Dr. Dee Miller on Designing Beyond Compliance

0:00From the Association of Registered Graphic Designers, this is the Design Thinkers podcast. Design Thinkers is Canada's largest graphic design conference. Originally launched to mark the conference's 25th anniversary, the podcast now reaches a global audience of designers and creatives. Join our host and RGG president, Nicola Hamilton, as she connects with Design Thinkers speakers to explore the work and ideas shaping today's design industry. This week on the Design Thinkers podcast, I'm joined by Dr. Dee Miller, Director of Product Strategy and Insights for Product Equity at Adobe. Dee is a researcher and innovation leader working at the intersection of technology, design, and equity.

0:42She's focused on turning real-world insights into products that are more accessible, usable, and culturally relevant. In this conversation, we unpack what inclusive design actually looks like in practice, and why it's not just about compliance, but about building better, more innovative products. Dee shares how her work at Adobe translates research into real product decisions, from improving accessibility features to shaping how AI tools like Firefly are developed and deployed. We also talk about the tension between speed and accessibility, what it takes to advocate for inclusive design inside large organizations, and why designing with communities is critical to building trust and accountability.

1:21This episode is for anyone working in product, design, or research who wants to move beyond theory and start building experiences that truly work for more people. I'm really excited for this conversation. I think that so much of your work is this space that a lot of designers are still sort of figuring out that intersection of like research and product strategy and accessibility equity I think a lot of designers like understand a lot of these principles in theory but putting them into practice or getting buy-in to put them into practice is a lot harder and I think you're actively doing that we are we're trying to here we are trying to it's not easy right um yeah that's the thing it's not easy and sort of before we dive into that like I want a little bit of background about you I want to make sure that our audience has a little bit of background about you so how did your career actually evolve into this space of research product strategy equity I've looked at your educational path and like you were you were building yes yes so let's let's it's you know it's interesting you ask this question um I went to like initially it's it's really a very interesting story in itself because I am a lefty so I write with love I pretty much do a lot of stuff with my left hand and when I went to undergrad school there was a program called science technology and society and it was really about how technology and science, you know, how society absorbs it.

2:58And they use the example of lefties. And I never thought about it. It's really of how left out I am in the design of technology as a lefty. And so I got really enthused about engineering design. What does this mean? How do you design? And I am the only lefty in my immediate family. So my parents are both right hand or like I learned from a right hand people to, um, it was really weird. My siblings are all, and so that is where I started trying to understand the constraints. How do we, um, bring in other voices that are not what quote unquote would be considered the norm. And then what really got me into this is my grandmother had diabetes and, um, she was using a blood glucose monitor and And she didn't understand the information on it and how it worked and things of that nature.

3:53And that is when I was like, OK, this is what I'm really into understanding and how to improve the overall experience, especially from a health perspective of like what information we put on there. How does it look? How is it accessible to people older? And that led me to saying, you know, as we get older, we lose certain abilities. And that is where I'm like, oh, there's this whole world. And in a part of my PhD program, Professor Grand Van der Heiden had a course on aging and disability, disabilities and aging. And I was like, oh, and that's when my career really started looking into that and bringing in the health aspect as well as just understanding disabilities overall.

4:41So that's where I got, this is where I am now. I'm pretty excited about being in this space. And I think it was really in that class where I realized that accessibility and the way we look at it is from the dimension of disabilities, but there's other layers to it of what access really means. And so that is where I am now. So I often say I'm product strategy and access. I'm taking it a little further than just the strategy, but how do we make sure it's accessible in multiple layers? Yeah, I mean, I think what you touched on there was this idea of we will all encounter access challenges at some point in our life.

5:21They might be temporary, they might be systemic, they might be as we age. Yes, as we, I mean, it's, and that's the thing, it's situational, right? And so at any time, even when, to your point, when you say temporary, when you get sick, sometimes you lose certain abilities. What does that look like? And it's understanding that whole, like, when it's temporary and when it's not temporary, what is that experience is supposed to be? And how you're not left out of the experience just because of a missing ability. Okay, so then maybe tell us a little bit about this role at Adobe, because a good chunk of your background was sort of on the research side, and now you're like actively implementing it and leading it and strategizing.

6:04So how do you sort of make that leap from academia into product, I guess? You know, it's interesting you say that because I remember in academia, I told my advisors and when they hear this, they'll be like, yep, she told us this multiple times. I do not want to go be a professor. I was like, my professor life will be when I'm towards retirement. I want to go in and do translational research. And that's what I get to do. I get to work with Adobe research team and product teams and translate research, but look at it through the lens of equity. When you say translate research, what do you mean, just for folks who maybe aren't as familiar with academic language?

6:44Yeah, when you say translate, so it's like, look at these research findings. So oftentimes, I'll take it to some simple, you know, medical research, oftentimes we have a cure, but it takes forever for it to come out. In this world, it's looking at we have different capabilities that oftentimes we're not putting into the product. And so what my team is allowed to do, for example, if we're looking at how do we leverage AI to improve alt text, we're now looking at that and saying, how do we translate that and productize it into something, into our product so it can auto generate. Similar to the work in Microsoft.

7:25Today, if you go to accessibility, it gives you the option and ability to add or generate alt text. It's taking that technology and making it useful for the user. And so sometimes it's in a lab, it's in a bubble, no one's really looking at it. I get to work with the teams and say, okay, what does this mean for the user experience? How do we make sure it's accessible? And work with the product team and say, how do we implement this into the product? What is that experience? And so that's what my team gets to do. We love it too. We get to look at accessibility through the lens of usability, design and innovation, and really take accessibility from the compliance realm, which a lot of companies and everyone does, and bring it to the quality realm of things.

8:16I'm really saying accessibility is about quality of the product itself. So that's what I'm excited about. I do want to talk about this idea of compliance versus actually enabling innovation or access. But first, I want to stay on this role at Adobe for a second. And I think the idea of product equity is something you talk about a lot. And I think one of the things I really want to get out of this conversation with you is what do some of those things actually mean? And how do some of our listeners start bringing some of that into their work likely at smaller organizations. So what does product equity actually mean inside a company like Adobe?

8:52Yeah. So product equity is ensuring that our products work for the broader spectrum of human experience, identities, and abilities. In our world, from an Adobe perspective, it's about removing barriers to creativity. If, you know, Adobe's mission is empowering everyone to create. Our mission is to remove the barriers to creativity for people to create. And so what does that look like? Particularly on my team, we're looking at two main areas. That is from an accessibility, making sure our products work with people with disabilities, as well as culturalization. When we expand our products and our base and go global, what does it mean to have a culturally accessible experience as well.

9:38So that is really what we are looking at when we say product equity on those two realms at Adobe, removing those barriers to make sure everybody's able to use our products. Are there like specific projects or maybe innovations you could actually talk to us about that your team's working on or has pushed recently? You know, I mean, like if you look at some of the work that Firefly does, my team had a very much impact prior actually to me joining under the leadership of my manager at TB. They were able to look at Firefly and look at how do we reduce harm in what images are created or generated. And the way they did that was to increase the amount of content we were obtaining that were diverse.

10:26So we know, so oftentimes I often say for responsible AI, you really should be questioning the models, what data they're learning on, and really understanding that. Well, my team was able to also say, okay, looking at this, these are from a human's perspective, some of the criterias that we should be looking at, and then also figuring out how do we source that content in that the models can be working off and learning while a user is inputting a prompt. And so things like that is what we've done that's out there today with Firefly. Now, I'm not saying it's like the best, you know, there's things that we have to, we're still improving as always.

11:08But a lot of the images you see today is because of the work that my team did in helping our content team source additional content from a diversity perspective. Yeah. I mean, I think that that's such a fascinating example of inclusive design that isn't necessarily about compliance, right? Like nobody was like, tick this box that was about innovation or a recognition that the tool was doing more harm than it needed to. Need to do. Yeah. No, I mean, you know, it's interesting because it's like, what is the definition of harm, right? when it comes. It's big and broad, like, I don't know, right? Because it will, it depends on, it depends on the ecosystem or the environment in which you're talking about harm, right?

11:50Like who's it harming and who's worried about who it's harming or what is it harming? Right. And that's the thing. So oftentimes, like, so when I think of product equity and, and, and this is, this is a dichotomy that I even talk about, like even for accessibility, there's those that create, there's the creators and the consumption. And when we think about harm, there are people who are going to consume what you're creating and then there are people who are creating and so it's oftentimes important what my team look at is like who's the creators these are the these are simple questions is there anybody missing from the table like what do we need to like those are that that's a major question what model are we using what data was used in this model even looking at that model what additional data points do we need to add to it Harm is an interesting, like you said, it's large.

12:44It depends on the concept. Amorphous. Amorphous. It really does. But something as simple as, I'll bring it back into accessibility. One of the things about accessibility and AI that I love and I question also is I feel like AI has accessibility can tremendously benefit from AI. However, I say all that to say it also can continue the bias that is often ignored in current data that it's learning off of. And so I'll give the example of alt text. Alt text presents a very interesting example because I love to use AI because it can help me generate alt text very quickly. And so now I know that all my images will have some type of alt text.

13:31However, what is the alt text describing? How is it describing it? doesn't know of the nuances of what culturally is acceptable or not acceptable. So things like that is where it's like, if it's learning off of data that is already has bias, and yes, we now can automate all text, it still can cause harm because of the way it describes the image in itself. So those are the things like an example where I love AI, I think it's great. And it's also an example where we got to pay attention to like what it produces. I imagine you're spending a lot of time as a leader, like actually thinking about that balance, right?

14:11Is it net positive or not? How much positive impact is it having versus what are the obstacles we still have to overcome? Yeah. And it's a hard thing because it's like, like I said, AI has the ability to help us scale. And that's a good thing. But it's like, how much is the scaling going to harm? So it's that balance that we're trying to figure out. And sometimes it's us making decisions that, hey, we may avoid talking about certain human characteristics like skin tone and the description in itself. So we're building the model, but we're making sure the model avoids skin tone. I don't know if that's a good thing.

14:52Like, these are the decisions and conversations that we're having where we're trying to make the product accessible. we're trying to make it available to screen readers and so they can understand but then we're like oh but can we really describe it at the level that a person that can see is going to have and understand what's going on so like are we giving the same experience that that is something and that's that's true equity right like that is true equity that is true equity yeah yeah and it's really hard to do that without without human interference without human sort of being able to add that nuance it is very hard to do without human interference for sure talk to me a little bit about some of the things you're sort of like really thinking about or working on right now like are there projects or topics that the team is like really rolling around yeah I mean the thing is is about where you know adobe is uh about creativity for all and so what that that in itself is a major mission so we always are thinking like what does that mean how does that show up um and particularly it's like as we look at the next generation like what does that mean um and this is the thing we're in a weird situation we're in this next gen technology is moving fast everything's going that scale.

16:14And then we have currently people are just trying to figure out how to use the current technology. Everything's changing. AI is changing what creativity really means. And I don't know if that's truly true. I won't say that because I do believe that someone asked me the other day, what are the skill sets or human skills you still need with AI? I haven't seen AI take away creativity. I think it augments and it helps and assists, but I don't see a takeaway creativity skill set. I still think that is with the human. I still think critical thinking is with the human and things of that nature. Yeah.

16:53So in saying that, it's like, what is the experience now then when you're thinking about Photoshop, Lightroom? What does that really mean in the realm of AI as well as making sure it's accessible. And what we're trying to do is how do we meet it? I think that the projects that we're looking at is how do we meet personalization and customization at scale? Because we got so much, right? And then the other project that we're - Is that kind of what Adobe's working with as like the sort of future? Like there's a couple of schools of thought, right? For like, where are we going in our tools? Is personalization sort of one of the paths that Adobe Chase say?

17:37I mean, there's a whole thing out there already from Adobe. We talk about personalization at scale and what we're really looking at. So yes, that is an area. I mean, that's an area of where everyone is going. Like everybody wants it to be personal. I don't know if that's the, so I always say personalization and customization are different. And then like I can customize and not personalize, right? Customize is more like everybody can have the same thing to some extent, but personalized is like really understanding human behavior. And what I will say is AI gives us some of that, but to what extent, how far do we go with personalization that it doesn't make it hard for other things.

18:19So I'll give the example, like if you personalize too much, am I going to be able to call support for help? Yeah. Are they going to know what my tool looks like because I've personalized it so far? Like these are the things, these are the complications. But yeah, Adobe has an article out there that talks about what is like the future of personalization in itself and like what it looks like from a customer experience perspective and customer orchestration perspective. I think personalization is going to be very important as we look at marketing. Think about when you get something in marketing that doesn't match what you are, who you are.

18:59You're like, why are they sending this to me? So, you know, it just takes target marketing and things of that nature onto another level. Can they do it without harm? And that I can't really say yes or no to at this time. I mean, it does make me sort of so much of the technological innovation. It's a process, right? Like we're not really going to know where we are until we're there. And that makes it really tricky. But I guess it's like, it's a, you know, it's tricky, but it's exciting. I think that that's the thing that I really like is that, yes, it's tricky, but it really is exciting to see where we're going.

19:45And, you know, it's always interesting because I was talking about this to a friend recently and I was like, yes, gender of AI, you know, where AI is, is new. But I've been working with AI forever, you know, prior, you know, like AI has been around. It's just that everyone's catching on to it. It's accessible to more people. And so that's where it's like, but that's also where some dangers comes into play. And so from an Adobe perspective and what we see is like, how do we, and particularly my team, remove barriers but still leave the human in control? Does that make sense? Like, I often don't like the word human in the loop.

20:33In the loop. And the reason being is because if I go back to even inclusive design, you can have human in the loop. Human in loop often means I ask some questions to some people within the community. It doesn't mean that they design with you. It doesn't mean the with versus for. And so I often say human in control because there's a need for the human to be in control. And especially as we accelerate AI overall. That designing with instead of for comes up often, I think, when we're talking about inclusion. And when we're talking about inclusion in lots of different ways, like not just the like technological access piece.

21:15And I'm curious from you how, you know, you're at Adobe. It's big. There's more money. There's more time for things. Designing with becomes easier. Have you come across any opportunities, like small ways for designers, agencies, teams to design with communities or with audiences that maybe don't cost as much cash? Like, I think that's one of the big barriers. Like, I think that's one of the reasons that a lot of these studio spaces designers are still sort of treating accessibility as a checkbox is like to be able to do it properly with like proper intention. Yeah, it takes it takes time and money.

21:53Yeah, no, I think I think you're absolutely right. I think one of the things I often like share as it relates to I think that what's interesting, you're hitting on one of the major barriers to accessibility, access to disabled users. Teams may not know how to find or recruit disabled users for research and testing. And so oftentimes one of the things that I say is first look inside your org. You know, even if it's a small organization, if you're an independent researcher, think about partnering with different nonprofit organizations that you can work with. But within the organization, you have people that have certain disabilities in itself.

22:35At various companies I've been at, I look at the ERGs. I see how we can do research within the company itself. Now, I'm not, as a researcher, I say this isn't a way to use, but I still want you to find someone external outside because you have the bias that we bring in working for the company itself. But I do say that this is a great way to start looking at the research, like trying to do research and building with communities. So partnering with disability organizations, a lot of different organizations, especially locally, can help you in that. And then I also say there are specialized platforms that have free tiers initially that allow you to get on and you can find people that way.

23:29But mainly it's just really going out for those employee resource groups. And then I also say, take a step back as a designer and just do some simple, some real simple things that some people don't even think about when they're designing. It's like, zoom your design to 200%. See how it looks. What's that? What does that look like? You know, like there's things you can do as a designer that is helpful as well. Just that simple. Try to use your design with keyboard only. That in itself will improve access. Like you don't always need, although I do want you to research, I want to say with people with the disabilities, sometimes you can start that small just to get, just to improve accessibility before you even reach out to the nonprofit organizations or your ERGs to help and support you.

24:28A lot of times, even within small companies, people are willing to just raise their hand. And I've been at a company that's particularly small. And I say, hey, I want to do this study. I want to do a study with people with disabilities. Are there any people that you know in your network that would want to participate? And if you can get one to two people to participate, you actually get a lot of information. From a usability perspective, and I believe accessibility is good usability, all you need is five people to find 80 % of your problems. So you can get one to two, two to three people to test out your design.

25:12You're halfway there. And I want people to also understand it's a journey. so don't feel like you have to get all the you know address everything it's always it's an iterative process and it's a journey I think so much of it too is is the sort of curiosity or openness to what you might uncover too right you don't know what you don't know you are absolutely right and I think that's it's interesting I love the way you phrase it as curiosity but I also think it's a fear too of not knowing what you know so you're like if I don't have to open that can I'm not going to like let me go over here and continue working so but the curiosity is what gets us to and that's why I want people to look at accessibility as experience quality and not as just compliance that's that shift that needs it's about and for designers I often talk in my talk really think about it from compliance to empowerment design.

26:10So from compliance design to empowerment design, you're empowering somebody. And I think that takes on a different, for me as a designer and a researcher, that takes on something different. Oh, I'm empowering somebody. I want to help. I think that attitude probably changes the way you approach communities too, right? Like I think one of the big fear factors for some folks is also why would this community trust me? Why would they want to participate in this, right? Like how do I, How do I build trust to actually get people to communicate with me in any like real way?

27:05You can seek advice and participate in our Slack workspace, meet regularly with others in your niche through our virtual communities, become a mentor for an emerging designer, attend free weekly webinars, make friends at our local events, and so much more. While the RGD is a Canadian association, you don't need to be Canadian to become a member. Find out more at rgd.ca slash membership and join us in our mission to build a graphic design profession that is broadly valued for its contributions to life, commerce, and society. Yeah, and that is really, it takes time and consistency. And so I think it's really important to build into your overall process.

27:47One of the things that we do at Adobe and I've done at other companies is we go back to those researchers, to those participants and let them know what the findings were and what's the plan. That is the trust. That's that time and consistency. It's not just doing the research and saying we're going to do something. It's really communicating and being transparent about what the research led to and what we're able to do. And there's going to be things that we're not going to be able to do, but just being honest and transparent is helping you build that trust because trust takes time and consistency.

28:27That kind of transparency about like, here were the findings, here is the plan forward. the like responsibility cycle of that is so interesting because then your team would care so much more about actually being able to implement those changes into a final product. Because I think that's one of the other things, like sometimes you can do this incredible research, find these insights and like you can actually ship them. Like everything just sort of gets lost in the mushy middle somewhere. Yes, you're absolutely right. And I, I think you're, you're, you're pointing out, it holds us accountable. Yeah.

28:56Yeah. And that's literally, and that's what But that's one of the reasons why I ask our team and other people to actually go back to the community for that accountability. Because, you know, companies don't want bad reputation, right? I always say there's certain risks that you take. And these are the risks. This is one of those risks that I'm like, reputational risk is not a good risk. Once it gets out and it just goes on its own, we have to really think about that. So that's why I say keeping the, it's not designing for, it's designing with from the very beginning of like including them to the very end of like when we release a beta, they're involved in the beta, giving us feedback.

29:42Do we address everything? Letting them know what we couldn't address. Because sometimes we find things and because of the tech stack or other issues, we're not able to address it at that time. But we can at least share that information and continue to get feedback. Yeah, I think that's so interesting. Yeah, the accountability piece makes such a big difference. It makes it really makes a big difference. I think like I was reading something yesterday on LinkedIn and they were talking about because in the United States, like ADA compliance, Title II is supposed to go into effect in April. And they talked about how accessibility has been talked about at an exponential forever.

30:23Like, well, I should say more recently in these last two years since that ruling came out. And I was like, because people know they're going to have to be accountable for like, they've had two years to get things right. And here we are. I think, unfortunately, some of that is critical mass, too. Like, I think some of that is the sort of aging baby boomer population, right? Like, what is it? It's like something like 2035, a massive amount of our population will be. When the baby boomer, yeah, will be a certain age. Agree. Yep, totally agree. And I think that's what's pushing it, right? And so they won't be able to use your website.