, right, getting your cross-capability. Teams ready. And as I work at a marketing agency and, you know, I come from a tech background, I was at startups, I've been at enterprise companies, this is my first time working for an advertising marketing agency. And the caveat there is you're not just building a product, you're servicing clients. And when I took a look at our client landscape, you know, we're all global now, you know, Digitas, I work for Digitas North America, but we're a global company under Publicis. Looking at our big clients like, you know, Kellogg's, which is now Kellanova and WKK, major banks like Bank of America, Samsung, none of them were thinking about accessibility as part of their business as usual work. So I saw a, you know, an egg to reach for there, and I built a program around that.

So that's a very organic appearance of a, I want to call it like a business process or like a business feature. And I have to effectively add from my own background. So I worked for the startup, which was called LoHomswap. It was a service providing different ways of traveling. So effectively, you would rent your home and someone else on another end would rent their home. So you would just swap homes. It was literally based on a movie called vocation, I think. So, so we recognized as well eventually at some point that we're not serving wheelchair people because they need accessibility to the actual house being rented. And then our product people sort of went out to the market and they start chatting with our customer base. And we realized that we actually don't have any ability to blind people to use our platform. And the ability to deaf people to use our platform. So it became a very huge product— well, not a product, but a project to actually cater that set of market. And it has a huge impact on the revenue, which, you know, I'm sure you have success stories like that. But that, you know, that leads to that very first question, Michael, is like, you know, like, how does it really open new markets? If you look at your product and you go like, I have no accessibility, okay, let me do this. And what one should expect when it's done and, you know, how you actually put it out to the market in such a way that it actually yields that benefit.

Sure. We'll start by talking about how we engage with clients, right? So luckily working for Digitas, we have, you know, hundreds of clients to work with. So I work with our client leads and I look for organic opportunities first. So not really going in cold. And what we'll do is there's a couple ways in, right? One is the typical way in is when a brand is litigated against, right? So here in the United States, the Department of Justice enforces, you know, the WCAG, which is, you know, the guidelines, the web content accessibility guidelines that are enforced by the World Wide Web Consortium, which is the governing internet body. And if you have an infraction, what's interesting is currently today it's small, right? You could be a billion-dollar corporation and that litigation is probably $75,000. You might be missing alt text for an image or something small. And there are law firms out there that go and that's their sole organization is looking for accessibility infractions. But what's interesting is, I don't know if you've heard about the European Act 2025, the European Accessibility Act, otherwise known as EAA 25. That's coming next year, and that's coming in June. And what's important about that is all the European member states— I think there's 45 of them— are being forced to adopt, similar to GDPR, adopt these standards based on the WCAG. And each European member state will have their own Department of Justice and how they enforce it. What's very interesting and what's helpful for me working and selling this to our clients is you could be sitting in Germany and you could be on a website that's produced out of the United States, but you can still be sued or litigated against even though you're not in Germany. And the difference here is that the fine is going to be based on your yearly revenue. Right now, they're still— I mean, it's codified, but each Department of Justice for each member state will enforce it in a different way, but it could be up to 10% of your yearly revenue. So if you're generating billions of dollars, we're no longer talking $75,000 US, we could be talking millions in fines. So back to what you were saying, that's one way in, right? Is they're litigated against.

Can I pick, Michael, can I pick a little bit here? So you said it's gonna get enforced. Um, so are you saying that any website online will need to have accessibility features, or is it for like e-shop? Like, is there a, you know, a sort of an area at which point you're doing business and that's going to become a must-have compliance checklist?

There's a cutoff. I think if I recall, I don't have it in front of me, it's 100 employees and over $2 million in revenue per year. So if you're under that, you would not be subject to infractions, but So any brand I work with is essentially at that scale, so they would be in trouble.

So, and we're talking about it doesn't matter what you have online. Are you selling an item? Are you providing a service? Or is it just a blog you have? If you're making money and you're making revenue, which is at a certain level, you will have to have accessibility features. Otherwise these, let's call them headhunt lawyers, who's going to be looking at you not being compliant with those, they will be making money basically by just headhunting these websites.

Correct. Yeah. Like you said, it doesn't have to be an e-commerce site. It could be a marketing site, you know, with no user journeys or path to purchase. So let's see how it's codified. Let's see how it's implemented. I'm happy about it because it really drives what I'm about is, you know, ensuring that there's accessibility for all and And I think that's going to be great, but I think it's going to take time. Right. Like what I've read is certain member states are adopting it quicker than others, like Ireland is compared to Belgium. So we don't know how it's going to work, but it's a good thing no matter what. Do you want me to continue on that?

Because it's okay. Yeah. Sorry to divert a little bit. So we were going like, how do you actually push that out? If you build it, how you push it to the market that it actually yields a positive revenue impact.

Yeah. So again, it matters what's the client's need. Are they aware that they have accessibility issues, right, which has typically been the case. So let's take a client at scale. We'll take a retail client that has, you know, let's say $10 billion in revenue a year. They have e-commerce sites. They might not know— and this is a real case— where they're at at all. So what happened there was I worked with our client lead and I said, look, we have this— Center of Excellence. Um, and we have this linear product, uh, suite. And I'll just state it first, it's— you start with assessments and audits and a competitive analysis, and we'll go into them if you like. Um, then we go into, um, process gaps and discovery through workshops. And then we go into, um, unpacking playbooks, delivering playbooks, unpacking them, uh, training. And then we sit on a retainer and governance. So That's sort of the linear progression. We can also sell these as a product à la carte. Like, we can come in and just do a retainer in governance. For an assessment, so let's take a website that might have 3,000 URLs. They might be direct-to-consumer. They have 3,000 PDFs. Within 2 weeks, what we can do is look at your production environment, so it's very light touch. We'll use a blend of automated tools, right? Tools like SiteImprove or, um, aXe, uh, things that look at your CSS, your HTML, and your ability to be accessible. But that's not the whole story, right? So screen reading— I'm not sure if you're familiar with screen reading or our audiences, but those with low vision and blindness utilize assistive technologies like JAWS or NVDA And then on macOS, they would use VoiceOver. On mobile, use VoiceOver and TalkBack on Android. So it's a way for someone with low vision or blindness to navigate using a keyboard and hear auditory signals of the page around them. So there's a lot that goes into that, right? We need consistent navigation. We need code behind all the elements within the page to be able to read properly. So what we do is we do an assessment of looking at it through automation, so we can look at all 3,000 pages very quickly in 4 hours. But then we need to screen read, and how do we do that? We analyze the pages throughout your website for consistency, so components and templates. I don't want to get too technical here, but, you know, most of the companies we deal with have a major CMS like Adobe Experience Manager or Drupal or Salesforce. Some might have Shopify or WordPress, but if you know anything about how websites at scale are built, they use a CMS, which usually has common components. When you think about like a calculator on a page or a footer, most of your calls to action. So when we look at these 3,000 pages, most sites at this scale are going to have 30, 40 components, and they're going to have 10 to 20 templates. So what we'll do is we'll identify those pages that have a good representation of that, and then we'll screen read them. Because screen reading one URL can take up to 8 hours because you have to do the desktop, both Android— I mean, you have to do both, excuse me, PC and Mac, and you have to do both Android and iOS, and you have to go through it, you have to listen to it, you have to understand it. So we do that in about 2 to 3 weeks, believe it or not, and we can do that for free, right? Depending on our intentions as an agency to get in the foot in the door, we can do that for free or we can do that, you know, at cost. Once we do that, we provide a list of Here, here's 900 issues that we found, and then we calibrate them. One thing to understand about the WCAG guidelines that, um, that tell us what accessibility is, they're binary. They say pass or fail, but we come at it with a marketing lens and a user impact lens, because when I present 800 issues to a major company, they're like, I don't know how to fix this. Like, we can't fix 800 issues. We have releases and campaigns going out right now. Tell us what should be fixed, so we'll prioritize that roadmap for them by user impact. You know, we've got— like, if you know anything about typical software development, you find a functional bug, it's severity 1, 2, 3, or 4. We do the same with accessibility, but it's based on the user's impact. For example, severity 1 is, you know, an accessible barrier that doesn't allow them to complete the path of purchase through an e-commerce site. Like, there's a, you know, there's no— the screen reader, the element's not working at all, or the So once we do that, we'll organize that backlog for them and we'll leave it at that for the assessment. We'll unpack it for them. That typically leads to the next step, which is why are we at this way? Like, we know, we see we have issues. Can you tell us how we got here and can you tell us how to fix that? So we will come in and we will interview. We have a templated process. We'll interview all the stakeholders working for that organization. And when you're talking about a billion-dollar organization, you could be having upwards of 10 vendors, right, that are doing development, that are doing QA, that are doing creative, that are doing internal IT. So we'll interview every— we'll interview legal as well because they need to understand what's happening when a brief's brought against them and they need to understand their current enterprise solution for accessibility. So once we do those interviews, we're able to assess where the breakdowns are. You know, is it process breakdowns and how they're communicating? Is it testing breakdowns? Is it development? Is it business? Do they have enough skin in the game to care about accessibility and fund it? So the answer is usually yes to all of those. So we'll unpack those, we'll provide it. That usually leads to us needing to come in and help with training and designing playbooks around all these. So we'll design specific playbooks for each capability across the way and we'll unpack them in training sessions. That usually lends credibility to us as a marketing agency, as an accessibility expert. Then we come in for governance. We'll do monthly audits across their digital ecosystem, and remember, it's not just websites we're talking about. We're talking about CRM, we're talking about email that gets sent out, we're talking about your social marketing, we're talking about your internal business-to-business tools, right? Your timesheet tool, whatever you're using. We analyze your third-party tools, so you might be interfacing with like a Bazaarvoice or Pricefighter, someone who's setting up surveys for you. That is an important way to look at the holistic picture of accessibility. And then the last product we offer is consultation and retainer. Typically we'll sit on there, we'll work with the team. Are they doing a redesign? Do they want to know how to make a new carousel? Are they finding issues and bad trends within their content? How do we help them work and fix those bad trends? It could be that a legal brief is brought against them. We'll interpret the brief and we'll put together a process of remediation, how to respond to the client. So I said a lot there. I don't know if there's anything you want to pick at, Mike.

Yeah, I do. I do, Michael. So I was taking some notes. So this is what Digitas does. So imagine our listener, and by the way, I love the fact that you will do an audit or assessment and that, you know, can be done free. So effectively just opening that relationship, you're showing that value and then, you know, it leads where it leads. So let's say some of our listeners are maybe smaller companies and they go like, you know what, like, I don't want to open B2B relationship. If that would be the case, how they could just use the playbook you, which you just gave? You mentioned some tools for the assessment. Are these pay tools which you guys use? So you mentioned like a bundle of tools. Is it like open source tools? Could anyone just go ahead and use that tool right now? Okay.

Yep. Most of them open source. Uh, for example, Axe, A-X-E, it's a Chrome desktop extension. You essentially go to a page, you press the button, you get a report of your issues. Google Lighthouse is another one. I don't know if you ever use Google Lighthouse.

I do.

Part of its accessibility. Yeah. But you know, it looks at SEO, looks at your performance. Accessibility is part of that. The caveat there is if you're a small business, that's great. I get an understanding of where I'm at. But if you don't have a subject matter expert on hand to be able to make those changes, and to interpret those results, you're really not getting anywhere. And also these tools are not going to do screen reading. So I've read somewhere that, and in my experience, you only find about 70% of your accessibility issues using a tool. You have to do screen reading. And screen reading is a heavy lift to onboard and learn, especially on the mobile side with all the gestures and swipes.

Okay. So let's talk a little bit about this because that's my second question. So I'm at home, I'm listening, I'm like, okay, well, I need to jump on this, uh, you know, digital accessibility train. I'm going online, I'm looking for some free tools because I just want to understand, you know, uh, exposure I have. So I run through my website, you know, for sake of argument, let's say I have an e-commerce which is making some money. So I run this and I get, you know, here's the report. And, uh, that, that next step for that screen reading— so we effectively saying there's no free nor paid tools, it's an actual manual labor process to actually do that?

Well, there is free screen readers. NVDA is free, but the, like I said, learning it is difficult. You could do it. You could train someone to do it, go through YouTube and whatnot. The two major screen readers right now are JAWS and NVDA. JAWS is paid. The market share is like 50 to 50. Most of our clients recommend that we use JAWS or they want us to use JAWS once they understand both. So it is, NVDA is open source. And then again, on your phone, which we know the market share of phone usage has surpassed desktop, they're both native, right? So TalkBack for your Android, VoiceOver for your iPhone. Those are free, they're already on your phone. Yep.

So the screen reading as a process for potentially listeners who don't really know what that process is, how that goes? Imagine you're doing this right now for Brand X, how does the screen reading process actually go?

Yeah, so essentially, you know, we would host our JAWS or NVDA, you know, in the cloud typically. And what you do is within that platform you would access your URL, and then we have trained folks who go through it, and it's typically keyboard only, and you're tabbing and you're navigating. There's a whole list of 40 to 50 keystrokes that you use to listen. So if you think about the way a website is built, you need consistency across your pages and consistency within the page. You need the correct heading structure, right? So think about it that way, because if you're screen reading and and you're going through it as a low vision and a blind person, you can't be skipping from top to bottom. You can't be all over the place, right? You need consistent navigation. That's one. Elements need auditory information for them. So for example, if there's an image on there, it needs to be— there needs to be an alt text behind it that tells you what this image is because someone who can't see an image needs to understand what that image is, right? And it might be an important image. Think about a graph or an infograph or a chart, right? We need to have that information written some sort of way for audio description, right? So that's— it's quite complicated, believe it or not, when you look at it. And what's interesting is there's this company Fable that we have used in the past that they employ people with disability. And they will come through and screen read for you. And I've watched them do it. And what they'll do is they'll send you a video. So we do this for our clients. We do now and then for our big clients at Skill to get also help foot in the door is we'll ask Fable to go through and screen read, let's say a path to purchase for an e-commerce site. So logging in, if there's any authentication for your loyalty or whatnot, building your product page, removing items from cart, adding them to cart, checking out and paying, right? Something that someone should be able to do within 5 minutes. If my team who typically doesn't— we don't have many folks with disability on my payroll. So we have trained people to do it and it can take them, like I said, up to 8 hours to go through a page and analyze it. These folks from Fable, they'll send us video of them going through it at incredible speeds because they have low vision, they have blindness. They've been doing this their entire life or since the internet's been out there. So it's a complicated system to use, and kudos to those who use it every day and what they've learned and gone through.

Okay, that's a good shout. So effectively what we're saying is that if you're thinking you're going to save money by doing that yourself, you're probably wrong. It's just that the learning curve and delivery and execution of that process without having any prior sort of training and understanding is just going to be way too high. And if you have like one website, you know, one product, you actually looking either to open a department within your business or actually much better idea to actually outsource and get a partner which does that professionally. That's pretty much your notion with pushing. Okay. Yeah.

And there's interesting, there are small companies out there who are looking to work with these smaller brands who usually are on Shopify or WordPress, right? Interesting. I met with some folks from MIT yesterday, actually. I've been having communications with them. I met them in accessibility forum. I believe the name of the company is Patrol. They're brand new, but that's their attack. They're looking for companies and brands that are small and have small budget and looking to be able to stay with them, you know, not at the cost of like a Digitas or Publicis. And they're doing some fun things with AI that What I've always been interested in is, can— on the topic of screen reading, can we use AI to screen read without having to be able to go through all the steps of this talk about JAWS and NVDA? Can we? Mostly for me— yeah, I think we can. The problem is auditory, right? We can automate most of accessibility testing and most of regular testing, functional, you know, NFR, but The problem is that auditory response. So when you're tabbing through and the screen reader is spitting back everything on the page to you, there needs to right now be the human ear that listens to that auditory response because there can be nuances and issues with what's being read out to the human ear. So where this company I spoke to yesterday, they're looking at ways to automate that via AI. I don't know if that's going to be great for the end user. They might still want to use a screen reader, but for me, in validating, that would be a huge plus for us.

Yeah, that would add a lot of acceleration to, I mean, all the businesses looking to embark the digital accessibility phase. So when you were talking about that potential, you know, human aspect, which is still needed to actually listen to what is being read back, I instantly thought about AI bots and AI support bots, which we currently have. You know, we can see enterprises doing, you know, Amazon, is a good example. If you try to reach customer support, you will get AI bot and they still not at the level of an actual human being when it comes to help provided. You often get annoyed by what they're saying and you go like, just give me an email, just give me a person. And we actually had a guest who run a company when it comes to customer support and call center support and their play on AI is that these bots and these help support bots, they will become better than the people once they will learn, you know, your profile, your personalization. They will know exactly what you're about. They will actually gonna be able to give you a better support than people can, and people will actually want the bots than the people. So when you were giving that example, I was like, will we reach a certain area or a time when, when these accessibility, you know, AI bots will be so good that, you know, people will go like, you know, like, no need for humans to listen, you know, let's just give to the bots and, you know, all this dandy and cool.

Probably. I mean, I, I'm excited about the future, and not to go on a tangent, but when I think about augmented reality, think about— and there's things out there that I've seen, like a pair of glasses for, for a person who's blind or low vision walking down the street and it's able to assess what's in front of you. Oh, here comes up a street because it has the GPS behind it. It has the visual authentication, what you're seeing, you know, stop in 40 feet. And then there's an auditory signal that says stop, and it tells you— you look at a building, it'll tell you what the name of that building is. This is that address. So I think about that. But back to AI, I'm on the Digitas AI steering committee, and I'm working on a few client projects right now with AI. It's really changed the game, right? And one funny thing we're doing right now is we're using AI to check AI. So like, when you think about what we're trying to do with AI, and I don't want to share too much because some of this is in development, but you know, there's what does creative want to do with AI in content and copy? What does development do with AI? We know about Copilot, you know, I'm sure you've heard about that, and I think I saw one of your podcasts, someone was talking about Copilot. And then from QA, what we're doing right now is, um, we're using AI, uh, and specifically OpenAI LLMs to generate test cases and use cases, right? So, um, when you said that eventually AI, you know, will— the chatbot will surpass the human, yes, but don't forget right now, and I still see this for a while, there's a human behind it indexing it, embedding it, um, ensuring that the output training it, uh, is specific to the needs. And I find and my experience with AI over the last 2 years is that content ingestion into an LLM and into, you know, your database, whatever you're using, Pinecone or ChromaDB, um, and the massaging of that data to make sure you're not in the old GIGO, you know, garbage in, garbage out, uh, is the hardest aspect of making a chatbot or making an AI tool. Um, and that's what, you know, for our test cases, right, to be able to generate a test case where you have to tell it the format it's supposed to come out We need to have— tell it, you know, we do this for accessibility, we do this for functional. We need to tell it, hey, you know, these are the type of requirements we're gonna be inputting, so to understand it. So in your system prompt, we're constantly updating that, we're constantly indexing it with requirements and with all information that QA person would have in their mind over their years of experience and spits it out. They spit 'em out fantastically, but then there's still that human element of reviewing them and massaging them.

So yeah, we definitely need to go on an AI tangent. No, it's all good, Michael. I mean, because we wouldn't be technology podcast if we wouldn't talk about AI. There was many episodes about AI, especially about, you know, applicability of AI within the businesses and what it actually does when it comes to automation, what value it actually brings, and, you know, what, what are the trade-offs. And, and that trade-off which you just mentioned has been mentioned by by many guests who actually have businesses built on the current AI capabilities. It's a very real challenge and it's interesting to see how, you know, how the market will solve that. So come back, you know, to your blueprint of, you know, doing digital accessibility. So we are at point 4. So we run these reports, you know, we did the screen reading. So we got this report. We know where our issues is. And you mentioned this thing which stuck with me, which is the impact or user impact. So can you talk a little bit about what is that? Like, how do you guys calculate this? Is it like in revenue loss a year? Is it about, you know, customers dropping off? Like, you know, what is that impact and why business should care about it?

Yeah, that's a good point. Really what it is is looking at the impact to that end user. And there's multiple types of disability, right? So You mentioned at the start there's over 1.3 billion and it changes every day because a lot of people don't understand there's temporary disabilities. Think of concussion, think of a broken arm, you know, mobility issues, temporary blindness. Someone might have just had cataract surgery. And then neurodiversity spectrum, a lot— that's something I'm really focusing on a lot now. So the autism spectrum, those with seizures, there's a lot to think about. So when we look at impact, It's looking at a defect that comes out and what is the impact to that person? Is it a complete block of usability? Like the screen reader doesn't work at all or the screen reader doesn't work on iPhone but it works on Android. Like that's like a level 1, right? And then we would look at a level 2, which is there's a portion of the screen or a portion of your path of your purchase that doesn't work. It's not a complete block. But the person's probably going to abandon their journey or the website because of that block. We always recommend that those two should be fixed immediately. And then you have 3 and 4. 3 is, there is an issue, but there might be a workaround. Most likely there's a workaround, like the heading structure is off and the person has some confusion tabbing around, but they're able to finally work through it and get through it. And then severity 4 is usually enhancement, where we think it's a— we think there could be a better experience. And when we talk about accessibility, one thing I wanted to mention was when you fix issues for those who have disability, you're fixing issues for all. There's this thing called the curb cut effect within accessibility. And what that means is when you look at a curb in the street, right, they have the ramp for those with disabled. But when you make a ramp for those who are disabled, now the postman who goes up that ramp can push his cart. A kid with a skateboard is able to go up it. An elderly person is able to walk up that ramp instead of having to go over a curb. So the same within the digital ecosystem. When we fix issues for accessibility, we're making a better user experience for all. So that's how we look at impact is how does it impact the end user. We're currently— you asked about how does it impact the client for their end revenue. That's a difficult one because, you know, there's some concern with looking at privacy of— you know, we can see who's using assistive technology to engage with the site, right? We can look at that data analysis. Being able to understand the various disability of each user is not something to really be able to quantify. How do you know? We don't know if someone's not using assistive technology if they may have a neurodiversity or a mobility issue. Possibly they might be keyboard only for mobility issues and/or neurodiversity instead of using a mouse. We don't know that. It's hard to quantify that. The other aspect of that is that if you want to— like, if a client asks, if I fix my website to be 100% compliant, how much more revenue will it generate? We've tried to quantify that, but when you think about a major brand, take any major brand, and they're constantly putting out campaigns. They're constantly changing things, right? So the website might have, you know, 400,000 visitors per day in June, and all of a sudden it upticks to 1 million per day in June. Is that because we fixed accessibility? Or is that because they launched a new campaign for a new sneaker that everybody wants to look at? So, um, I've struggled with that. That's something that I've worked with our data and analytics team to do because a lot of clients are asking that. But the one way I do it now is that I look at your brand and your demographic, right? Every marketing, uh, agency or every CMO for a company has a target for each product, each campaign, or each brand. We can look at the statistics for Males 10 to 30 in the United States have this many disabilities. And we can sort of say, well, if you become more accessible on your D2C, um, site for this demographic, you're opening it up for 500,000 people to be able to access this that weren't able to access it before because you had a user— a severity 1 impact that prevented the path to purchase along the payment. So that's how we do it today.

I get you. I mean, it's almost like pseudoscience. It's like there's an educated guess you can make, but it's going to be very hard to really quantify it and create a formula or a way which is replicatable in every single case with every single brand and actually be consistent with your projections. So if we try to flip that question or the answer I was looking and actually talk about you know, success stories, which could be anecdotal, not necessarily seen the data, but you work with the brands and, you know, they have these accessibility issues. You've seen those reports and, you know, had huge impacts. You know, you guys did your thing and then you got feedback, which is, you know, more customers staying longer sessions, you know, more revenue. It's like any sort of like, you know, success KPIs to share.

Yeah, sure. So I think there's a few. There's organizations that have been litigated against multiple times. And, you know, going over time, if I'm engaged and we're engaged with a client for multiple years, which we are with a few clients, no litigation. That's one, right? That's, that's a big step for them. Two is the actual auditing of the site. So when we come in and think about this too, and this is how I try to tell a customer or a brand, is if you like— when you are going through a normal development software cycle and you have issues and bugs, do you release with them? No. Why? Because they're bugs and it's a bad user experience. They didn't know that there's accessibility bugs and they're releasing without them. So what we'll do is we'll do audits for them and we'll show them, or we'll train their internal teams to do audits, and we'll look at the trends going up. So when we— for example, there was this client when we first came in 4 years ago, they were at 68% compliance, and we have a proprietary— our own proprietary scoring system that we use for them. It's based on automated testing, it's based on screen reading, it's based on their e-commerce journeys, and we scale it based on user impact and number of issues found. And what we've shown them is using that same audit and scoring system, they went from 68% across 20 brand sites to 78% within 6 months. And then 2 years later, they were at 98%. And that was being able to fold in all these remediation of these 800 to 900 issues alongside their normal business as usual scoped marketing work and their scoped design work. It didn't cost them much to fold it in. And that's a success story in my, in my opinion, that they have fixed over 800 issues. They're preventing issues to go into production. They're preventing litigation. The other success for me that's sort of personal and instills pride is there's a huge culture shift within an organization. So as you probably see when you talk to people in the organization you work in, the shift for DE&I in every company in the last 5 years, um, DE&I is being funded, right? I work for a very progressive forward company at Publicis, so, uh, DE&I has its, um, lens over everything that's happening and it's wonderful. And when you get into a company, um, like the ones that I'm working with and they have a DE&I backbone, um, shifting the culture for accessibility is a win for marketing, for the business, and for IT, and for the development teams because it, you know, dovetails into everything DE&I wants to do. And what we try to do is say it's not DE&I, it's DE&IA, because that inclusivity part focuses a lot on gender, on race, on sexuality, but accessibility sadly sometimes gets left to the So what we always try to do is just add that A and let people know that that's part of it. You know, it's part of inclusivity, but it's also part of the DE&I structure in general. So I think those would be the way it's successful. And you bring up a good point, Mike, because the fact that we're not able to right now quantify, you made an extra $2 million right now. We can guess, you know, we have the figures out there and there's been studies that show like you talked about, the trillions of dollars that are in the hands of those with disability and their families across the planet. Like I said, we can break down demographics and tell you, you probably made, you know, extra $2 million this month because of your inaccessible barrier that you removed. But I don't like going into with that. I like going into with what I discussed.

Yep. I love that, by the way, the DUI and putting A, And that's why in the presentation as well said, a highly overlooked aspect of tech growth, which is the digital accessibility. It is highly overlooked. It's one of those things where no one thinks about until they are forced to think about. Like, you know, it usually comes in these compliances areas. And before that, people don't really realize that they're literally leaving money on the table. So—

Yeah, just one more note on that. Sorry, Mike. So you remind me of— we also do a competitive analysis during the assessment. So if there's a major brand that's selling a phone, you know, take Apple or whatever, we'll go look at Samsung, we'll look at Amazon, we'll look at Best Buy, look at other, you know, Google Pixel, their websites, and then we'll show to the brand this is what the other brand is doing very well. Like for example, Apple does very well with accessibility. They've made huge strides in it. So That's a tool that we can use to also show your success. Like, hey, you wanted to be as good as Apple? You've made— you've gained a fraction of that by being accessible because your competitor is doing it.

Yeah, yeah, 100%. I mean, so easy to imagine, like, you know, say you're blind, you know, you go online and you go, I want to buy a phone. You know, you're tabbing through the Google results, it's reading you, you know, this website, that website, you go inside the website and nothing works. Well, you instantly going to leave and go to another one and then you land on Apple and it works and you buy product because the happy path towards that ability to purchase is just there. It's very simple, right? It's not rocket science. So before we switch gears, Michael, so we have this question which is in the spirit of the show and we ask every single guest. So when I say mastering tech growth, what comes to mind?

Mastering tech growth. So for me, number one, you can't master tech growth in your capability without a passion for it. When we talk about accessibility, like my grandfather was a quadriplegic from polio, believe it or not, there was a great outbreak in the '60s in Massachusetts at the polio— he was a Postal service man. And I grew up in my grandparents' home early on in my life. My parents weren't married, and I had to watch my grandfather suffer. And, you know, everywhere you go now there's disabled parking spots. They used to be called handicap, but now we call them disabled spots. And that, you know, when I was— I'm 50 years old, so when I was growing up, I watched my parents struggle to get and helped my grandfather get out of the car. And I just— this is before the internet, but I still saw that. And then when I went to college, one of my jobs there was— and I just applied for it out of the blue— reading and doing tests and homework for those with disabilities. So we would sit in a room day after day and do the actual tests and read to them and got their responses and wrote it down and used the computer for them. This was in the mid-'90s. So I had a passion for accessibility from the start. From the other tech perspectives like quality assurance, automation, I'm passionate about it. I, you know, my girlfriend will tell you that I like picking apart her flaws, you know, and she wouldn't be happy about that, but that's QA, right? I started out as a, you know, I majored in computer science and geographic information systems. Um, but I soon realized I wasn't a good developer. Um, what I was really good at was finding issues and picking out things that were bad. So I had a passion for that. Um, so passion's one for mastering tech growth. Um, the other is, um, having a good team. And what I became best at was team management, right? So, um, right now I oversee, I think, 68, um, different QA, accessibility, and engineering people. I've worked in companies where I had teams of over 100. I find myself to be— that's what I'm best at, is strategically leading teams and understanding who needs to do what. I look at it like, you know, you're from Scotland, are you?

No, Lithuania.

Lithuania. Geez, I thought you had a Scottish accent.

No, thank you.

But soccer, football, the best people who manage teams manage them like they're managing a sports team, right? There's people on your team that are A+ to your star. They need to be given and trusted with certain things. They need to be treated a certain way. You have people who are on the bench who have a role. What's their role? What do you trust them with? And how do you utilize them? So that's how I've always looked at it. So passion, having the right team, and then innovation. We talked about AI. Everything that my team does and I do, I'm always challenging them to what's next out there, right? What are you looking at for resources, right? So are we constantly training our team? Are we looking through Reddit and subreddits? Are we using YouTube? You know, are we using TDLR? Are we, you know, are we using— what are we doing to be able to understand and innovate? So when I look at how to automate something, I always— or when I look at how to test something, so let's say we have a new project. The first thing I ask my team is what can be automated and what can we use AI for? Before you do anything else. Then we eliminate that. We say, okay, these are what we're automating, because you can't automate everything. You can't use AI for everything. There has to be some manual approach to it. Um, so, so that's how I master tech. I have passion, team management, and push innovation.

Very great answer, Michael. So what if I make this more difficult on you and I say you can only choose one?

Ah, Uh, team management.

Team management. Okay. So even, even if you don't have passion for what you're doing and you're not driving—

so choose one. I'm thinking, let me make it difficult on you. Yeah. I'm looking at scale, right? So I'm, if it's a startup passion, if you're in the organization I'm in at scale, it has to be team management. If your team isn't rolling in the same direction with the strategy of those above you, you're going to fail no matter what. So I think it depends on where you're working.

I love that. Yeah, that's 100% true. 100% true. There's a, there's a different skills and different requirements, um, required to, to the business maturity cycles, which basically change, uh, it's called Dunbar numbers. So you have like 5, 15, you know, 50. Anytime you pass that headcount area, I mean, it's a guide. It's not a rule. By basically means that the old rules or whatever, you know, skills and people took you to that level, it's very likely that it's not going to apply to the next level and you need to—

it's a tipping point. Yeah, like Malcolm Gladwell. I don't know if you've read anything about Gladwell, uh, but he has a book called The Tipping Point. That's exactly right.

Yep. Um, okay, so I love that. So let's go back, uh, you know, to our key topics, which is digital accessibility, QA, and automating both of these. So yeah, I left the last bit, which I want to take us back to your blueprint, but I left this to QA because I feel like, you know, it has this relevance. So you mentioned that, you know, big brands, they will have partners and you might, you know, they might have like 10 vendors feeding into that online product, which needs to be digitally accessible, which means, you know, if they're feeding some type of content, that content itself needs to be you know, checked and validated, but it's not your content. It's actually content which comes from your third-party partners. So all that, all that picture for me, you know, looked like that's going, you know, that's where a lot of human error, you know, process error and things like that, which could appear, which led me to that question or leads me to that question is like, how do you manage quality on all of that input? So the output is still in a fully digitally accessible product.

Yeah, it's a big question. Uh, standardization, um, experience is the way to do it. So I, I try to, and I've always been this way, um, have standards on how we do everything, whether it's a test plan, whether it's your automation framework. Everything has to be scalable and reusable and adaptable. And so it doesn't matter the type of project that I go into, I have a standard set of templates standard set of test strategies, a standard way to automate that can adapt to anything we go through. What makes it difficult, and what you just brought up, is when you talk about the cross-capability, the cross-agency, the cross-vendor aspect of it, and it's setting up the workflows and the processes between them. And luckily most organizations are using like an Adobe Workfront or Atlassian's Jira to communicate They're using a wiki and a Confluence. And again, everybody's using it differently. I take my own organization where we use Jira, but the templates for how a defect is entered or the workflow and the definition of done is different per client, per capability. When I came in 7 years ago, I tried to fix all that and it just, it was a no-go. We're too big of a company. Everything's organically grown that way. That throws a monkey wrench into things. So it's understanding that and documenting the way things are working. But to answer your question, it's communication, it's standard workflows and processes, and it's assigning the right type of person to govern these things. And it can't really be done— I mean, there's certain comms you would have, right? You're on Teams chat, you're on Slack, you're on email. You need gold source records of things, right? So I always get mad at people who are not keeping gold versions of something in SharePoint. Like if someone is— they have a document and they take it and they download it and they work on it and they upload it, right? You need single source records of things across the capabilities. You need a good program manager, you need a good project manager, right? So for me as a QA leader, I can't do all that on my own. So when I'm driving QA and accessibility, you need people with skin in the game across the way, those program managers, those project managers. You need someone, especially if I'm engaging with a client, there has to be usually our champion, right? Whoever brought me in to drive accessibility or, you know, we also come in and just do QA for organizations, right? Whoever's championed me, I need their help a lot to push down what I am saying from the top down. It's usually the C-level, the CMO level who brings me in. Like, I need them to push it down. Mike DeCastro, Digitas, is here. This is what he's trying to do. And I'll put together a set of guidelines that I expect. Like, this is how we work. This is how we're going to use Jira, or this is how you use Jira now. Let's make sure we're continuing to use this way, but we're also going to use SharePoint over here like this. Does that answer what you're asking?

Yeah, it does answer. I was expecting very similar answer, Michael, because I was very keen to understand if you have something which is, you know, more automated. But reality is, and I have that in my professional life as well, is that once you have a lot of people and quite complex workflows, you know, there's a tendency to someone forget things or not deliver data on time. You know, having a source of truth, or as you called it, a gold record, is very key to be able to go there. And ultimately what you can do is establish your workflows and just advocate these workflows and Sure. And what I call police the workflow. So you need to have the program managers to really police in that workflow if you're expecting everybody to work with it. Okay.

So, um, so just on that note, um, I'm looking at AI to do that. So I don't know if you're familiar with like a Langflow, um, or a Flowwise, which, you know, incorporates LangChain. So if you've used these, you know, no-code, low-code tools, and I saw you had a, um, I forget his name, a gentleman who talked about no-code, low-code. They're great for, for where they can be used. And we look at that and how can we use Langflow or Flowise or Zapier or N8N to, to, to do these type of things. And we're experimenting, experimenting with that now. And that would help with a lot of what you're talking about, automating the processes. We do come in and, for example, internally we have a lot of media teams and, you know, Digitox makes most of its money on media. And they have a lot of processes that are like, we get all this campaign information from our client and we put it in a Google Sheet and the budget changes and, you know, this is going to YouTube, this is going to Instagram, this is going to Snapchat, and that's all manual. So we've come in and we've set up a framework that can automate that for them and it's been very helpful. So I just think that when you talk about your main point when you said, I thought you'd have like an automation. When you talk about just working with like 10 vendors at an organization at scale, um, we have to think about automating the view internally. But to get everybody on board to what we're doing, it's very difficult, you know, to get everybody to use the tool that we're implementing. It's just, you know, it's hard.

It's climbing an Everest while someone's throwing snowballs at your face. That's the analogy I use. Yeah, it is very difficult. Come back to the workflows which you mentioned, Michael. Yes, we did have the automation guru. His name was Greg. He runs business outside of other startups. He actually started way back and they were looking at automation and helping companies to go from Excel spreadsheets to an actual, you know, UI-driven automations or backend-driven automations, you know, when Zapier appear, a lot can be done. You know, these pipelines of work, uh, if you have a DevOps team, you know, who has an automation mindset, you know, people look at processes and they will go like, you know what, this part can be a bundle, this part can be a bundle. And then, you know, they put it in the code, they put a standard on top of it, and you have input and output, connect your, you know, stepping stones as I call them. And you can definitely automate a lot of these, but solving fully, I never seen that done yet at scale. So, you know, you can't—

no, we automate a lot. If you want to get into that a little bit, we were— so when I came on board, there was a small automation engineering team, and a lot of organizations, especially a marketing agency that has a tech component, they building an automation team, automation engineers are expensive. Number one, automating things, you need to have the right approach, right? Like you need to have a good ROI, like the business critical items, the redundant items, the repetitive items. But when I came here, we were Selenium, Java, and I stuck with that for a while. The framework wasn't scalable. So I hired a few of the best folks I could find and I challenged them and I said, look, what we have, our framework isn't scalable and adaptable from client to client right now. It's very custom. We're using Java. Why didn't I not want to use Java? Because most of our engineers are front-end engineers. Front-end engineers, what do they know? JavaScript, mostly. Python, we'll talk about in a minute. But so I transformed our team with the help of good people. We're now all Python. We're now all JavaScript. And we're utilizing a tool called Playwright from Microsoft, which is very easy to use. I think it's important because I have a— like I said, I think I have 10 automation engineers and 50 manual QA folks, and some of those are managers and leaders and analysts, globally dispersed, of course, for cost budget efficiency. I want to be able to have more engineers. So how do I do that? When we build a framework using Node.js and when we build a framework using a tool like Playwright, which is easy to use, and when we use Python, now DevOps can help. Uh, now the front-end engineers can help. So I don't no longer have 10 engineers. I have 25 now because everybody can use our framework. Everybody can incorporate and help our framework. They can go into, to Git and start iterating on what we're doing. So I thought that was very important when you talk about— to me, that was team management, and that was what helped us scale, was identifying ways to increase our automation coverage and efficiency without hiring more folks. Let's get the, let's get the other engineers who are doing other work, and now they're unit testing their own work using our framework and our scripts. Now they're helping us build new scripts. So we do a lot with automation. I think we're currently at 35% automated across all the projects. And how do we determine that? We have a test case repository, which most QA teams will have, right? And you can use many, many tools for that. And we look at them and we say, is it automated? Will it be automated? Can it be automated? And then we check that box. And then we look, we have 700 test cases, and right now 150 are automated. Next year— and so I challenge my team with that KPI. I say, I want, you know, 45% next year, at the end of next year, we have to automate our test cases because that reduces human error, obviously. It allows us to cover more. You know, think about all the browsers, operating systems, and platforms that we're contractually obligated to serve for our clients. And remember, we're not a product company, right? I've worked for product companies. We're a service company. We work with major billion-dollar corporations and we build software for them and we develop content for them and we put it in production and they expect it to work on Edge, Safari, Chrome, Chrome, two versions of macOS, two versions of PC, all that stuff. You can't manually cover all that. So what do we do? We build software within Playwright, sometimes Selenium still, sometimes Cypress, depends on the project. We build that, that calls libraries like LambdaTest or libraries like BrowserStack and that can emulate with real devices and go through all our scripts. So I'm super passionate about automation as well because—

It's clear.

That's how we save money. That's how we move forward.

It's a huge leverage. It's a huge leverage for any business to look at their processes and ask the question, how can I do this one time? How can I build the automation one time? It will take me longer, it's probably going to cost money, but then never ever come back to that process ever.

And well, there's maintenance to it.

There's maintenance.

Data changes, um, functionality changes, right? So you need a good maintenance strategy and That comes down to people, and that comes down to a process. So for me, the process is, um, it basically costs nothing to run your automation in a pipeline, right? So you talked about DevOps. We work very closely with our DevOps team. They're part of our team, essentially. Um, we set up pipelines, and our scripts are running all the time, and we look for failures. And is the failure because of a bug, or is the failure because the script is out of date because there was a new requirement? That's one way to do it. The other way to do it is constantly working with your product managers, your development team, right? Looking early in a project and say, what's changing in the code here and in the function? And then we go back and we proactively update our scripts. What I'm waiting for, and I think what we're very close to, is either GitHub, like Copilot or something, being able to analyze the code of what we're developing, analyze the code of my scripts, compare them, and see if there's any changes, right? I think you need proper annotation and commenting as well. Maybe we can do it looking at comments of this has changed, and then AI can look and say, okay, your, your automation script's no longer going to work because we see there was a change on this variable or this object. So, so maintenance is a thing, Mike, for sure. You can't just automate once and leave it. Maybe some, some places, some things on a legacy system that's going to be evergreen and never change, okay, but I don't live in that world currently where things just sit.

I get you. Yeah. Yeah, you're definitely going to spend time maintaining that code 100% because it is a code base and there's no— yeah, there's no do it once and never forget when it comes to the code. I mean, probably for nothing really, it's always a maintenance, right? So there's changes in technology, right?

And then languages. Yep, right. Like I said, we, we were Java, then we went to JS, and now we're mostly Python.

That's a very interesting journey, which I, I want to ask because you drop these couple of lines. One was You know, I came, it was Selenium Java, instantly saw we can't scale that, so we went and changing. Why, why there was presumption of like, yeah, we can't scale that? It was it because it was very custom and for every single client it was different and Java-driven? Like, what was your, um, thought pattern of looking at this and going like, yeah, that's not scalable?

Yeah, it comes down to people again. Um, you know, I, I'm sure there's people listening to this eventually, or maybe today, that are Java, Selenium, Java, any other test execution engines that are like, "Ah, that's bullcrap." And they're probably right. For me, it was about the people. We didn't have a lot of Java experts on hand. We don't have a lot of backend engineers. We work within a CMS. We work within Adobe. I have a lot of frontend engineers. They don't know Java. So that's why I thought the switch for us to scale would be able to scale to people. I also think it's— setting up your IDE, your onboarding of a Python and a JS is easier than Java for me. I think it is. And if you look at AI currently, it's all being developed in Python, right? Yeah. I think that just made sense to me going forward. And that's essentially why. Not because Java and Selenium are bad. I'm sure it works perfectly for a lot of people. For us and my situation and my team management, it was to get away from it.

When I think about Copilot, I always have this thought is whatever is the most amount of code online which is open source, that will be the best AI Copilot you're going to get because AI needs to learn on big amounts of data. And, you know, let's say Java is still— not still, it's, it's very huge. You have loads of systems being built like, you know, banks and healthcare and— but all of this is actually closed source. Like, you can't really go and train AI on that because, you know, there's a lot of business knowledge in that, and businesses just don't put that and don't allow, you know, GitHub or GitLab to actually go and train the modules on that information. When it comes to Python, it's just everywhere when it comes to open source. So I do believe anything Python-related will be like the strongest AI copilot. Just a side note and personal opinion.

I agree.

I'd say I'm right.

And I have very, very smart people under me, and like you can't succeed in business without having smart people working for you, right? So I give my success at this company, uh, and in my career to trusting the right people to do the right things, right? And bouncing things off them. And I love being told no, you know, I love being told all the time. I love suggesting things. And if someone tells me no, tell me why. I'm gonna, I'm gonna trust you because I trust you as a developer, as a QA person, Yeah, because when you reach a level, you know, vice president above or whatever it is, there's too many things going on for me to get down in the weeds and make that decision. I need to hear from my people. And that's why, again, I say success comes from team management, mastering tech. I need to listen and trust my people because if I can't, I'm going to fail. I can't scale without doing that.

I'm going to get you back on the show, Michael, and we're going to talk about leadership because You're going to that level a couple of more times. I feel like you want to talk about this. So I'll bring you and we'll talk about leadership and team scaling. That's a very interesting topic by itself. And we had a couple of guests answering the question of what is one most important factor for tech scale, pointing to the people and pointing to the culture and just pure leadership. Okay. So I do have one more. You mentioned one of the clients where had, you know, a lot of social activities, you know, where they're repurposing content, they need to put it in Snapchat YouTube and all the other social platforms. You said you built the framework for them to automate that part. Can you give us an idea of what, for example, you know, listeners thinking about this could do today to actually automate their own social, you know, distribution engine?

Yeah, funny enough, that was built in Java, uh, and it still is, um, but it works. Um, yeah, so let's see how to approach this. Um, when you think about multiple— the way this project worked was there's multiple tracker and campaign tracker sheets that have data that needs to tie together, whether it's budget, placement, geographical information on regions where the, the media will be placed. Um, these teams were all doing this manually, um, through email, um, through Google Sheets, through communication with the client on Teams chat. And every time something was updated, they would go in and manually change a column or a row. And eventually they would take a look at all these different sheets they had, and then they would have to sort of conflate them all and put them into one master sheet that then was the main thing that they would look at for budget, for execution, everything. And things would change along the way. So what my team did was go in there and use Java and Selenium and some custom— it's a very custom system, but it works for the media team we were working with— to be able to ingest this information from the various sources validate it and then place it, like normalize it. So what they did, all these Google Sheets, they had different formats too. Every, like the budget, the campaign tracker, the regional tracker, they all had different formats. So we devised a system that normalized it into one format because they didn't want to change. There were different teams all working within this media, and I find this all the time, um, like people, organic capabilities organically grow and they have a way of working and they don't want to change, right? They want to use that way of working. It's been working for 5, 7 years. So what we essentially, what this did was it normalized all these different formats into one format that the automation engine put together that gold record and they were using that to be able to go to market with their campaigns and adjust their campaigns. They kept adjusting all these other formats, And the system we wrote would constantly look at all these forms and check it against the normalized form and then update it and update it. And then they use that as a gold record. I don't know if I did a good job explaining it. I'd have to bring on one of my engineers to do better, but that's the big picture, normalizing different formats.

Yeah, that gives a clear picture, Michael. So what I was looking, it was to understand if you guys build something custom, which is definitely sounds like that.

It was custom.

Sort of looking and waiting, you know, until you give a reason. Because when you were talking, I was like, why can't you just use like, you know, Buffer or any other sort of like social, you know, media automation? But then you said this key thing, which I think— not a thing, but it's just true to many big organizations— is that when people develop processes they like to work with, it's actually much harder for them to onboard new tools and get them, you know, uh, accustomed to those new tools and develop new processes. Even if it's bigger, that IT transformation or digital transformation, as you call it, actually takes years. And big organizations will opt in to build custom tools to facilitate whatever the flows they have right now.

And now you hit the nail on the head. Yep. I wasn't very excited to build a custom tool, but they were willing to pay It worked. And it's exactly what you said. Sometimes getting the zebra to change its stripes is nearly impossible. Sometimes we're looking to build something that we're going to make money, they're going to be happy. But to your point, and Charlie Perrins actually, who we had a conversation with all the time, always challenges me when I talk to him on things like this. I was like, why would you continue to work in Excel? Why would you do this? We have those conversations. So what I try to do with a client is explain, like you said, well, we could probably use Buffer, or we could probably use this, and maybe they iterate on that going forward. And that's the hope, right? Like, this tool's working for you now, you're very happy with it, but maybe we come with a new proposal next year. Maybe we say there's a better way to do it, it's cheaper, it's faster. So it's a good point, Mike.

All right. So that was a fascinating conversation, Michael, and I'm sure our listeners want to know more. So if they wouldn't want to just you know, have a chat with you, maybe look closer into your work, you're clearly very passionate about QA and digital accessibility, and if they have some unique questions, how they would get in contact with you?

Sure, you and I are connected on LinkedIn. I don't follow LinkedIn too much, you know, I have been more and more, like for example for accessibility, Jenny Lay-Flurry, who drives accessibility at Microsoft. By the way, if anybody's interested in the best tech industry accessibility leader, it's Jenny Lee Flurry at Microsoft. LinkedIn's one way. I don't know, email? Anyway, how do you—

I wouldn't share an email in the show notes, you're going to get—

LinkedIn. Find me on LinkedIn. I have been checking it more and more lately. Especially since you and I connected and we've met, interested in the things that you're doing, Mike. So I'll be there. Message me on LinkedIn, connect with me. Weekly I'm checking my messages, so it's probably the best way. But yeah, happy to have conversations. I'm always learning new things from people who learn things from me, so just like we did today.

Excellent. Thank you, Michael. So anyone interested, just scroll down if you're watching YouTube or listening to the show notes, you'll see the links to Michael. And one thing we'll ask of Michael is to provide the list list of tools which he mentioned today, so we can list those as well. You mentioned, I think, Playwright from Microsoft. You mentioned some, some accessibility scanning tools. So I think we'll put this in for people so they can take those tools and do something today to start the journey into digital accessibility. And as well, thanks everyone for tuning in. As Michael shown us, digital accessibility, quality assurance, and automation are not just tools. They are actual strategies to make sure that your tech venture growth grows with inclusivity. So if you find the value in the show, please do subscribe. We are on Apple Podcasts, Spotify, I mean, pretty much wherever you're getting your pods from, as well on YouTube. Leave us a comment if you have any questions to Michael, you know, do that as well. We would love to have a chat with you and engage. And as always, until the next time, Stay safe and see you online.