Reed Enger is an artificial intelligence experience designer at Google. He’s also the founder of Trivium Art History, a digital platform that shares the stories behind 30,000 years of creativity.
After graduating from the University of Northwestern in 2009, Reed left behind his training as a print designer. Employers at the time were more interested in hiring web and user experience designers. “The term ‘experience design’ was kind of frowned around early on,” Reed remembers. “It seemed a little indulgent, like, “I’m an experience designer.” Well, come on, man. You make websites.”
Reed describes his post-graduation years as the best time to build a big portfolio quickly at many of the local agencies—back then, user experience was still maturing as a discipline.
Reed Enger, the man himself.
As Reed’s knowledge expanded, he realized that experience design was more than just building websites: “I have to never take it for granted that I might have to learn a brand new language, metaphorically, to tackle new problems.”
In 2017, after spending a few years embedded in New York agencies, he moved to Seattle to join the Google Cloud team. Here, he was tasked with driving a seamless experience of design systems for one of the technology giant’s most important divisions at the company.
“You should have fun when you’re designing. I would love to think of a design system as just a box of toys.”
This involved having a direct impact on how people interact with products and services like computing, data storage, data analytics, and machine learning every day.
We spoke to Reed about design as a form of patience and caretaking, contemplation as a kind of counter culture, and why it’s a good thing to remain skeptical about the future of design systems.
[Be secure in the future of your own system with InVision Design System Manager]
ID: Why is it important to keep human-centred design in mind when building design systems?
Enger: It’s that idea that you need to design for a person and, to do that, you have to understand what they need and hopefully some of how they think. That’s, I think, what you can do with a design system.
Because a lot of design systems sort of grew out of code repositories, because developers were way, way ahead of designers when it comes to modular, flexible, reusable kind of componentry. So, I think that as part of that legacy you do wind up with design systems that are pretty formal and sort of optimized for Ctrl+F. They know exactly what they’re looking for and then know how to get to it. It’s a resource and less of a guidebook.
UX can make design systems more of a guidebook—an invitation.
Design systems can get a bit of a bad rap because they’re seen as a constraint. The thing is you can’t do whatever you want, because you’ve got to work from the system. But, just like a box of Legos, a design system applied correctly is a creative tool. It’s not a limiting tool.
ID: You described yourself as a caretaker for the design specifications of the Google Cloud when you first joined. I wondered what made you use that word and how you take on the responsibility of the caretaker role in your design practice.
Enger: Caretaking is humble work. It’s not particularly flashy. It’s playing bass, not lead guitar. Working in design systems I found to be very patient work, and I don’t think that that’s a bad thing.
I’d come out of the sort of rock-and-roll New York agency scene, which was really, really fun. It was an interesting adjustment to then go to a place where a huge part of my job was just listening and identifying the challenges that future teams were working on and then supporting them however I can. My job was not to create something that had real sting, real flavor. It was to melt it away until it was strictly function and could be applied anywhere.
It’s a very invisible sort of design, working on a design system. I really came to love that. I found it to be kind of egoless. You just want the thing to work really well. You want it to be clean. You want to sweep out the cobwebs in the corners. You want to Marie Kondo that design system. It’s that same satisfaction you get from lining up your shoes in front of the door in a really nice straight line. It’s that sort of thing. Caretaking, I think, has its real benefits.
ID: How would you explain a design system to someone if the phrase didn’t
Enger: I would call it a box of Legos. I would say, “Here’s a box of toys.”
You should have fun when you’re designing. I would love to think of a design system as just a box of toys. Lots of possible combinations. And then somebody comes along every now and then, gets creative, and lights all of the toys on fire and builds a new one, which is great.
It should also be participatory. You should involve lots of people. You should share your toys. If you have a really nice toy, you should add it to the box.
ID: In your experience, how can you make a design system less intimidating
and more useful?
Enger: Edit, edit, edit. Documentation can be so long, and most of the time you don’t need the majority of it. Describe everything as simply and concisely as possible. Use visuals when you can rather than words. Provide very simple overviews that somebody can breeze through and glance their way around. Make it fast, and make it simple.
“I think that you have to get comfortable with the fact that many things can be true at once.”
It’s funny, a design system is like any other experience. You have to design it for elegance and optimize it for efficiency in the same way that you would any other experience.
ID: Can you talk a little bit more about the process of optimizing for efficiency when building a design system?
Enger: I think it’s about providing entry points that speak to different types of users. I’ll think of an example here. Someone who’s really familiar with the design system is going to want to get to what they want as fast as possible. So, you need to build in a search page. You need to have lists with the keywords that they need so that they can just get right to it. You’ve got to make sure there’s power to use your features.
Icons from Trivium’s design system
Even all the different heads on the same design system can be a big factor in how efficient something is. We had an engineering-focused portal for the design system that was basically just a huge, long list, and the engineers loved it. It was super fast. But then, for the designers, especially because the team that was growing really fast, we needed to onboard people. So, we had a visual overview that showed images relative to components.
The system should present itself in a way that is clearly understandable to the user that is using it.
ID: How do you think your knowledge of UX shaped the way that you approached living in the world on a day-to-day level?
Enger: I used to worry that being a visual person and a designer and a writer was making me dissatisfied because so much of trying to develop taste is learning to identify when something is off. It’s that carefully curated sense of dissatisfaction that lets you improve as a designer. I used to worry that it was going to turn me into a grouch.
I’ve been very relieved to discover that what it’s actually done is just develop my curiosity more, because, if you see something that feels off, that feels wrong for some reason… Maybe it’s type that’s kerned poorly. Maybe it’s a door that opens in instead of out. When you start to pay attention and notice those things, sure, you can be a grump about it, but you can also be curious about it, and you can try to figure out why that was made that way. Was it an accident, was it intentional? Maybe there’s something going on that you don’t know about.
It opened up this sense that there’s always something to learn. So I try to exist in that space of being excited to learn and keeping my eyes open to see what I can learn next.
ID: Earlier this year, you gave a talk at the Awwwards Conference in New York on how design systems have souls. Your talk alluded to the fact that everything has a soul—what do you mean by “soul?” How do you know it’s there?
Enger: When I talk about the souls that are in design, I’m using the word in an esoteric sense. There’s a variety of long, dead philosophers and scientists and nuns who referred to objects having an inherent purpose and unique and specific purpose. And everything had that purpose. A person had it, but so did a tree. So did an apple. I love that idea. And I think the thing that I like about it is that it forces you to ask the question, why is this here? What is the purpose of fill-in-the-blank? Apply it to anything at any level within a design project.
I think it’s also a really kind of lovely way of equalizing many disciplines, If everything has a unique purpose, then that purpose accounts for not only its functional behavior as a UX person would craft, but that unique purpose needs to be served by the visual design. That purpose needs to be served by the copywriter who creates the snappy call to action. That purpose is applicable to everyone who touches it, and everyone needs to be responsible to that purpose.
Then that purpose is also something that relates up and down. It’s that as above, said below hermetic thing. So, you have the small element. Is it harmonious with the larger themes of the product, the website, whatever it is that you’re working on? Everything has to make sense from the top to the bottom, the bottom to the top.
You’ve got this symphony of elements and components and design styles, and they all have to sing in tune. Their souls have to be aligned for the product to feel good.
When everything’s lined up, when all the souls are being expertly crafted and they’re all singing the same tune, then you wind up with something that feels complete and finished.
“If everything has a unique purpose, then that purpose accounts for not only its functional behavior as a UX person would craft, but that unique purpose needs to be served by the visual design.”
That kind of goes back to that idea of dissatisfaction. If you’re crafting that dissatisfaction, then you can start to see where things are out of alignment. I think that that’s what all good designers do. That’s what all good writers do. That’s what a good UX designer does. They identify where things are not serving their purpose. That’s it.
ID: In that same talk that you gave, you mentioned the word ‘glanceability’ as a key component of design systems. Why is that important?
Enger: People don’t read. People just don’t read.
ID: Why do you think that is?
Enger: I think it’s because we’re in a hurry. You know? We’re on a deadline.
A timeline of 40,000 years of art, taken from But also, documentation and design systems tend to be pretty dry. I think that’s kind of the other component. It’s really rare to run across writing that is pleasurable to read in a design system. It’s not impossible. It’s out there, and god bless the people that put in the time and love to do that. Make it glanceable so that somebody can whip through it as quickly as possible.
I think it’s just a matter of kindness. To try and describe a dropdown menu in words is … It’s terrible to write, and it’s terrible to read. If you’ve just got a little animated image of a dropdown opening and closing, you get it right away.
It’s just easier on everyone. I mean, even better, if you’ve got a little live snippet of code, and you just have the dropdown menu there. Then somebody can go in, and they can click it, and they can see it open, click and see it close. It’s just merciful to everyone involved to try and keep things visual and short.
ID: When it comes to the history of design systems, key moments have included the publication of Christopher Alexander’s A Pattern Language, Jenifer Tidwell’s view of design systems as shortcuts in the 90s, Brad Frost’s theory of building systems, not pages, and the introduction of Google’s Material Design. Let’s look to the future, perhaps 50 years from now: what do you see as pivotal—and hope will exist—in 2069? How far could we go?
Enger: We’re in a weird spot right now with design systems. They’re primarily used by large-scale products. A lot of enterprise-level products. A lot of products that are used by millions if not billions of people.
What we’re seeing is a radical standardization in design systems. And how could you not standardize? These things have been exhaustively A/B tested, optimized down to the pixel.
So, now what we have is an extremely narrow range of highly-performant elements that are recognizable and efficiently usable by the greatest number of people possible. We have arrived—or we are close to arriving—at a point where there’s something like a design supersystem, and everybody’s kind of just using variations of basically the same thing. That’s extremely, extremely efficient.
“A good system would know when you want to use tool A versus tool B versus watch YouTube. That’s what a hyper-personalized system would allow.”
I think it’s going to flip from being a standardized design system to being an individualized and generative design system where every individual person has their own design system [assigned to them by the product], and they’re probably not even aware of it. The system’s probably being generated based on the patterns that you yourself have as you move through space, both digitally and the real world, when you interact with your devices; when you walk into a Starbucks.
The massive awareness that is being sort of slowly created, the coverage that we have now with sensors and transactions I think is going to be enough to inform a hyper-personalized design system for the individual that will then build your experiences and your tools, probably from some insane fleet of microservices.
A good system would know when you want to use tool A versus tool B versus watch YouTube. That’s what a hyper-personalized system would allow.
ID: Do you think there is such a thing as healthy skepticism about all of this?
Enger: Oh god, yeah. I think that you have to get comfortable with the fact that many things can be true at once. The same slice of code that could be hugely empowering could also be used to reinforce a surveillance state. I think skepticism is really, really important.
The potential is so incredibly, incredibly strong. It could be absolutely, radically misused. It is every day. But it could also be wonderful. You want to keep making stuff that pushes it in a wonderful direction.
“A good system would know when you want to use tool A versus tool B versus watch YouTube. That’s what a hyper-personalized system would allow.”