The edge case is everyone

Most communication technology was built for an able-bodied world. For more than 11 million Americans who are Deaf or Hard of Hearing, a phone call can still be a barrier. T-Mobile set out to rethink that experience.

Willie King, Business Development Manager at T-Mobile, and Lex Liley, Associate Director of Product Strategy at Kettle, took the stage at Config 2026 in San Francisco to share how their teams partnered with the DHH community to challenge assumptions about voice communication. Watch the full talk above or read the transcript below for a look at inclusive design and why solving hard challenges leads to better experiences for everyone.

--

Willie: Everyone, the voice that you're hearing is Brandon. Brandon's not just my interpreter. Brandon's our interpreter, and he's here to assist all of us so our stories can be shared today. Hi everyone, my name is Willie King and I'm T-Mobile's Business Development Manager who focuses on accessibility.

Lex: Hi everyone and I'm Lex Liley, associate director of product strategy from Kettle, T-Mobile's design and agency partner.

Willie: What you all just experienced at the start, that confusion of us not being able to communicate and you not having access to what I was saying in the moment is the reality for over 11 million deaf and hard of hearing Americans every single day. And today we're going to show you how our teams have been working to change that experience and level the playing field not only for the deaf and hard of hearing but for everyone.

I work at T-Mobile, one of the largest telecommunication companies in America with a long-standing mission to weave accessibility into its culture and deliver inclusive communication to our customers. We began about 35 years ago with one of the first relay services for people who are deaf or heart of hearing. And today we're the nation's leading relay provider and deliver a variety of services that enable communication tools for those who are deaf, heart of hearing, deaf blind, blind, low vision or people who have speech disabilities. But there's still a ton of unmet user needs in the market today. So in the last few years, we've been partnering with Kettle where Lex Works.Kettle is a human- centered design agency on a mission to make product design highly personal.

Lex: When our teams began this project, we had a ton of excitement for wanting to improve the call experience for the deaf and hard of hearing community. And we knew we had to start from the ground up.So we began by learning about the tools that deaf and hard of hearing people use today to communicate and why those tools matter to them. And we began to ask ourselves, what does an equitable call experience look like between a deaf and hard of hearing person and a hearing person? Or perhaps to put it more bluntly, what if a phone call finally worked for everyone?

One of the things we learned was that features like text-to-speech and speech-to-text technology were some of the most effective forms of communication either via your standard phone features or via a third party app.

Willie: That's right. But even with these tools, there are a number of pain points that I and millions of others experience daily.There is always a barrier when communicating outside of our own terms. For example, when I want to call a business, I'll need to determine which app and communication mode I want to use or even if I need to coordinate with an interpreter. It's an extra step I need to plan for in what is already a busy day. And perhaps some of you are thinking, "But what about sending a text?" Absolutely. There are a ton of workarounds out there in the world today. But what about when you need to call your insurance company and work your way through a dialed menu or call your child's doctor to hear about your test results? The way the phone operates today doesn't support me in making those types of calls.

So, to help you understand a little more about my background and why I do this work, I was born deaf into a hearing family. And now, my wife and I are raising two deaf daughters. Communication barriers aren't theoretical in our house. They are part of our everyday life. Let's be honest. The truth is, we were not built for this world. And this world wasn't built for us. Everything around us, especially communication, assumes you can hear and speak. Every day, my family is working around systems that were never designed with us in mind. Some things as simple as making a phone call become complicated. I can't just pick up a phone and connect. There's always an extra step, a delay or a workaround. So, this is why designing better accessibility products matters to me. It's not just another app or another feature. It's a step towards a world where my family doesn't have to adapt to communication. It's where communication actually works for us. And that, that changes everything.

Lex: Building on Willy's story, accessibility means something different to everyone, which is why it's also important to design communication tools with more than just your primary audience in mind. For instance, my story highlights a very different user group in this space. My father has had partial hearing loss since the age of two, which incidentally is the age of my son. Dad has since gone through surgery over the years and now has a cochlear implant. And though we don't sign when we communicate, speaking to him at times can be incredibly challenging. He's from a generation where if technology fails him, his instinct is to just push it away. It's associated with failure and frustration.

But sadly for him, he's the one who becomes increasingly isolated. So even though he has partial hearing, something like seeing a transcription of a conversation during a phone call is an absolute game changer.

Willie and I both feel this problem in very different ways, which I think gave us a unique perspective when we approached this design challenge.

So when we zoomed out on this problem space, we started to realize this actually went beyond just the deaf and hard of hearing community. About one in seven people in the United States experience some degree of hearing loss. That's more than 50 million Americans.

We're talking about a future where what we know to be a typical phone call may become increasingly inaccessible for millions of people in this country regardless of whether they self-identify as having a hearing disability or not.

Lex: And that's when it clicked. This isn't just a deaf or heart of hearing problem. This is a both sides of the phone call problem. And it's growing for people outside of this community too. Suddenly, the design challenge became clear. What if the phone screen could become the canvas that made a seamless call possible for everyone?

Now, friends, you might be wondering, hm, I wonder if this is the point where they do a demo for us.

Unfortunately, our work isn't live yet, which is the reality, I think, for a lot of new technology that's trying to go to market because we have to move carefully and it has to be safe. We really want this to work for people rather than creating a whole new barrier for them. So, we don't have the finished product to show you here today, but instead, we would love to walk you through what we discovered.

Three key learnings from this process that we believe are truly useful for any practitioner in this room, regardless of your role or job title. And who knows, maybe something we share here today creates a little ripple effect out there in the world. And that's where learning one begins because being deaf or heart of hearing does not mean silence. Communication across this community is incredibly diverse. And so designing for that range meant we really had to design for more than just one type of user. Take my father for instance. He has a cochlea implant, so speaking in his own voice when making a phone call matters to him. So outgoing audio and speaker support were essential. But hearing the person on the other end is harder, which is where real time call transcription became critical. And allowing him to read the conversation as he speaks is also important.

Willie's different. He would prefer not to use his voice at all and instead type his response, which is delivered through text to speech. And like my father, he also relies on the transcription to follow the phone conversation. Same community, completely different needs. Some users want to speak, others prefer an automated voice, some rely on hearing with supportive text, others on reading. And those preferences can shift midcall.

So, it wasn't a question of which features do we build. It was how do we surface them on the screen without cluttering the space for people and allowing people to adjust their settings without leaving the conversation.Through user testing, we found three core inter interaction patterns kept rising to the surface. And what we landed on was a very clean call interface that allowed you to switch between one typing your response and it will be spoken to the other caller, two speaking your response and reading what comes back in real time or three hearing the response and reply through your keyboard.

These were simple patterns, but getting them right and making them feel seamless together was honestly one of the hardest design challenges that we had. It took many rounds of iteration to find that beautiful balance between flexibility and clarity.

Willie: Learning two accessibility is a practice, not a feature. Our design system was built with accessibility in mind, but real users kept surfacing needs we hadn't anticipated. So, we needed to keep systemically circling back to validate whether our features were meeting the use case. Dark mode is a clear example of this. Our first instinct was to simply just invert the colors. logical on the surface, but it left us with a black text bubble on a light background, which isn't true dark mode. So, we partnered with our design systems team to define text and background color pairings that actually hold up across both light and dark modes. Then, we brought in T-Mobile's accessibility team to pressure test the full range of scenarios our system would need to flex for. And it kept going.

Once we learned that certain color combinations, yellows and blues in particular, can meaningfully reduce the cognitive load of reading on screens, we started experimenting there, too. Something we're carrying into our backlog. The thing is, none of these were edge cases. They were fundamental to building an experience that actually feels intuitive. So the bigger lesson here, don't wait.

Accessibility of any kind, shouldn't be treated as a late-stage checklist item. For the deaf and hard of hearing communities, many of whom are older, features like dark mode and scalable text aren't just nice to haves, they're essential, and they need to be in the conversation from day one, not added in at the end.

Lex: And finally, learning three, design with your intended audience, not just for them. Now, I know we've all heard this one before, but hearing it and living it are two very different things. And such small shifts in your design process can make a world of difference. As a practitioner who doesn't experience these disabilities, I had to be really honest with myself. Insight and empathy are not the same thing. And even though I had my father to turn to, which really helped, my team and I had to fundamentally shift the way that we worked to meet this challenge properly. What made us lucky of course was Willie. He's not just a member of the deaf and heart of hearing community. He was the visionary behind this project. And having him in the room not only shaped the product, it changed how we operated as a team. For instance, cameras were always on and we would introduce ourselves every time we went to speak on a Zoom call because Willy's eyes were always on the interpreter. We also cut out any side chat in conversation in the side chat because we didn't want to create a cognitive load for Willie during the meetings. And what we found over time was that our users weren't this entity we periodically return to. They were there alongside us. And honestly, it made us sharper, more focused, and a lot more present. We were also really lucky to have T-Mobile's dedicated accessibility researcher who helped us cultivate a community of deaf and heart of hearing participants whom we tested with throughout the project.

These were people that went from suggesting new features to helping us fine-tune them and creating this really consistent continuity of change and improvement. Now, how you bring your user groups into your process might look a little different. You might want to think about co design sessions, usability testing, or simulating the call experience using people, using prototypes.

Because what Willie modeled for us was something we had to actively build into our process.

Our designers would literally pop into separate rooms and recreate the call experience by having one speak while the other type silently. And that hands-on approach is what led to some really excellent breakthroughs. This is Annie and Courtney who's testing the prototype of our project uh over Zoom. And they led the design of this work. And one session they had in particular changed everything. They noticed two things in particular. The first is that the deaf and hard of hearing person trying to keep pace by typing their responses found it exhausting. Not just slow, not just tiring, I mean exhausting. And second, the hearing person on the other end was sitting in silence. No: “Mhm,” or “Yep.” Essentially none of those sounds that say to you, I'm here. I'm listening. I'm still with you.

And we started to worry that perhaps without that they might hang up, become disengaged, or worse, act negatively towards the deaf and hard of hearing individual. So this observation led us to one of our favorite features, quick replies.

One tap responses that keep the conversation alive in real time. I'm using voice recognition technology. Please bear with me. Or, hm, that's interesting. Let me think for a moment. They were small moments of humanity that put the texture of a real conversation back into a phone call and what we hope set both parties up for success. So, whatever that looks like for your team, the goal is still the same. The method matters less than the mindset. Get close enough to feel your problems.

Lean into them and let what you find shape what you build. And hand on heart, the project experience moves from something you might feel very comfortable with to something that feels incredibly dynamic and fosters genuine curiosity.

Willie: So, there's another dimension to this that I wanted to share because it speaks to something we don't always talk about in product design, and that's the emotional experience of picking up something new. We put a lot of thought into onboarding, not just as a setup flow, but as a way to help people feel comfortable, especially those who might not be familiar with this kind of technology or maybe those who have had frustrating experiences in the past. We wanted our users to be able to choose the gender of their synthetic voice or go completely neutral to make the call feel like theirs before they'd even made one. So, our onboarding screens walked people through the core features in a low stakes way. So, by the time they were in a real call, nothing felt foreign. Call logs gave the whole experience an organizational layer. Past calls became searchable, reference-able, less like a communication tool, and more like a tool for life. The goal was simple. Make the system feel approachable.Make it feel less scary.

And here's what Lex and I would encourage you today to take with you. Even if accessibility isn't your core brief, treat the edges of your audience like they're the center. Because designing for accessibility creates a ripple effect that makes the product better for everyone in ways you won't always see coming. Our work isn't finished, but the more we design alongside our users, the more it evolves, takes new shapes, and surprises us. And honestly, that's the part we're most excited about.

Thank you!