Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
/empathy: the importance of feeling heard
60 Plays6 months ago
Yum! Brands UX Manager Erica Price joins Aaron and James today. They talk about using empathy with users, brand managers, and subject matter experts to make sure everyone feels heard. They also discuss user-forward design, UX testing, and and how AI can support all of this work.
Recommended
Transcript
Empathy in Collaboration
00:00:01
Speaker
Once you start changing your tune and you empathize with the people that you're trying to work with, the conversation becomes a lot more, a lot easier, because the moment that the ah brand feels like you're not, like your motivations don't align with theirs, that they're going to feel threatened. And that is a really challenging, that's a really challenging thing. So you have to like put in the work. You have to be like talking to the brand, talking to the people that you've lost the trust with, and you have to dive in to like what is motivating them and how do you pair it back
00:00:41
Speaker
to them so that they feel like they're being hurt because ultimately we all have the same motivations. It's just our delivery could vary.
Introduction of Hosts and Guest
00:01:02
Speaker
Welcome to the forward slash where we lean into the future of IT. I'm your host Aaron Chesney with my beautiful co-host James Carmen. And today's guest is Erica Price. Hello, hello. And Erica Price is a UX Manager at Yum! Brands, where she leads the design of user-focused systems across the restaurant industry. She co-founded Sci Summary, an AI tool for simplifying scientific research, and Helix Booking, a wellness booking platform. Erica has
00:01:39
Speaker
over a decade of experience. She's passionate about creating inclusive digital experiences as a teacher and speaker at memorizely. Did I say that right? ah She shares her expertise in UX and product conceptualization, UI design and accessibility, helping others build user first products. So welcome, Erica. Good morning, James. I feel like she You know, wrote that bio just to mess with you. There were some some parts there that were tough. Definitely some 50 cent words in there. That was the whole goal. To be honest, chat GPT wrote it. so User focused systems across the restaurant. And there was a lot of R's in there and a mess you up. Yeah. theres that's fun Thanks a lot, Eric. Okay.
Overview of Yum! Brands
00:02:30
Speaker
So just because I'm from this area, some people may not know what that is, but Yum Brands.
00:02:37
Speaker
What is Yum Brands? Yeah, so Yum Brands is one of the one of the largest quick service restaurant companies. They are the parent company to talk about Pizza Hut, KFC, and also a smaller ah smaller one called Habit Burger. So they essentially are that parent company that really holds all everyone. And we build technology.
00:03:07
Speaker
to support these larger larger brands, basically. You know, I need to pitch them my idea for a new restaurant on the moon. You know, great food, no atmosphere.
00:03:25
Speaker
and don't Alright, since it's Halloween, I'll tell one of my, I think it's my favorite, my wife's favorite dad joke, you know. sir Skeleton walks into a bar and he says I'll have a beer and a mop. It doesn't it does it's not landing We'll think about the same reason a same reason that a skeleton didn't cross the road He's got no guts hi Yeah. So you can think about a skeleton drinking a beer, right? And then it's all over the floor. He has to mop it up. I see. I see. My brain is not in logistics mode.
00:04:08
Speaker
Too early for that. Not not enough coffee for that. All right. of the Of the brands that Yum Brands Owns what's your warfare? Where's your where do you go to get lunch or dinner or whatever out of out of the options that you have? So I am a I would say a strong Taco Bell like person when I was pregnant I went to Taco Bell Every single day for the last like month and I got a potato griller which they have now It's not it's not a thing anymore
00:04:40
Speaker
But it got to the point where the employees knew my name and I knew that that wasn't maybe the best sign, but I blame it all on the pregnancy. that's fair Yeah, I went to talk about until I looked like I was pregnant.
00:04:53
Speaker
i mean Let's be honest, the price is amazing. The food is pretty good and they've really upped their quality. And their brand, man, they they are they're doing really well with their branding.
00:05:09
Speaker
I think the habit burger, if you've never had it, Californians would probably say they have, but habit burger would probably be my next one, but I've only eaten there like four times, so.
Challenges in Brand Expansion
00:05:22
Speaker
Well, since Young Brands is from, I live in Northern Kentucky, since they're from Louisville, Kentucky, and the way we say it here is not Louisville or Lewisville, it's Louisville. It's like you swallow your tongue, it's Louisville. ah It's from Louisville.
00:05:35
Speaker
ah hopefully they'll bring a habit burger to this area and I'll have to check it out. Cause I hadn't, I don't know when the next time I'll be in California for anything. It's very good. I really hope they do expand. I think they're on the, they're on the like trajectory. um It's actually surprising when they first bought habit burger, everyone was a little confused because it's just a much smaller comparing it to the large giants of KFC Pizza Hut and Taco Bell. So it was really, we were, I was like surprised to even find out that Yum had habit burger inside of their sweet, but now they have all of the, or not all, i they have all the different types, the most common types, burgers, pizzas, and tacos. Yeah, it's it's interesting how long it takes for certain restaurants to franchise out into other states. Like it took us forever to get a Chick-fil-A in Michigan.
00:06:32
Speaker
So and it' just um even though like in in it's almost like they they start expanding from like their epicenter, like wherever they started, and they just slowly kind of expand. And and I'm not i' not quite you know sure. like Like for instance, you said you know it's ah it's a California thing. like Because we're in the Midwest, it feels like it takes forever for it to get this far east.
00:07:01
Speaker
And there's lots of rules too, like franchising anything. Um, there's a lot of brand, like they want to make sure that it's not just popping up everywhere every two seconds. So it's almost like there's lots, limited amount of franchisees. So you have to like find the right place, the right people, the right market. Oh, but there's only a certain amount of.
00:07:28
Speaker
franchisees available or allowed in Whatever state ah and it's really interesting. It's actually a fascinating space the space of franchising Restaurants because I didn't know this there were so many rules that you have to follow if you franchise with a larger org and There's just like just so many rules like specifically. I think my husband and I are looking at like Culver's um And they They're like, you have to spend at least X% on marketing ah every month. And I'm like, wow, that's very granular operations. So I feel like my understanding of what a franchisee owner is like really changed when I was like book and that just looking at it. So it's fascinating. It's hard work, though.
Empathy Training at Yum! Brands
00:08:20
Speaker
Oh, my gosh. Owning a restaurant, man. And being a team member in the restaurant, that is ah is some hard work.
00:08:29
Speaker
Absolutely. Yeah. I have been an employee of, well, they weren't young at the time. It was PepsiCo at the time, but they spun out young, but yeah, I was, I was an employee. I worked for Pizza Hut, slinging pizzas for many years. So as a, as a younger person, I guess everything I've done in the past is when I was a younger person by definition, but you get what I mean. yeah I actually did. I actually worked at Taco Bell twice.
00:08:58
Speaker
Oh, nice. I still love their food. That's a testament right there. you're right I have i've ah never worked officially, but I have done training. What's cool about Yum is they make you do training. So you'll go in store and you'll train um with all the brands.
00:09:19
Speaker
And it was it was honestly, I think, really, really fun and really eye-opening. But it's really cool that they like force you. It's like, you need to go empathize with the people that you're building products for. And training is the best way. I think it's a it's really eye-opening experience, too. I've never worked in like, I was a waitress. That was my closest food. um But it was really fun working the drive-through. If I did that, man.
00:09:48
Speaker
and's That's where I would live, is to live in the drive-thru. You've mentioned that you've had some involvement with in-store systems for you know different of the the different brands. is Is that part of your role as you know developing products for in-store? What all is the scope and the breadth of what you do for companies? I do a lot, I would say. I have been moving around in many different teams where I was ah first actually on our menu management team, which essentially is the the person that I was designing for is a data manager. So the person who is ah the one who's in inputting and managing all the complexities of national menus and even store customizations, because fun fact stores can do almost anything that they want.
00:10:43
Speaker
up to a certain point, and they can price anything to what they want. So like the data manager's job is to like manage thousands and thousands of stores, basically, and all the requests that franchisees have. So like that was like the first team that i was um that I was on. And then I went over to the point of sale team.
00:11:01
Speaker
um where I got to do a lot more like KFCUS, Taco Bell tacobel u s experience design, which was really awesome. i really My eyes opened into what we call service design, which essentially is how the environment that the users inside of the store impacts their experience so intensely. And it's fascinating because you can actually ah make an improvement by like changing the placement of your store or changing the placement even of a button and that could actually encourage or discourage like forgetfulness which is quite quite fun. So I did that for a little while and now I'm actually back over in the data management system we call that admin portal and I'm also ah very close with a project called tracks, which essentially is the inventory and labor management. And then we have teams for what we call the super app, which is more of like the day to day operations. Everyone has an app
00:12:05
Speaker
and they can like post and get routines. And it kind of walks them through, handholds the team members through operating the store efficiently and effectively. So I really touch, I would say, everything in some way. um And it's also a logistical fun thing to try to like bring everyone together just because it's just inherently challenging. like we have people all over the world. YAM is really cool because most, I think most everyone is remote, specifically in the technology space. We're all over, like we've got people in Vietnam, the UK, and I actually even work with ah the consumer facing brands. So that's going to be the the dot .com. So kfc.com, talkabelle.com, like the apps, like I know the teams and we're slowly building out a centralized user experience model.
00:13:03
Speaker
where we could where we could like more effectively speak to each other and understand what everyone's working on. So then we can be influenced and hopefully create experiences that
Balancing Brand Uniqueness with Standards
00:13:14
Speaker
are similar. So a team member, whether you're a team member or someone above office, you feel like you're working on a product, a young product, kind of like Adobe. Like Adobe is a single brand and they've got lots of different apps.
00:13:30
Speaker
I think that's kind of like the direction that we're trying to to head in, especially because four brands, they're very different tacos, pizzas, chickens, but there is a lot of overlap. And so there is efficiencies, I think, that we can build. And that was kind of an aside. So yeah, I touch everything, I would say, mostly. you can You could say you've touched all of our food.
00:13:55
Speaker
I have touched all of the food with my little grubby fingers. So what was maybe not so finger-looking good. at each So what was it like the conversation of, you know, trying to convince folks that, yeah, I mean, because we face the same thing and, you know, just software development, right? But ah from a product and experience standpoint,
00:14:19
Speaker
trying to convince people about that overlap. You mentioned there's overlap, but everybody everybody everybody thinks they're they're the snowflake. Like, oh no, no, our brand's unique. Our customers are unique. Our experience is unique. yeah there's There's nothing that pizza has to do with tacos. There's nothing in common. How were those conversations? Did you get a lot of pushback? Did you get a lot of like, oh yeah, I can see that. Well, how did that go? Good question. ah Lots of pushback, still pushback, to be honest. Okay. And I just want to say like,
00:14:46
Speaker
Yes, there there are snowflakes in their own right. Technically, even in the data model, like the pizza and chicken in and tacos, they all use similar terminology, but they are different. And so I think that difference is what makes each brand special. And our intent is not to wipe that away. I still think Taco Bell has a kick butt branding and they're they've got a really specific audience that they're getting toward. And so they can do more fun, like crazy, crazy experimental things because it makes sense for that market and makes sense for that space. And the that the rules might change inside of pizza and chicken. And so I'm not invalidating and that, I would say.
00:15:38
Speaker
But I think that 10% of customization is what really makes a brand stand out. It's what makes them successful. And so communicating that piece, if it's like, hey, I'm not here to wipe away what you have already created. I just want to take the things that are overlapping and I just want to optimize them. Like let's share the knowledge. We all have stores. We all have drive-throughs. That's all similar to language.
00:16:09
Speaker
Where are their optimizations? Taco Bell is doing it really well for this instance. KFC is doing it really well for this instance. Can we learn from each other? It's more of a group effort. And I think that was inherently really painful at first, because of course it is. I think any white label solution where you were working with many brands, there's going to be this fear of like, oh, well, they're just going to make me a generic brand.
00:16:38
Speaker
And I don't think that's the intent and it should should never be that should never be the case. I think it's up to us as people prescribing the experiences to make strong correlations to what is shared, but also make very strong like customization recommendations. So what makes a brand a brand? What makes them special? How do we keep that? And I think changing the tone, making the brands feel like we weren't trying to, we're not trying to become yum. like We're not going to make KFC rebrand to yum. like We want them to be special. like they are They've been doing it for a long time. They're doing really well. We want to keep that.
Collaboration through Empathy
00:17:22
Speaker
So it's once you start changing your tune and you empathize with the people that you're trying to work with, the conversation becomes
00:17:33
Speaker
a lot more, a lot easier. And also it's really important to set expectations with those types of conversations and goals too, because the moment that the ah brand feels like you're not, like your motivations don't align with theirs, that they're going to feel threatened. And that is a really challenge. That's a really challenging thing. The first time it happens, it's really hard to get out of it. So you have to like put in the work. You have to be like,
00:18:02
Speaker
talking to the brand, talking to the people that you've lost the trust with. And you have to dive in to like, what is motivating them and how do you pair it back to them so that they feel like they're being hurt? Because ultimately we all have the same motivations. It's just our delivery could vary. And that's even harder for a larger organization because they are, we are not just speaking to one person.
00:18:27
Speaker
We are 1,000 people speaking to 1,000 other people. but So that word of mouth is very challenging. And honestly, I don't think that's ever going to go away. I think that there will always be that challenge. And I think that's true for anything that's white labeled. When I was at Brad Financial Airlines data at the time, it was the same thing. They they didn't want an experience or a product that was generic.
00:18:54
Speaker
That was not limited. Sorry. And you have a lot of tradition with these companies too, right? Because they, they're, they're not like new kids on the block, like as far as, and I don't mean the boy band, but the, you know, the these, these brands have been around for quite some time. Um, like I remember as a kid, all of them being around and, and still, so they're at least, you know, 50 year old brands.
00:19:24
Speaker
Um, not that I'm 50. Uh, well, yeah, wait, I am. Um, but, uh,
00:19:35
Speaker
so you know, so you've got a lot of tradition there in, in breaking down, you know, those traditional barriers has got to be an issue too, right? Yeah, it's huge. And and like, there are franchisees who have contracts written with the Colonel, like KFC with the the Colonel himself.
00:19:54
Speaker
his signature on a napkin, like that's still a valid contract. So I think it is a a challenge. And Yum, we are much newer. We are young. like We are technically the parent company, but in reality, like we are learning from them. And I think when we put that hat on of like KFC, Taco Bell, Pizza Hut, like you are the experts.
Franchising's Impact on Strategy
00:20:22
Speaker
like let's work together, teach me all the things, and I'm going to be the bridge between all of the brands. And I'm going to try to spot these efficiencies that other brands are doing. And we're going to just see if we can't work together ah to do that. So it's an inherently very challenging relationship, but I think it's a very worthwhile a worthwhile endeavor to to work through that. i'll I'll give you a freebie. Get them to have a CoLab product. That's it a chicken taco pizza that all three of them carry. i That's so funny because that like literally comes up a lot. like but like It would be amazing and I could get it at any one of the restaurants. It's all trademarked. They have to get
00:21:14
Speaker
They have to get the chicken from KFC and they get the taco stuff from Taco Bell and then they get the pizza stuff from Pizza Hut. And they're all working together, feeling each other's pain. We do have co-branded items, but I really wish we could like build bundles for those co-branded stores to be like, would you like a bucket with your pizza or your bucket ah to a double taco with your pizza? with your chicken sandwich? That would be a great thing for families like mine. Like I have, I have twin boys and they, and I call them ying and yang because they are so polar opposite of each other. So one likes Taco Bell, the other one likes pizza. So it's like, well, what do you want? Well, I want pizza. I want tacos. And it's like, and we're kind of stuck in the middle. It's like,
00:22:13
Speaker
maybe I want chicken. So it's like kind of one of those things where it's like, if you, if you could do all of that from a one pickup thing, maybe like, you know, that that would be amazing. Cause then you don't have to run to three different restaurants to please three different appetites. There's, there is a, that would be really awesome for the customer. And I'm just like, I'm imagining all of the logistical problems of that way. Oh my gosh. So many logistical problems.
00:22:42
Speaker
going back of a store though, my gosh, to have yeah all of them in one. It would be like a food court. Yeah, basically. yeah Food court in a box. You'd have the yum food court. Yes, yum food court. Maybe that's the future. I must say I love the the the Taco Bell, like the co-branding stuff they do with like Doritos, with the Flaming Hot, with the Uh, now they got, what is this? The Cheez-It thing. I haven't tried the Cheez-It thing yet, but this like monster Cheez-It, like, how do I get a box of those? Like, yeah um I want, I want to eat just the Cheez-It itself. yeah I'm constantly impressed with what Taco Bell and honestly, like their tests, the innovation kitchens are like, that's so cool. Like that's, it is so fascinating and I love listening to.
00:23:34
Speaker
just like the things that they're, they're creating. And I'm like, wow, there is, they put so much energy into that, which I think is amazing because that newness I think really keeps the brands relevant. Well, how, what is it like to support from a, when you worked on point of sale system? So I would imagine a menu like Taco Bell, literally Pizza Hut and and KFC probably don't, and maybe a little bit more.
00:24:04
Speaker
Uh, I would guess that Pizza Hut is probably the most static menu where, and then KFC might be second, but Taco Bell has got to be just a hot mess to try to keep up with from a point of sale system standpoint.
Designing Universal Systems
00:24:15
Speaker
is Is that true?
00:24:16
Speaker
So it's funny, I think they almost all are equal in my in my eyes, just because they all have different, it just depends on the problem, right? So Taco Bell generally, Taco Bell has a lot of customizations, like they are, and they're, but they're generally like they're probably the most used thing is like, I'm going to swap out ah ground beef for beans, like that's a really, that's like pretty popular. So like, that's very specific. And this meat plus bean, yep. From the old, from my old register days, I remember hitting that a lot, you know, oh my thirty you know, plus bean. Yeah, exactly. And so like, but then, but then you have KFC, which is like bucket and most likely, they're going to be changing
00:25:03
Speaker
chicken types and chicken flavors. And then for pizza, you're going to have the complexity of a half and half pizza where you have and also like the weights of it. So to me, it's like, everyone has complex modifications, you just might have more weight on it's either agree, it's ingredient swapping, it's flavor swapping, it's side changing,
00:25:29
Speaker
It's like they're so they're all kind of like modifications in my brain, but they all kind of have it's just their complexities. In their own right which makes it challenging because if you're creating a singular point of sale system. You have to account for the extremes in all species because that's like the one of the big things that we've been.
00:25:51
Speaker
trying to do is create a point of sale that both Taco Bell, Pizza Hut, and KFC can use. So we're white labeling it. um And it's fascinating. It's a really fun problem, because we want it to feel like the brands, because obviously the team member is working for that brand. So we want to make sure that we're flexible for that. We want to make sure that the modifications are flexible, just because each brand is going to have different needs. And it's It's kind of an interesting challenge because we have to weigh these requests constantly. It's like, okay, is this a KFC ask or is this a entire everyone ask? Cause that's going to change greatly how fast or what methods we use to update it. Cause there is this like, there is this push and pull of like customization that we want to keep, but we don't want it to be too complex.
00:26:49
Speaker
Uh, so yeah, it's just like a balance, but I would say that it's, for me, it's like all of them have their very specific needs. So you're going to run into similar pain points, but they're going to be in different subjects and different parts of the point of sale, if you will. So on the, on the point of sale screen, where is your, like your, your prime location? Cause you talked about being able to find things easily, um, earlier.
00:27:17
Speaker
And I know that when I was, um, a cashier at Taco Bell, that one of the things there was always that one modification that was like, how do I do that again? And it was like, you know, Oh, you got to press this and this and this. And then that I'm like, why isn't that just a button like the other ones? Right. Um, so how do you like de define your prime real estate for these different brands? Cause obviously like you said, there's, there's different, you you know, feels overall, you've got kind of your build your own, your customize what's there. And then, ah um you know, this is how I want it assembled type of thing.
00:28:00
Speaker
um Yeah, I would say that's a big challenge real estate in general. And honestly, it's fascinating because every, so you might say like, okay, this happens six times a day for me, or maybe like 30, 40 times in a shift. And so that is like fodder or weight that can be put into like a business case to say, Hey, this is going to have a big impact. But the thing is you also think about location because the location of your Taco Bell,
00:28:31
Speaker
in California versus um like little little small town in Ohio, you're going to have different prices, demographics, the customers are going to have different motivations. So like some some markets or some parts of the country or even just cities, it can change like from one city to the next.
00:28:53
Speaker
It's like the customers might be more worried about how expensive it's going to be. So they're going the customer is going to ask the team member, how can I make this cheaper? like Is there a combo that I can get? But that may be not the case for a different that same exact brand in California and San Francisco Bay Area, but because they are probably going to be in a different economical situation. And their motivations and their requests aren't going to be driven by like how can I bundle this in a way that's going to give me the best deal, it might be more like just make it just make all the customizations for me. So that's it's a really interesting problem because we
00:29:36
Speaker
get these requests, and unfortunately we have to like weigh them across the country, across the economics of of these particular areas. Because what's what is common for you might be not common for another another market.
00:29:54
Speaker
And so we have to then come in and be like, okay, well, how do we support? We support both parties because we have to think about the most complex and we have to think about like the least, the least resistant. And so we do a lot of analysis and discovery.
00:30:10
Speaker
And as much research as we possibly can get on like where the request is coming from, that's why it's so important to understand the environmental factors. A lot of people don't think about, oh, creating an experience on the point of sale is going to be the same across the country. It's like, no, no, no, no, it's good. It's going to change. it's It's going to be very different. And so there is this like balance of like, do we create it for everyone or do we make it a customization?
00:30:40
Speaker
And then these different markets can actually take advantage of that. But then you also have additional problems you might have that might arise with that, like ah the findability of that particular feature and whether or not it's turned on, that customization. So there's all these like, I guess these like weights that you have to you have to think about. And it's a really, really hard challenge to decide what gets prioritized. This is why research is so important because without it, we're guessing. Even if we talk to, um like we can't just talk to like one market. We can't just talk to the people in Columbus, Ohio and expect the experience to be the same. And so that scale is always a challenge. And I totally forget the question that you had asked before. I totally want to.
Importance of UX Research
00:31:28
Speaker
Well, I think, yeah, you you hit on something that was very interesting to me. I, because of my,
00:31:33
Speaker
My example in my brain was like, I'm guessing Boulder, Colorado, you're going to see that minus meat plus bean more often than you would in South Texas, for instance, or something like that. So do you, if there is a prime real estate spot for the more common things, do you put the minus B or minus meat plus bean button in that spot in Boulder, but you don't do that in the South Texas? I don't know. And I can't, I can imagine you can't like,
00:31:59
Speaker
use statistics to measure and then evolve the menu and move things around accordingly. Cause then, cause most of these, if I remember when I was on the cash register, you're like, you're just listening and like, I'm, I'm just, you know, autopilot, right? So if you move things around, you're going to mess me up and exactly they miss slow so that's interesting. Yeah. That's another fun thing. Muscle memory is a constant, is a constant, um, thing that we are not fighting. We're trying to work with, we're trying to.
00:32:28
Speaker
play around and like ah almost it's so fun because like the mental the mental model of a human it's invisible but it's really cool about like UIs in general and then UX you can actually you can change a user's mental model and it's fascinating what the experience can do to the person using it. Like I could change someone's behavior by like moving a button. I can make it more likely to or make people less likely to forget about sauces by moving the sauces button in a area that they're going to have a lot more like eyesight into. So it's just really fascinating what
00:33:17
Speaker
We as designers, like I always say like we have so much power because visuals are so important to the most to most of our, ah not everyone's, but to like most of our experiences. And I think that that is terrifying.
00:33:35
Speaker
because you could like potentially hurt a store if you move a button. It's like, oh, they lost $4 million dollars today because this button moved and they had to retrain on where this went. So the impact is insane. So we have to do a lot of research and a lot of talking to people to make sure that we're de-risking a prescription or a solution.
00:34:04
Speaker
Which is inherently challenge. Yeah. I remember some, some user tests that, um, I was sitting in on and just, you know, in, in, we framed our questions in a way that we were monitoring, uh, you know, their screen at the time. And, and so we, we basically were.
00:34:28
Speaker
asking them to perform a task. So like, how would you, you know, submit this answer type of thing. And to watch them move their mouse to the button and never click it and then move the cursor off of it and then go back to it. And then in back and forth, it's like, there's so much learning and just seeing that and they never said a word and you're like, Oh my gosh, they're They're right over the button and they don't realize that's what they have to hit. I'm like, come on. It's like being a parent. Like you just want to, yeah, you're like, you're like, this is a learning moment. I can't interrupt. I can't, I should, I cannot interfere with their, uh, with this memory, like with whatever what's happening. It's like being a parent. It's like, I want, I want to push the pedals for my child because he's having such a ah such a hard time like pushing the pedals, but I know.
00:35:26
Speaker
It's a pain. That's it. That's a huge pain, but user usability testing and just like observation in general, especially when you're so familiar, like you work for months, weeks, hours, hours, just like working on a design with all these hypotheses and assumptions and then watching them do it. And you're like, you're so close.
00:35:52
Speaker
yeah But it's great because that just tells me it's like they're not confident right now. So how can I make them more confident? What affordances or visual feedback can I give them to make them feel confident? like i The mental model is there. like I'm aligned because the first thing they did was hover over this button. But what's missing for them? And that, I feel like a detective.
00:36:17
Speaker
Like I am just constantly being like, tell me your inner thoughts that you yourself don't even know. Because they don't know. People are so reactive and there's so many, there's so much processing power happening and in the background. And it's up to us to like ask questions that will at least get us closer to an answer.
00:36:39
Speaker
that is going to make them more successful. And that's what I love about user experience and just products in general. It's like, it's always learning. You're always learning. There's, you can be doing something for a hundred years and because the environment and culture changes.
00:36:53
Speaker
job security. Well, I have a couple of questions I want to make sure I ask because ah because it's just interesting to me. um One would be, so Aaron talked about user testing in in the age of the mouse in the web, but I would guess just based on being, you know, ordering tacos a lot,
00:37:16
Speaker
I would guess that the user interface that they're dealing with is not mouse and cursor, it's touch screen. So how do you like figure out like, what are these people doing? Cause they don't, they they don't interact until they look at something, think about something, and then they touch what they don't like move their finger around. They touch exactly what they want to touch. How do you observe that in a scientific way? What what does that look like when you're doing testing? Good question. So that is inherently always been a bit challenging. I think with the newer.
00:37:45
Speaker
newer technologies, um, you can like experience that hovering a bit. I know that like Apple pencil has that. You can like be close to the screen and it kind of acts like a hover, but honestly, that's not like the case for like most, most devices. And so I would say the solution to that problem is a physically observing the user in the environment. So you're watching where, and actually what I do is I have the, I have the screen recording and then I have a camera.
00:38:15
Speaker
behind them, um watching where their fingers move, because you they do tend to do this like um follow, like looking around, not always, not in that, and that also depends on the person. So not everyone is going to have, yeah, they're they're going to be looking more, they're going to visually scan rather than having like a finger actually do it for many reasons too. And so that inherently is quite challenging.
Observing User Interactions
00:38:43
Speaker
when I'm doing usability testings, when it's not an observation. So when an observation, that you just have to like take what is happening at face value and you can't like interfere because you're trying to understand what are the experiences, like what are they doing? Where is their finger going? Are they leaning on their desk? Are they leaning on the screen?
00:39:04
Speaker
What's the placement? Like what's the environmental impacts? Like are they writing things down? Some people do that. Like they'll have a sticky note and we'll write it down because they know that they need to do it, but they don't want to forget it. Um, and they know they have to do this complex task. They're trying to remind themselves. So there's all these environmental factors that absolutely cannot be observed without being in person, which makes it a very expensive endeavor because you have to like travel and get people out there.
00:39:32
Speaker
and potentially have a disruption in the restaurant. It is not natural to have usability testers or like just user experience designers like inside of a restaurant. but Like you're going to have some type of negative disruption. But yeah, that's like a constant talent. You have to physically be there. You can also um make some assumptions on hypotheses based on user patterns and user flows. So if you can't If you're not in the store and you're just like looking at session replays, then you can analyze the steps that they took versus like the end goal. so If you were to say, hey, this person ordered a 16-piece bucket with a side of fries and a Coke, um you can say, hey, this was the end goal, and this is how many flows it took to get them there. and From there, you can start inferring and then you can start digging.
00:40:27
Speaker
maybe doing some usability testing, maybe doing some user interviews, maybe doing some user observation to get down to like the core problem. I feel like oftentimes solutions and user experience in general, like the the core problem is not usually addressed with like most solutions.
00:40:48
Speaker
And that is inherently very challenging to get down to. And you really need a research expert to like do that effectively. Because if not, you risk asking a biased question and making the user feel like they need to answer in a certain way or making them feel uncomfortable because someone from corporate is here and they're watching everything I do, which is uncomfortable and it's going to make them act in a different way.
00:41:16
Speaker
And it's up to the person facilitating that conversation or that experience to put them at ease, to not ask biased questions, to make them feel like it's a safe space to like make mistakes, and empower them to like talk to us, which is quite challenging, especially if you think about the team member. um Maybe they're younger in their professional lives,
00:41:39
Speaker
And so that speaking aloud, speaking to someone higher up, they might, there might be a lot of barriers in their, the mental barriers in their mind of like, Oh, I have to be perfect. So I don't get fired. But I'm like, in reality, I don't have the power to do that. Like you're better than what I can do. So you're the expert user. Teach me your ways.
00:42:02
Speaker
And yeah, that could be, that's, it's quite a challenge, I would say. You get a lot of that observer effect, right? the sh thrilllling or just yeah Just the, just the, uh, the act of observing changes the behavior. It's like, you know, you're doing things that normally wouldn't do. and And there are things you can do to combat that. Um.
00:42:23
Speaker
and the Kind of like asking bias bias questions, right? is it's That is so hard. that is And honestly, I even still deal with that as a problem for for me. And I love chat GPT because I'm just like, can you tell me what's wrong with this question and tell me what's biased about this question and help me create a unbiased question? And I i love doing that. But honestly, it's it can be challenging. Well, because if you have a question that's going off script and you're in the store, you're in the restaurant and you're like,
00:42:53
Speaker
okay, how do I ask this in an unbiased question? Like off the cuff. Like you're just like think saying the question four different ways inside of your brain and the user or the person you're observing is just like standing there waiting for you to ask the question. And I'm like, I promise I'll get there. I just have to ask this in a way that's not going to put my biases that I have into the question. And it can be quite challenging ah to to do that for sure.
00:43:24
Speaker
Well, in allowing them giving to have the space by not asking too many questions too. Because the more questions you ask, the more likely you're adding a bias. Because if you say something like, how would you save this? OK, well, you're telling them to save it, right? So you're it's not really biased. Now they're looking for a save function. And if you've named it save, you've just clued them in to exactly where they need to go. And that used to make me, I'd be sitting in the back of the room going,
00:43:53
Speaker
It's not a good question. Throw that one out. but yes but and i that was That was the part of UX testing that I loved. was was you know
00:44:09
Speaker
just ah hearing, it was almost like I was watching two different shows. It was like screen and screen type of thing. I've got the user screen over here and the UX tester over here. And there, I was seeing mistakes on both sides and it was the kind of like great. It's like bad question. Oh, they can't find it. Oh, you are failing hard. yeah and also like You were talking about like, Oh, I just want to do it for them.
00:44:38
Speaker
Um, that's another hard thing too. Like when a user, when a test has failed, the user has not failed. That's very important because the user is always correct. Um, it's the prototype or the design that has failed. And if that happens and they're looking at you, like, what do I do? Well, wait, let's, let's be honest. There are some users that just fail. Like you have 10 good tests.
00:45:04
Speaker
And then you sign this perfectly that can't spell ABC. No, no, no, no, no. No, you're just experiencing a diverse perspective. Look at that. Look at that. I feel very passionate about this. There is no possible failure.
Challenges in Usability Testing
00:45:24
Speaker
It's usually, I mean, there could be environmental failures too. Like that could be a potential thing that's happening. Like, I don't know, the prototype froze.
00:45:34
Speaker
um That's a possibility. In general. It's a good perspective. And also tests aren't supposed to be 100%, right? There's always going to be varying. If there is a test that's like 100% correct, like 13 out of 13 users have successfully tested this, I'm immediately red flagged.
00:45:59
Speaker
um like there was a there was an error in some type of conducting or there was a bias question or and maybe this is just me being because i feel like ux is all about over analyzing everything because our biggest i would say our biggest superpower is de-risking experiences and if something comes back at 100 i'm like Did we ask it in the right way? Did we allow enough openness for high level conversation? Like what other things did we miss that could potentially add a bit of color to the situation? And I mean, there have been times where I've gotten like a hundred percent success rate and I'll give it a good hard look. And if you can't find it, you can't find it. That's fine. And you can go ahead and ship it. But I would never ever say to like a brand that
00:46:52
Speaker
I am a hundred percent confident that this, that this is going to work and this is going to be effective because of the five users usability testing that I did. There's always going to be some risk and I don't, I just wish people wouldn't be so perfectionist about it. Well, and that's the thing in a hundred percent doesn't happen in nature. Right. That's, that's why there are bell curves. That's why there's, you know, normal distributions. It's why, you know, in anything you do.
00:47:22
Speaker
It's not going to be a hundred percent. Um, and that's, and we combat this all the time with, you know, test coverage. You know, somebody learns about test coverage. Like we have to be covered at a hundred percent. And it's like, no, that's not. Yeah. I know. I've heard you say it, James. I was, I was thinking of you when I'm, uh, when I started, you're you're talking about me, you do need to be posted at a hundred percent. I'm like,
00:47:51
Speaker
yeah No, it doesn't. Because there are always those things where the test is going to cost you more than it did to create the code in the first place. And it's it you you lose the benefit. um Or it's going to be so intricate that it's going to be difficult to maintain. So it's it's not worth that that extra. Yeah, it makes it brittle.
00:48:19
Speaker
That's a huge value, like thing that you have to really take into consideration. Like I would probably never test a settings feature to like update a profile picture. Like that is not affecting my bottom line. Um, and so I think it's really important to research, not don't research everything, everything, every single, single aspect of your design, because like you said, it might be more expensive.
00:48:47
Speaker
then to just go and like build it. And at that point you might as well just like build it and then test it. Cause you know, just, just like skip, skip the design process and go into it. But like, if it's not going to have a major impact on your overarching, like objective, then it probably doesn't need to be tested or and to yeah, I think it, yeah, it adds, it adds complexity. It adds like money. And I think it's really important for researchers and product folks to understand when when and where it's actually good to do like appropriate testing. like Yum isn't sending us out to Taco Bells in California every quarter. like That's a once a year type thing where we do a like like a big chunk of testing. and I think that's just a good example of research done in an appropriate way where
00:49:46
Speaker
If you have something small like, oh, changing a product or changing a, changing a profile picture, maybe add that into one of your larger scale usability testing, like modifying a chicken bucket, which literally is seconds equivalent to dollars because every second it takes to modify something and doing it incorrectly is like a cost. And you can quantify that.
00:50:09
Speaker
And so that's a really awesome, like super power to have to be like, Hey, we can save X percent of, or not like confidently. We're 70% sure that we can save X percent of time on a particular complex order that we think will add value. And I think that's a hard thing for product, UX designers, developers. It's really difficult and challenging to, to like remember that.
00:50:37
Speaker
And you shouldn't, you don't really need to have a test for everything, but you do need to test appropriately where it makes sense. How is it going to impact? If I don't do this, how expensive, like if, if maybe like, is it how much is the risk going to be like, Oh, this, this feature isn't good. How much is that going to cost me if that's the case? And if you can make a business business case, then I think it's fine. It's sales like UX is.
00:51:07
Speaker
basically selling an idea.
Cultural Considerations in UX
00:51:10
Speaker
And then it's up to the business to like advocate for themselves and be like, okay, but you want to spend X amount of dollars. How much money is this going to give me back? What's the ah ROI? The return. It just has to be worth the squeeze kind of thing. Right. yeah i right You know, i with that, I kind of, um, leverage the, that 80 20 rule, right? You want to, you want to make sure you're focusing on that 80% of the use, you know, right type of thing.
00:51:36
Speaker
And instead of that, that 20% that's hardly ever going to get used. Um, yeah, there's a level of complexity too, even with that though, like, uh, let's say you go and propose a new solution. Um, there is this like time in which it takes the user to learn something. And then it's almost like you can't like measure the value right when a feature goes out. So you have to like give it a a little while to let people like start to learn it.
00:52:05
Speaker
I am constantly impressed by how adaptable people are to learning complex experiences. Like data management is insane. Like the amount of variations of a bucket ah ah of one singular piece is insane, like astronomical. And they're going in managing hundreds and hundreds of lines of many items, of many variations, of many types. And that has always impressed me. i'm like Are you, people can do anything if you give them enough time to learn. And I think a lot of people forget, forget that there's a balance because you don't want to spend too much time on training because that can be costly. Well, I must say I am just deeply disappointed in your answer to how you test touch face to interface. I'm here in my mind, what I had, like my mind immediately went to like,
00:52:58
Speaker
You know, there's that helicopter that like, it can observe the pilot's eyes and shoot at the thing that it's looking at. I'm like, you guys have to have these systems, like a camera based systems that monitors my eyes and knows what I'm looking at on the screen. I think that does exist. I'm sure it does in some somewhere, but that's what I was imagining. Like you had that. Cause I mean, if we can do that and that's, I think they haven't built into cars now, like to see like how alert are you? They're monitoring your face. And yeah, you get a warning. If you're, if you close your eyes, it'll be got you to wake you up.
00:53:29
Speaker
yeah Yeah, I think my mom my wife yeah has that. Yeah. Yeah. I think that you have an alertness thing. in this Yeah, they can do that. Like with your, if you're on a phone, like, so another, a good context to touch devices in the point of sale, we are on very specific devices. Like they're, they are built for point of sale and it's built on an Android system. And yeah, it's, it's for that. But like your phone, like if I were to build a consumer facing iOS or Android app,
00:53:57
Speaker
you actually might have the ability to tap into those technologies simply by purchasing a product analytics tool because take the the cameras now have LIDAR and they could like see you, which there is like this question of whether or not it should, I think is another and ethical debate that could probably be talked about forever. and But I think that is the case. like We have heat maps, too. and you have heat maps on mobile devices, and it's either going to be like how often things are pressed. You can create those experiences, basically is what I'm saying. And I'm pretty sure they exist depending on the device that you're using, but not everyone is going to have that device. no one's gonna Not everyone's going to have the newest iPhone. So we had we used to have the analog
00:54:50
Speaker
uh, metrics that we would capture. It was how worn out the button was that you pressed the most. It's like, you couldn't read, you know, meat anymore on that button because it was pressed so many times or cheese. I feel like that's, it's still, that's kind of the case sometimes with that. That's where the big grease spot is on the glass, right? A thousand percent. I think that that's totally still a thing.
00:55:20
Speaker
We had one of our consultants ah at Calibri, they needed a new laptop because their keyboard was failing on their laptop that we had we had issued them. And I was like, okay, well, you know, tell me a little bit about what's going on. And so he sends me this picture of his keyboard and the ASD and W keys were all worn off. There was a but there was a but like ah a worn off spot in the middle of his keyboard and the arrow keys on the right side were all worn off. I'm like,
00:55:48
Speaker
That's not coding, dude. Come on. At least buy an external keyboard. If you're going to be playing games with your machine, I don't care. I don't need to know that. But like, come on. We're on your keyboard playing video games. Come on. yeah So you you burn out an interesting thing earlier. you You were talking about like, oh, the meat button or whatever. So of course, my mind always goes to movies. I think of the movie, Idiocracy. Have you guys seen Idiocracy? So when they're at the doctor in, you know, 500 years in the future, the guy's standing there and is like, okay, I'm gonna have a headache and point to a button where it's like head, you know, it's like a graphical thing. Like it's a picture of somebody's head.
00:56:23
Speaker
Do you, and this is not to say that the people working at registers need to be dumbed down to, but like, is it faster to provide pictures of things on the menu or do you use words or do you use pictures? What makes it actually faster to help those people at the register? This is a very passionate subject of mine. I love it. Yes. This is, I would say, um, and this also actually brings in like the accessibility of it all too. So if you think about the team member,
00:56:51
Speaker
These are people that have varied levels of professional experience. They might be really young. They might have different levels of education. They might speak a different language because the country is quite large and there could be whole communities. Like, and I was even at our store that 90% of the time everyone was speaking in Spanish and They were like, oh, it would be really great if I had like a Spanish option. But we didn't we didn't necessarily have it at the time. um So that, I would say, is a inclusive design experience. So we had there was a lot of research, I would say, that says visuals, you're going to be able to scan a lot quicker and be more effective if you can use visuals.
00:57:48
Speaker
to like get places quicker, basically. But there is this inherent risk of like getting to the place that you need to go. But before you tap it, you have to feel confident in that decision. And so ji in general, we have larger images on our tiles of our of the menu items, but we always incorporate the text with it. So we have that additional affordance. So I'm a big proponent and my biggest thing, anything, I think anything that is tappable must have two to three affordances or like visual cues to let a user know like you are selecting on the appropriate button. And so having a combination of image and text supports both parties. It supports people who
00:58:41
Speaker
Uh, maybe they are readers. Like that is just their learning style. They're really quickly at scanning and they can like jump between words fairly quickly. But maybe someone, um, is in a very loud environment, like the team member is and like reading. Cause we all know that like, if you're experiencing a loud noise, this is why when you.
00:59:05
Speaker
When you're doing a serious like parallel park, you're going to turn down the music so you can focus, right? Even though your eyes are the things doing the work, but in reality, your whole brain, your whole body is doing the work. And so so by supporting quick navigation or quick scanability using images,
00:59:26
Speaker
And also having the text with it, usually in a bit and a bit smaller of an area, that kind of gives you both best of both worlds. So as someone gets really good at the point of sale, they're going to just be trained to look at the images. And they might notice really minute details in every image. um And that's really great. But the thing is, though, marketing images and just Food images, like how do you tell the difference between an extra crispy leg and an original recipe leg? That is a very slight color change.
01:00:02
Speaker
the whole shape is completely the same, but the color changes. And what if there's a big bright or big bright light like shining in from the drive through onto the screen and now you don't have enough saturation to know, to see that all you see is like two chicken legs and you're like, I don't know which one is which. Because you're thinking about it, like you're going really, really fast because you're incentivized to make orders, make make customizations as fast as possible, um and also not make a mistake. So if you just had images, now you're like setting up the user up for failure or potential failure.
AI in Design and Accessibility
01:00:42
Speaker
um But having those letters will be like, okay, I found the legs, fast, speed, but I'm going to verify that I select the correct one using the words. Now, if you're someone who's ah who mostly speaks
01:00:58
Speaker
Spanish and that's going to be like your primary language. What you can do is you can be trained on the image and visual cues and then start learning those ah most English words and start trying to like spot the pieces that are like, okay, this is the unique ones. This is what makes sense to me. Hopefully though, we'll get to a place where we can actually have multilingual, which we do have multilingual for like data management and things like that. So eventually that's going to have to be the case. But in the meantime, if maybe they're new and they can speak really good English, but reading isn't the the best thing, we can still like support them. So it's all about like understanding the different experiences and how do we combine both of them to create an experience that's going to allow the user to be successful with the objectives we that we have in mind.
01:01:55
Speaker
Yeah, that language barrier has always been something that um it's ah it's a bit of an annoyance for me and and not in like the that severe right wing kind of view, but but more of... you know so one and You know, if if you think about somebody that speaks multiple languages, they're multilingual. If you, somebody that speaks two languages, they're bilingual. Somebody speaks one language, American, right? ah So, you know, that one a one of those things that's always irritating me is that there's, you know, the the thought of adding additional languages is always kind of an after thought.
01:02:44
Speaker
that's brought into a product after it's already been launched, after it's been out there and used several times instead of on the forefront of accessibility. And I think that that's it's very to me that seems very narrow scope. And it's also much more expensive to do because I've done i' done some translation projects um in order to make them multilingual. And doing that at the end is so much more expensive because you have to redesign the UIs and put in those placeholders and take out those places where you've used a literal string um for what the value is on a button or in a text.
01:03:38
Speaker
and and extract those out so they can be internationalized, localized. And um it's so much more expensive than if you just would have started with it that way. Yes. It's your, you have to have really, you have to define your intentions. Yum. um I think we always, like we've always known that, like we know the, we understand the behaviors. KFC and Pizza Hut Talk about they've been them along, they've been here for a while. So,
01:04:06
Speaker
they what's great about yum is that we have that like knowledge and so we are thinking about it like even if we're not necessarily implementing like right away having bilingual flipping between each one we know it's coming and that's a superpower because we could then build things in a way that's like okay how is this going to work for left to right languages how is this going to work in this culture how is this going to work in whatever it might be and those are questions that I think we're always asking ourselves. And it's something really challenging to to get right because we have to do our research in appropriate ways that allows us to be as diverse as possible because we are a giant global company. um So luckily we have that we have that ability and we like kind of we can do some pro-activeness. And now with like new AI models, it's almost
01:05:03
Speaker
It's like, why would you not have multilingual at the beginning with its, tra like the chatty-be-teeth translations are amazing. I mean, there's still 1000% going to be issues and you might have to change a few things because German words are a bit longer than U.S., not U.S., oh my gosh, English words. And there's going to be more characters. So how do you create a a pattern that is flexible enough. But also, every time you add that complexity, it could potentially take a bit longer to like a button could maybe take months. If you're like, this has to be multilingual, it has to support at least 60 characters, it has to have an icon of this size, what if the icon doesn't show up? And so you're essentially
01:05:56
Speaker
going through all of these situations and these scenarios to actively make a really good customizable button to be flexible. And that is challenging work. And that's expensive work, to be honest, ah because it's a lot of upfront thinking, but it's, I think more, it's less expensive than, than like, like you said, doing all of the English language and then retrofitting it.
01:06:23
Speaker
not having a strategy in place. Cause now you're essentially trying to build a strategy and you're working within constraints that you might not have had if you just had built it at the beginning. But the question is like, what do you focus on? And that is going to vary depending on the product. Well, I think, I think too, that there's going to be so was something we we've seen in other areas of programming as we see patterns emerge. And you grow accustomed to these. And these are instead of thinking about certain things over and over again, they they emerge as patterns. And once once that pattern is established, they become standardized and then simplified. you know So if you think about where UIs were even 10 years ago, they were
01:07:21
Speaker
It was a much different playing field than what you've got today with, you know, the responsive design in, in these flex spaces that are adjusting themselves based on the, on the form factor that you're on and in that kind of thing. It's, it's, it's amazing to be able to just go. Just put a button here and have it size and color and text and all that automatically when You know, 10 years ago, we were having to code all of those different things. I'm static. Like it was all static. I would remember oh yeah putting my first website in text edit and making my body 800 pixels wide because like that. Uh, and when I got to college and I was, that wasn't when responsiveness came in and my whole brain just blew up.
01:08:21
Speaker
because I'm like, what percentage? And then Flexbox came out. And then I'm just like, what is, man, this is amazing. And now like, I feel like with chat GPT and even like Figma, the the design programs that I use, they're starting to like implement these practices that are very similar to Flexbox and similar to these responsive behaviors.
01:08:45
Speaker
And that is just so cool. It's so cool because we have so much power. ah And I feel like there's so much so many cool things that you can do. But it's like, how do you think of everything? The answer is you can't. and Yeah, exactly. Well, it's interesting. So you know of course, if you're doing a technology podcast right now, you have to mention AI in every episode. so um
01:09:13
Speaker
How do you think, or what do you see? Because I know there's areas of, what did you say, quick, whatever, QRS, quick service restaurant, where they're experimenting with just the drive through as a perfect case. So you can just pull up to the drive through and say, I want a bucket of chicken, blah, blah, blah, blah, blah, blah. And it's kind of just piecing together your order automatically by listening to you, where the systems are getting much more sophisticated of listening to ah You know voice input or just typing you know whatever you want and understanding what you're talking about how is that changing interface design are we are we gonna have. You know just speaking in english to the computer and it'll just do what we want of that rather than navigating through a bunch of buttons like but how is that changing your world.
01:09:59
Speaker
Yeah, I feel like it's a it's a huge it's a huge change, I think, in our in just technology in general. And I probably have an over-positive view. I always tell this, like i I always say, if you are afraid of AI, don't come talk to me. But the thing is, like I'm biased because AI helps me in every way possible. On one hand, it helps me be a better designer. It helps me be a better facilitator. It helps me create more professional, like just communication methods, I can create PRDs, I can have, I can have it like, I can, it does so much for me. So I get so much benefit. And that way I can like take a step back and be strategic about all of these options instead of relying on like, all right, well, I have to go through these like X amount of user flows. So I can really try to test and see if this is,
01:10:54
Speaker
going well or not. so yeah in general In general, I would say AI has a big impact just on in in general. and I feel like it's UI at the moment. I don't see and don't think it'll ever get to a point where you're only speaking in voice to a technology because I feel like we've already tried that with like Google and Alexa, the Google Home, um and the S-I-R-I, I'm on a Mac, so it'll it'll cut it'll start talking if I do that. um but And all of our phones will be like, what? Everyone's going to wake up. So I feel like there is always going to be some, because people, most people, not everyone, um I wouldn't even say most people. I would say it because the majority, I don't know, majority it sounds like a negative presentation. A lot of folks have, ah like we rely
01:11:48
Speaker
visually on like our surroundings to inform our behaviors. so On one hand, I do believe the future of AI and user experience is going to be a more effective way to work with technology because I look at AI as this accessibility innovation because for me, like I'm ADHD, I'm dyslexic.
01:12:14
Speaker
so it has upped my communication game and now I have this I have more respect in my field because I could I can I don't have to worry as much about like all right I can let my brain accidentally change all the letters and I can let my crazy thought process go in whatever direction because at the end of the day chat GPT is going to be my editor and it's going to like distill everything down into the main points, and I'll correct it along the way, of course. So from an accessibility standpoint, and help it's helped me quite a lot just working within
01:12:48
Speaker
any ah any technology, I feel like it's been so helpful. Communication-wise, inside of Figma, I can like name my layers appropriately, naming design system elements and components. like I was even building out like a class system in ah like Webflow and that was really helpful to like ask and converse with chat GPT to be like, can you tell me like what makes class effective? What makes it variable? Tell me the properties that, or what are like the the base, the base things that I should know about. And then I can say like, tell me what, how I, how, what are some recommendations? How would you name this particular class of elements? These are my intentions. This is, this is all that. So I feel like on one hand,
01:13:36
Speaker
It's going to empower a lot more people to be more technical, which is amazing in my mind. um But I think there is always going to be some type of UI, even with the glasses and stuff. like They're ah going to have some type of visual visual cue. And maybe like it's more generative UI, which is a fascinating concept. But to the point we were talking about earlier with memory muscle memory,
01:14:05
Speaker
I don't know if AI will be able to effectively know when to change something versus not changing something. We could probably train it to do it, but like there's always going to be this risk. So from a UI perspective, I think it's just going to make humans interacting with technology more empowered. And it's also going to train us to be, as humans, to better communicate our intentions. That's already, chatTBD has already helped me with that. Like, I feel like I can way better articulate the things that I want out of an experience because I had to train my GPT to have the appropriate context, to be able to be effective in ah like just communication in general. It's actually, think I think of chat GPT kind of like consulting because consulting, you really have to dig deep on what a brand or what a client needs.
01:15:04
Speaker
It's like, do you want this because you like the solution or do you want this because you're trying to solve a particular problem? And let's try to open that door. And I feel like consulting in the past for me has really helped me better work with chat GPT because I can ask it those questions and say, tell me about this. I'm also testing its methodology too. So I'll be like, what makes a good unbiased question? And I'll correct it.
01:15:29
Speaker
as it as it goes to like modify the prompt to me. So it's like, I understand that this is like the best way to to ask a unbiased question, but for my situation, I want to ask it in this context because of this, this, and this. So it's ah it's really it's really fun. but yeah So I guess my non-answer is that there will probably always be visuals, but it will evolve and empower us as humans to interact with technology.
01:15:57
Speaker
ah a lot more effectively. So, excited to see what people of different backgrounds can actually achieve by utilizing AI. Okay, you said something interesting. We don't have a, you know, we're trying to keep this to an hour and a half or so, but but but I think we want to dive in. At least I do, because, you know, I'm interested in this. You said you think humans interacting with the computer would would teach us to interact with one another better. So that's an interesting and a provocative thought in and of itself, I think. To me, the the layperson, if I was just thinking about it, I would think, I know that chat GBT and these LLMs, I can misspell things, I can say things out of order, I can kind of just to be very loose and and rough with my input. And it kind of figures out what I'm saying. That would lead me to believe that im it was going to teach me to be lazy with my communication. So were you saying that
01:16:54
Speaker
The communication as our our written communication filtered through like a an LOM helps that become more clear, concise, maybe considerate of other things that you might not consider. Or are you literally saying like, I'm going to be able to sit across from a table with Erica and speak with her better because I've been interacting with chat. chi could tea Which one were you asserting?
01:17:14
Speaker
Yeah. So I feel like it's combination of both almost. So like the last, the last one where, um, it's helping me better understand, uh, like, or it's helping, it helps me better tell people or explain like the context that's missing. And I think that that is something that I like to do with chat TPT. I ask it like, okay, what is it missing? Like, what am I missing? What, what, how would you improve this? Tell me why you would improve it.
01:17:43
Speaker
So it's like almost like a very, it's like a tutor that is reading every single thing. So for one, on one hand, it helps me because the thing is I'm also very curious. So I don't know if everyone would want to do that. And I think there might be some type of laziness where like, now I'm going back to what I used to do in college, which just word vomit into notes.
01:18:09
Speaker
And then I would spend hours and hours and hours editing it, but now that process is a lot faster because chat GPT just does all the editing and then I just correct correct as I go. So it allows me to also understand like how how other things might be perceived more positively. And because I can ask that question and change my answer based on that, I think I'm more effective at describing what I want because I'm constantly reading the the newest results of my word vomit. And my word vomit, to be honest, it looks still like the word vomit that I had before. like it And sometimes I just do it in a voice and I'm just like, blah, blah, blah, blah, blah, everything and anything.
01:18:56
Speaker
And it's really, I think it's really powerful to pull that into like a singular sentence because it teaches me like, okay, this is what an appropriate structure looks like coming from my voice. Cause that's like the biggest struggle for me. Whenever I was word vomiting into everything, it was really challenging to fit the conventional standard because it doesn't, like that doesn't come naturally. It doesn't click with me. Like the sentence structures don't click with me.
01:19:22
Speaker
Punctuation doesn't click with me. Letter placement doesn't click with me. It's just how my brain works. And so that's just allowed me to be, to to see like, okay, this is what makes it effective. And you can start seeing behaviors. You can start seeing patterns. This also could be just as a person who loves research. So I'm like spotting all of these like similarities. And I'm like, okay, this is how I can communicate that better. Because at the end of the day, I'm actually most of the time drafting out a lot of my meetings beforehand, reading them, making sure I get all the points that I need, and then communicating that effectively. So am I really being lazy? Or am I doing like twice as much work?
01:20:04
Speaker
to make confirm that a meeting goes well It could make some of the things that we should be doing with our communications a little more, I wouldn't want to say accessible approachable maybe is a better way because you said you should be preparing for your meetings going through notes, but if you have a, you know, an LLM.
01:20:21
Speaker
going through all your notes and summarizing for you, it's not going to take you an hour and a half of prep time for that meeting. It might take you 15 minutes. So you will do that. And that thereby enriching your communication experience because you can go through it. Yeah. And now I'm two steps ahead because instead of spending a whole hour trying to prep for a meeting.
01:20:40
Speaker
Now I can spend that time on doing the next step after the meeting. I can spend the time ah proposing new solutions and working on harder problems. So I think that AI really just helps people do basic things that, sure, we could pay someone to to do that. um But if you can get that person working on not grunt work or work like intern work, like redlining,
01:21:08
Speaker
Um, instead pay them to solve bigger problems. I think that is way more impactful. Actually, I think one of my, one of my favorite quotes, and I don't know who said this might've been even Joe Park, the, the, um, uh, yum, um, CTOs, I believe. And he, he said like, AI won't replace humans, but humans using AI replace humans. Cause they're just going to be more effective. Like that's just the reality.
01:21:37
Speaker
So it's fascinating. It's how technology has affected human beings over the course of our entire history. you know Now that we have fire, we have more time to sit around and think about things differently. It's always about like freeing up your brain for more think time, I think, is what technology does, that we can get into deeper and deeper yeah thought patterns and thought processes.
01:21:59
Speaker
And then we create new technology and that frees up even more time, right? So it's, it's interesting how technology affects humans. It's like ah ROI too, right? Like, do I want, do I want to pay someone this amount of time, this much money, um, to just like edit, copy and change words, or do I want them to create a ah language system that allows my company to speak in the same brand voice across an entire.
01:22:24
Speaker
and and our entire global brand, like what's going to get you more value, the person editing words or the person defining like your brand voice and like ah hitting home and adding more value to your overall business. It's kind of how I like to think about it. I think that's a great segue into our next segment, which is ship it or skip it.
01:22:51
Speaker
Ship it or skip it. Ship it or skip everybody. We got to tell them if you ship or skip. So the first ship it or skip it type topic that we'll address today, since we were just talking about AI, let's do AI generated content. Ship it or skip it. Erica, since um i'm sorryre you are the guest, you get to start. You user or AI generated content.
01:23:20
Speaker
I would say, a ship it with lots of disclaimers.
01:23:29
Speaker
like Literally and figuratively, lots of disclaimers. If you're going to have AI-generated content, be transparent about it and and like get it peer reviewed. It's almost like research.
01:23:44
Speaker
like If it's AI generated, you must have it reviewed and make sure that it's in an actual, like it's telling the truth because LLMs, they hallucinate and that if anyone tells you differently as of right now, that it's, they're totally lying. Like they, it totally hallucinates. But what's great about AI is you can ask it, Hey,
01:24:07
Speaker
What about this is, what did how did you change it from my original prompt? Or um did you make anything up? And if you did, why? Tell me your reasoning. like Walk through that experience with me. So I think it can be really useful. And I'm a big proponent of it because I'm kind of lazy when I write.
01:24:27
Speaker
so yeah i'm I'm going to be at basically the the ship it with an asterisk, right? With the foot known, it's like, yeah, if it's all checked over. um Funny story, my my son the other day came up and said, co-pilot was gaslighting him because he had it he had it generate him an image. And then later in his prompt as as he was you know having the conversation, it said, I can't do that. And he's like,
01:24:58
Speaker
didn't you generate this image for me? And it was like, no. um that's funny I never did that. And I'm like, that's, that's creepy. um But yeah, I think there's a lot of ah There's a lot of value in, in generating the content, but like anything, you know, it, it's, I would consider a first draft, you know, it still needs to be edited down, fact checked, absolutely fact checked. Uh, because I have heard horror stories about it, um, making up answers that aren't true.
01:25:37
Speaker
and things like that. This information is a real problem. So yeah double check everything. Don't trust anything you read on the internet. Yeah. I'm i'm on the same lines as as you all on the shipper skip, ship it for sure ah with asterisk. the The hallucinations are interesting for a computer because like if if I'm interacting with someone, like if I'm talking to Aaron and he's making something up as he goes along,
01:26:05
Speaker
I can see his body language. I, we're we're kind of built, we, you know, over generations and generations of being able to pick up on that sort of thing. But you don't know that the computer, like it it says these hallucinations with the utmost confidence. You don't, there is no body language to tell you. I'm kind of BSing you right now. You know what I mean? So yeah, that that's pretty, that's pretty scary stuff. Our last, you know, our last episode, the guest we had, Matt Coteney was talking about like the hallucinations in the legal industry or You know, just made up cases like that never existed. Like, Oh, here's some precedent for you. Judge the, uh, no, actually that's not a case. It didn't never even existed, but it was very confident and it said, Oh, you should cite this. Uh, yeah, that's pretty, pretty interesting stuff. All right. Yes.
01:26:50
Speaker
But when it makes up its citations, like, I mean, you have to go and check
Context in Design
01:26:54
Speaker
them, right? Of course, right. but I think it made up the like, oh, you should reference, you know, finders versus keepers, whatever. I think I made that joke last time, but it's, it's pretty interesting. Okay. What's next, Aaron? Let's do designs as requirements. So this kind of leans into the UI and UX area of having a design as a standalone requirement.
01:27:21
Speaker
No. Like this is the visual, like the pictures. and Yeah. Like, you know, like your Figma, like this is the requirement. Make it look like this and do it. A whole hard, hard cancel on that. I think, uh, it's ah UI does not encompass business cases, research. It doesn't have the context to do it effectively or the foresight to do it well. So just building a button that looks like the standard button isn't going to account for language and color and label and like, also I think it's just unfair to have design as a requirement because like it's expensive. It's more expensive to create a Figma file than it is just to like write a document. Just get chatcpt to write you a PRD or product requirement document and parrot. So hard, hard pass for design being the requirement.
01:28:23
Speaker
i I have to agree. I don't i don't think just just having the picture is enough. there're there you know As a developer, there's there's nuance as you're you're going through and implementing a solution. Understanding the why, what what am I getting at here that you know developers are creative, right? Giving them a little bit of creative license too.
01:28:43
Speaker
Think along the way, uh, and not just try to be pixel perfect to an image. You ask for pixel perfect to an image. You guess what you're going to get pixel perfect to an image. You know, we've had, you know, you hear the horror stories, add a button to the screen as your requirement, just, you know, as the text or whatever.
01:29:00
Speaker
Okay. The button's there. You click it, it doesn't do anything, right? Like we can be very little jerks sometimes when, you know, um, but yeah, I think I agree with you. The context, the why, I think is always very important. Understanding that outcome you're looking for. What what are you trying to, what experience are you building for this user? All of that. Yeah. Implementation is more than just the, what it looks like at the end. Yeah. And i'm I'm going to be on board with that as a skip. Um,
01:29:27
Speaker
But I do think it's important to have them as part of your requirements. So it's kind of a, yes, have an idea because that picture is worth a thousand words. So make your requirement a thousand, one hundred words, right? So have your, have your, this is what I expect it to do written out and then have the picture to support it so that you can put all the pieces together. um Because I have found that knowing what what a product owner or designer is his thinking with their pictures is helpful. Kind of like you were talking about with the button, right? Having the image and the text is more effective than one or the other. So I think it's a it is a not having it as a requirement, but
01:30:26
Speaker
um having it support the requirement. Yeah, showing intent, right? We want the intention to be very clear and we'd rather have the intent clear. I'd rather have a fuzzy fuzzy UI with clear intentions because that means that, yeah, there's just some opportunity for improvement if need be.
01:30:46
Speaker
If you ever read those requirements, like for user interface requirements, where it is literally just words like, I need a button that is 160 pixels wide and has this sort of ah arc, blah, blah, blah, blah, blah, rounded edges. At that point, get AI to build it for you. just Yeah, like, come on, let's just build the build the stupid button for me if you're having to go through all that. Writing all that out is a lot slower than just and building the vision of the graphic itself. Just give me that, and I'll just pluck it in for you. That's fine. OK, the last one we'll do today um is one that I think maybe we won't agree on. I don't know. We'll see. Um, hallway testing, hallway testing.
Effectiveness of Hallway Testing
01:31:24
Speaker
I would say it's, it's another asterisk one. So hallway testing is really, I would say it's a very useful tactic when you don't have quick access to your users. So if you want like general usability testing, you can use a hallway test to get you like 10% of the way there.
01:31:45
Speaker
But what you have to understand, if you're testing with people literally in the hallway, they bring biases because they likely work at the company that you work at. yeah They might have deep understanding of how data models work. And so, sure, they might be able to get you in the right direction of like a placement of a button, but the moment that they get into like strategic, oh, I'm going to build a menu in this way, it fails because you're not you're designing for someone with a deep knowledge of the system, but it's not like true usability. Like if you wanted to actual usability, learnability, don't, and we even do this, like don't test with Taco Bell employees ah all the time. Not saying you should, you should always test like test with your users. That's going to give you the most ah risk adverse like direction. ah But yeah, I think,
01:32:41
Speaker
you can test with other brands as well to see if like, okay, for usability, can this be learned and without context? And then can this be learned with context? I think it's really important to like just make a distinguish between those. So it's good when you don't have anything else to go off of and you need some type of like second pair of eyes, because let's be honest, when you're designing in a vacuum or just looking at something for more than two hours, you're likely going to make decisions that you maybe wouldn't have if you had a fresh perspective. So it's like a yes, but. Yeah, I'm, I'm going to be on the skip it train on this one, because I think to a lot of the so the same points, you're not going to get an unbiased result from your peers because they already know too much, right?
01:33:38
Speaker
Uh, if you're designing something for most likely your hallways are going to be packed with people that know the project that you're on, know what you're working on and are expecting to see certain things or are used to an older version of the product because they've been trained on it at some point. And I don't think you're going to get that fresh pair of eyes. I will say though, it's important to get peer reviews of it is like, what do you think of this? Um,
01:34:06
Speaker
As, as more of like a review step, which, you know, anything that's design coded or thing, I think it's always good to have a second pair of eyes kind of looking at it's like, am I crazy? Or is this actually look like it's going to work? Um, so I'm, for hallway testing, I'm pretty much a, uh, skip it. Go ahead and hit your user base, get some fresh eyes on it and, um, you know, do, do your normal testing and.
01:34:36
Speaker
Hallway tests are better for like dry runs of your test that you're going to be testing with real people. Okay. So, all right. For the, for the listeners, both of them, um, what distinguishes a hallway test from a traditional user. Like what, I mean, I guess based on the context, I guess I can infer what it what it is, but like what explain, what is the definition of what a hallway test is? And for, uh, one of the hosts, uh, he might need education on what, what you're talking about. Cause I've never heard that term.
01:35:05
Speaker
Yeah, I guess hallway tests, and I think this really does determine, depend on like the situation. Holloway and test implies you and want to do a usability test on someone on your team and potentially you want to like, well, I guess maybe that there's, there's assumptions. So maybe I won't say the assumption, but like it's essentially you are testing and experience with a colleague instead of testing with, maybe not instead, but versus testing with a actual user or someone who doesn't have the deep understanding of the business and those operations. in in the In the term comes from, you walk down the hallway, you grab somebody and say, hey, I want you to test this. yeah User testing is more like, I um i know exactly who I'm testing, I'm i'm
01:36:03
Speaker
doing a giant like filter. like I'm only looking for tall, cobalt employees who've been in that company for more than a month and understands how to ah how the restaurant operates just in general in store with fulfilling orders. When a hallway tests, I talk to my dev who knows the schema.
01:36:27
Speaker
and You're not going to get the same results. it just dates it's not It's not real. You might get some perspective, ah but then you get into like designing my committee, which can be like really annoying and not an effective use of time. But there are some use cases, I would say, where like having peers look it over and maybe say, oh, well, this is going to be too expensive to build.
01:36:51
Speaker
who So that's like the difference. It's like the person usability testing, testing with the people that might use it or an unbiased audience that might not, that might not need to use it. Outside of your user community, someone outside the user. Yeah, I don't know. I guess it would be interesting. So like I think of, I'm a developer, so I think of in terms of feedback loops, you know, unit tests are quicker and you get your answer, but the the problem with, if you're going to you know rely on a hallway test to kind of short circuit getting to that real user test, you may throw some things out that were,
01:37:23
Speaker
the, you know, like the a fellow developer, I might go to Aaron and say, what do you think of this, this screen? Oh my gosh, that's disgusting. That's gross. and But, ah but if you really put it in front of the real user, they're like, Oh my gosh, that solves my problem. This is beautiful. I love this. So I would be wary of that. I would, I definitely wouldn't use it as a way to short circuit, but it could be more holistic view of the product, but, but ultimately you have to change those users lives. So you you really want to focus on that. So I would, I would say skip probably. That's what I would guess. Yeah. I think it depends on like your intent. If your intentions are,
01:37:53
Speaker
are clear and clean, like I'm i'm asking feedback um so I can get another diverse perspective versus I'm asking you as my coworker, if this is going to fit within the mental model of all thousands of our users, that's going to be the time where it's like, okay, your intentions are not good. You're trying to skip a potentially very expensive process. Yeah, but I mean, if I'm being honest,
01:38:22
Speaker
I'm just trying to get to the, you know, helicopter targeting system testing thing. Like that's, that's where I'm trying to get to as quickly as possible. I i want to use that. I would love to sit in the cockpit of, uh, the patches do that. Yeah. Like that would be so cool to be able to see that in action. Anyway, that that's what I'm trying to get to is that, is that, but yeah.
Lightning Round and Conclusion
01:38:45
Speaker
All right. Was that the last, uh, that was the last one we're going to do.
01:38:48
Speaker
So it's time for our most serious part of the podcast, the lightning round. Okay.
01:39:15
Speaker
So the way the lightning round works, we're gonna ask you a series of hard hitting questions. We'll each take five pot shots at you to see how you survive. Try and keep these short, no description or no embellishment is needed.
01:39:38
Speaker
We're looking for rapid fire answers. And some of these do have right and wrong answers. you And you will be graded accordingly. Yes, you will be graded accordingly. I was going to say judged harshly, by but graded accordingly is a little bit easier. Bring it on. I'm ready for challenge. All right. So I will start things off with your first question. What's your favorite carb? Bread, pasta, rice, or potatoes? Bread.
01:40:08
Speaker
Do you like the smell of gasoline? Yes. Do you believe in love at first sight? Yes. Did you go to sleep away summer camp as a kid? Yes. Do you like the name Charlie for a girl? Yes. Can you say good day mate in an Australian accent? And I had to like slow that down so that I didn't do it in an Australian accent. So can you say a good day mate in an Australian accent?
01:40:38
Speaker
ah Good day, or good day, mate. Nope, I can't. Good day, mate. Maybe I should add an R. R. Good day, everybody.
01:40:51
Speaker
yeah That's fantastic. Ask for permission or beg of forgiveness. Forgiveness. What's a country you'd be okay never visiting in your life?
01:41:05
Speaker
ah here I don't know. You're such an inclusive person. This is going to strain your brain. My brain doesn't know. i everything Every country that pops up in my head, I'm like, nope, nope, nope. I would say ah the moon.
01:41:27
Speaker
okay
01:41:30
Speaker
No atmosphere there anyway.
01:41:35
Speaker
exactly So, since since we failed on the Australian dialect, we'll try a different dialect. What sound does a seal make? um and See, there we go. Well done. Quick recommendation for any art schools that should be your, uh, your, your like mascot. Cause like art art, art art. I totally give credit to my husband for that. He's the one who made that joke. I love it. That's great. Uh, on a scale of one to 10, how much do you enjoy garlic? Oh my gosh, 40.
01:42:14
Speaker
Okay, we would accept it that as well. yeah That completes the lightning round. Did I get an A or a B? or what I think that was a ah solid effort. I think missing the Australian accent gets you in at an A-. I can live with that. I'll work on it for next time.
01:42:36
Speaker
This has been the Forge Slash, where we lean into the future of IT. Thank you to our guests, Erica Price, for the great talk on UX and UI at YUM. Also, thanks to my beautiful co-host, James Carmen. I'm Aaron Chesney. Please subscribe and stay tuned for future episodes, and always stay curious.