Greg Petroff is a renowned executive design leader. He’s built and led design teams for companies like GE, Google, Compass, and Cisco. In this conversation, we discuss the state of design and how designers can have more relevance in today’s changing environment.
Show notes
- Greg Petroff - LinkedIn
- Improbable Futures - Greg Petroff (Substack)
- Greg Petroff (@gpetroff) on Threads
- Dave Gray on Possibilities – The Informed Life
- A generation of design leaders grapples with their future by Robert Fabricant
- ONCE
- Figma Config 2024
- Figma’s AI app creator accused of ripping off Apple weather app by Alex Hern, The Guardian
- teenage engineering
Some show notes may include Amazon affiliate links. We get a small commission for purchases made through these links.
If you're enjoying the show, please rate or review us in Apple's podcast directory.
This episode's transcript was produced by an AI. If you notice any errors, please get in touch.
Transcript
Jorge: Greg, welcome to the show.
Greg: Hi Jorge. Thank you for inviting me.
Jorge: I am so pleased to be able to talk with you. I have known you for a long time. We’ve known each other for a long time. I’ve admired your work, your career, and we also, I think, share similar backgrounds in that you come to the design space from architecture. I could talk a lot about your accomplishments, but I think I’d rather have folks hear from you directly. How do you introduce yourself?
About Greg
Greg: My name’s Greg. I tend to introduce myself these days as a possibilitarian. And what does that mean? It’s a term that isn’t a real word, but if it were a real word, it would describe a point of view about looking at problems and recognizing that there are ways to solve them and that we should create the future we want. And we may have many issues that are sitting in front of us, but, if we’re creative — and we use design as a methodology of looking at the world — we can come to a future that’s actually something that we like.
I’m an architect by background, so I studied architecture. I have two degrees in it. I actually practiced as an architect early in my career, and then I have spent the last 20-plus years working in mostly enterprise software, some commercial and some consumer, as a designer and then, most recently, as a leader of designers.
Jorge: You’ve had what I consider to be very prominent positions in very important companies leading design. I think that I met you when you were at GE. What was your role there?
Greg: Yeah, so GE — I joined GE; they were building a group called GE Digital, and I was one of the first employees. I built a team from scratch, which was really a privilege and a lot of fun. And I was the Chief Experience Officer for GE Digital, and for a period of time, I actually also owned the brand for the company as well.
I was at Google for a short period of time. That didn’t really stick for me. I love Google. It just was an environment where I didn’t meld with the manager that I was working with. Then I was at ServiceNow, helped build their design practice from small to medium, large.
I was at Compass, a real estate company, which was amazing, super fun. And I’d probably still be there if the real estate market hadn’t tanked and they ran out of money and stopped being able to afford people like me. And then most recently I was at Cisco in the security business, running design as the Chief Design Officer for their group.
Jorge: My perception is that in all of these roles, you’ve done what you described doing at GE, where you’ve built a team to improve the experience through design.
Greg: Yeah.
“Possibilitarian”
Jorge: I’d love to talk about this word, ‘possibilitarian.’ I had a conversation on the podcast with Dave Gray, and he also spoke of possibilities. I got a little bit of that vibe when you were talking about it.
Greg: Yeah.
Jorge: What is the relationship between this idea of creating the future we want and design?
Greg: Yeah. I think there are two things that designers bring to the table that other disciplines don’t – it’s just not their first inclination. The first is that we are practiced at abductive thinking, which is the ability to generate multiple ideas and keep them alive for a period of time. You learn that in architecture school, and you learn that in design. You learn the ability to use artifacts to inform you and have a conversation with your work. You’ll often hear me talk about making to think versus thinking to make. I think in that conversation with the medium you’re working in – whether it’s paper, digital, AI, or whatever – the mistakes inform you along the way and allow you to keep a number of opportunities open that may surprise you. In their execution, you discover something new.
The second thing that designers do well is the art of juxtaposition. This means taking two things that seemingly are different and unrelated, and putting them together. The creation, friction, tension, or synthesis of those two things together creates new opportunities, new ideas, or new thinking. We bring that skillset to the way they work. At some level, I was really caught early in my architectural career while studying in Italy. An Italian professor talked about a tradition in architectural education there around a thing they called the scherzo, which means ‘the joke.’
The scherzo was an absurd brief that you would build. The intention around that was to strengthen your skills, have a little fun, and not take yourself too seriously, but at the same time, in its absurdity, some truth would come out that would be valuable for you. Design has that capability and ability to look at the situation and condition of humans, evaluate it, create something new, and in that effort, you can start to discern potential new features. From those, you can make an intentional choice about which one you find preferable, for whatever reasons that make sense to you.
So, I think those are the superpowers that designers bring to the table. Other disciplines may attack problems in different ways and find different solutions and paths to the future, but, yeah, that’s one that we do well.
Jorge: I’m gonna recap to make sure that I got it right. One was this notion of knowing through making somehow. You mentioned abduction. That’s a word that most folks might not be familiar with. I think that most people know of deduction and induction as ways of knowing the world, right? Abduction is one of those, right? Like a way of knowing the world. But, it’s one that involves making things that push the boundaries somehow or give you a different way of understanding how things work, but through making; through this idea that you’re gonna make something and then see how the world reacts.
Greg: Yeah.
Jorge: Then the other one that I heard here was the notion of being open to crossing the streams between different fields somehow. You talked about juxtapositions and collisions of ideas or bringing things together that might not be obvious to folks. What I’m getting from this is that in so doing, in using these skills of knowing through making and bringing ideas together that might not be natural bedfellows, part of what design does is it helps expand the boundaries of a possibility space somehow. Is that fair?
Greg: That’s a much more articulate way of saying it.
The State of Design Today
Jorge: My perception is that there has been a tendency in, let’s say, the business world at large to embrace design as a function of the organization. We’ve gone through a period where companies have invested in building teams of the sort that you have led.
Greg: Yeah.
Jorge: I sense a change in the weather regarding this stuff.
Greg: Yeah.
Jorge: There was this article recently in Fast Company by Robert Fabricant where, I think, the operative phrase that people might Google for is something like “the big design freakout” or “the great design freakout.”
Greg: Yeah.
Jorge: You were one of the people cited in this article. I just wanted to get your sense of the state of design currently with regards to the larger world of business. Where do things stand?
Greg: I think there’s a couple of things going on. If you look at the number of designers in leadership positions and the size of teams in organizations, historically, there are more people doing this work than ever before, and they’re producing value and creating incredible experiences and business opportunities for the companies they work for. I think there’s a fair number of people who are frustrated with some of the downsizing that’s happened. Many designers have been let go as organizations have shifted from growth mode to profitability mode based on market conditions.
But if you look at the aggregate, the community is in a much better spot than it ever has been. Yet at the same time, there are more of us who do this work. There is a supply and demand issue right now, which is there’s more supply of talented people than there’s demand inside organizations. That is causing a degree of churn and frustration in the design community, which I understand. If you’re looking for work and have been looking for work, it can be a scary prospect, and people are trying to understand it.
I also think there are many people in the design community who have grown up in the last 10 years and never experienced a downturn because we’ve had, since 2009, 15 years of growth until recently. I’m old enough, and I’ve been in this career long enough, that I’ve experienced multiple downturns before. They used to happen like every five years, not once every 12 years. So I do think we’re in a moment. That’s one thing.
I do think structurally, there’s a couple of things that are different this time. I think the market has placed a lot of pressure on SaaS companies, enterprise software companies, and companies in general to be profitable over growth at all costs. When we were in the growth at all costs mode, companies just hired people. They were like, “We need more people so we can do more things, and more things are going to allow us to grow faster, and it doesn’t matter if we’re losing money.” Now it matters. The bottom line really matters. So that’s one.
Impact of AI on Design
Greg: The second thing is, if you go back to November of 2023 and the announcement of ChatGPT to the world, you have sort of an AI arms race going on inside software organizations. Everyone is feeling stressed at the leadership level about being left behind and their products becoming irrelevant or consumed by a competitor’s product that may have AI in it.
So the investment, let’s say the strategy investment, inside organizations has pivoted hard towards either AI engineering or AI efforts or paying for AI systems, etcetera. And that has placed operational cost or OPEX cost pressure on organizations. The organizations look at what everyone else is doing, and they’re making decisions—whether the right decisions or not—the decisions that they’re making are to cut costs. You’ll see it in marketing organizations being downsized. You see it in design teams being downsized. You see it in some product teams that may not be in the AI space being downsized. You see it in engineering teams where certain engineering skills are not valued as much as they were before—they’re still valuable, but they’re not as valuable as the ones that they’re looking for. And I think a variety of those things happen.
I think there’s another aspect, which is a playbook that happens inside large organizations. The moment organization X downsizes 5%, their peer looks at it, and they have pressure from their board to do the same thing, right? And so, these are all sort of cyclical things.
Now, what does that all mean for us? My belief is this is also the accumulation of a change in tooling that’s happened over the last five years, even before AI. We’ve had AI for 10 years, but the kind of generative AI that’s more recent—within the last two years—the tooling has changed the way that we work, right?
The Role of Tools in Design
Greg: So if you think about Figma, Figma allows teams to collaborate and scale from a design perspective, work in high fidelity from the start, and move very fast. Then, you have the growth of product tools like Aha and Asana that have helped product management teams manage their backlogs more effectively. And then if you look at the growth of API-driven software as a service, right? So if you’re a contemporary, digital-native company, you build as little software as possible. You go to AWS, Google, or Azure, buy services, connect them, and spend your time making the connections versus writing code from scratch.
So all of these things are really changing the materiality of how software is built. And I’ve talked about this for a long time. Eight years ago, I had a talk at Enterprise UX where I was seeing these trends start to show up, and they’re here right now. What it really means, from my perspective, is that how to build is much easier, and you can build much faster. And that puts pressure on the craftspersons in the organization. That can be the designers, but it can also be engineering. The question that is most paramount is what to build and why, and the intentionality of it.
I don’t think we’ve had a good conversation cross-functionally amongst the people who build software—product managers, engineers, designers, and researchers—around what it means with these new tools and who does what. I think that people who can discern an outcome that customers or users value, show what that looks like, and then create it quickly are going to be the winners. And I think there’s a huge role for designers to play in that. However, I think a lot of designers over the last 10 years have been much more focused on the craft side of design, and there’s nothing wrong with that. In fact, craft is really important. But they have not had the opportunity to, or have chosen not to, or they’re not interested in looking at how to use design methods to see what those potential options should be—what the idea should be, what is the product, what is the product’s value, what are the downstream consequences of that product, what are the decisions, and how are people going to evaluate it?
And so I think we’re in this sort of seismic moment where we have to figure that stuff out. And once we do figure that stuff out, I think that unleashes all kinds of opportunities for net new things to happen in very different ways. I was joking with a friend of mine the other day that I should start a company called Generic Software. The idea would be direct-to-consumer versions of favorite software applications, sold at a significantly lower cost. Because you could do that. It’s like generic drugs, right? Instead of getting brand A, you get the generic, and the generic is just as good, right? Brand A took 10 years of effort to build and has a lot of incumbency around goodwill and a lot of things happening with how they’ve gone to the market and sales. There are a whole bunch of things around selling software that are really important. But someone could come in with a different business model and build something very quickly that actually adds value.
Jorge: Have you seen the stuff that the folks at Basecamp are doing with their ONCE brand?
Greg: No, I have to take a look at that. Yeah.
Jorge: Yeah, they’ve recently started a new line of products, which they’re branding ONCE because the idea, it’s like they’re doing counter-messaging to the SaaS model where the idea is that you pay them once, and you get a software application that you can install on your systems. But it’s very much what you’re describing here in that they are, the way I’ve heard them talk about it, they’re picking software categories that can be made generic. They started with a competitor to Slack, basically a messaging app.
Greg: Yeah. Yeah.
Jorge: I think they’re charging something like $300 or $350 for the license for this thing. But I have a lot of thoughts based on what you just said. You were recently at the Figma Config conference, right?
Greg: I was, yeah.
Jorge: I got the sense that it was the first big post-pandemic design gathering. It felt a little bit like a gathering of the tribe. And I thought that it was interesting that it was happening around a tool and a tool that I see as primarily being about screen-level design, which is, not like you’re describing design that happens upstream from that phase of the process where you’re deciding what kind of thing to build, not how it will be built, or how it will look and feel.
And I don’t know that there’s a question here, but I’m trying to feed you ideas to get your reaction on them. One of the things that was announced during that conference was a new AI-powered capability that would basically design screen-level stuff for you. Shortly after the conference, somebody prompted it to design a weather app, and it produced something that looked remarkably similar to Apple’s weather app. It seems to me like that’s par for the course for using generative AI at that level of design work. And people are commenting on that, on, like, this thing has clearly been trained on Apple’s thing.
But the other thought that I had when I saw that—and I’m wondering if this is one of the other factors that feeds into these trends that you were calling out—is that a lot of things are looking like they were designed by Apple. Even the ones that were designed by humans. I’m wondering the degree to which standardizing around things like design systems and frameworks has led to a homogenization at that level, at the level of screens.
Greg: Yeah.
Jorge: I don’t think it’s an accident that our AIs are spitting out things that look like things we’ve seen before because everything is looking like things we’ve seen before.
Greg: If you look at enterprise software, everyone’s got the left rail now. You have the left rail, and then there’s this sort of best practice. And so the difference structurally of one application to another is primarily in the decisions around color and branding. But the actual shape, the size, those things, there’s so much homogeneity across enterprise software for sure.
I remember, in the early days of building for the web, the degree of just wild ideas that were out there. And there’s some reason why we’ve converged on this set of overall experiences—because they work for people. They’re familiar, they’re learnable, they’re easy to support from a structural standpoint, and the engineering for them is more straightforward. So there’s nothing wrong with that. I think one of the things that people forget is that design is not just what it looks like, it’s what happens when you do something. And I think the difference between good and great is that someone intentionally figured out what happens next when I click this button. Or, what’s the step in the process?
As an example, one of the things I started doing at Compass, and then I insisted on at Cisco, is I wouldn’t look at screens without seeing flow diagrams. I need to understand the path that people are currently taking to solve their problem, and then I want you to show me the path that they’re going to take with our solution. And I want to understand where we’ve decided to innovate or make it better and where we’ve decided to just leave it similar and why. Because sometimes the reason is, it’s really hard to make that step better; there’s an engineering issue that you just don’t have the capacity to solve. Sometimes it’s laziness. Sometimes it’s people haven’t really considered… sometimes it’s a step that was added that actually no one ever asked for, and you could actually eliminate, it’s just adding friction to a system.
Sometimes it’s a step you need to add because actually, friction is something you want to have in that moment. You want people to take the, as an example, I think one of the things I think is really important in the AI conversation is if you want humans in the system, then you have to give them a moment to understand the response and then make a decision. And if it’s a decision that has importance, you have to make that clear to people.
So I think those kinds of things are still there. I think they’ll always be there, and there’s plenty of room for that. Figma doesn’t solve that. It solves it a little bit like—some of the demos they showed at the event—they showed building an automatic interactive prototype based on just a bunch of screens. The system understood that button X would probably drive screen Y, and so you could see it build flow maps for you. And that was cool, but it still doesn’t take the behavior design of a system. I think that’s one of the things that structurally we as a community should be better at. And some people are good at it. I think that there’s been a lot of conversation around—you and I have a bent on information architecture—there’s definitely people talking about information architecture again right now. And what does that mean?
I am a big believer in this idea of invoked experiences. I think what AI promises is more contextual experiences. It knows a lot about you. It’s following what you’re doing. It’s helping you along the way, assuming you’ve given it permission to do that. Why do you look for something? Why can’t it just make you an offer as to what to do next, based on something that you would value? And then that next thing that you do, does it need to have a permanent location in the application, or does it just show up when it needs to and is invoked? I think we have an opportunity to really change the kinds of experiences that we have so that they’re less linear and less “every feature has a location,” which is the homogeny that we live in right now, right? Every feature has a location, and we can simplify things for people.
Jorge: Yeah. One of the things that I’ve observed in interacting with generative AI—and I see this primarily in writing, but I expect it also maps to what we’re talking about here—is that because of the nature of how things like language models are trained, they’re trained based on an existing corpus, and the existing corpus, for the most part, reflects fairly conventional ideas, fairly conventional ways of talking about things, right?
One of the net results of this is that by default, if you don’t prompt it otherwise, the results you get are fairly conventional and middlebrow. And I’m imagining that an application that—I haven’t seen a demo of that feature you talked about, but I can imagine that an application that uses AI to predict what flows might emanate from buttons in a UI would do so in a very predictable, expectable way that mimics the way that these things usually work, right? So if you see, I don’t know, if it’s a button that says, “Contact Sales,” you can imagine what form that would lead to and what fields would need to be on that form. That would lead to—you used the word homogeneous—experiences that all start looking like each other.
But I think the broader point here—and this goes back to what you were saying earlier—is that this is a trend that suggests that design might need to move upstream to work at a higher level. Delegate the screen-level stuff to the tools and focus on deciding what it needs to be and how it’s going to present to users, but not at the screen level, how it’s going to—what kind of thing is it going to be in the world, right? Does this mean that we need fewer designers?
Greg: I don’t… Yeah, so I don’t think so. I think in the short term, it might. In the long term, I had one other Aha at the event. So they had a—first of all, Config was not a design conference from my perspective, it was a tool conference. But there was the convening part and the vibes of seeing your friends was great.
But it is not a design conference, as much as they invited really cool people and they talked about design, but they also talked about product management. They talked about engineering. They talked about… Figma is in the business of growing their business. But they had—I cannot remember his name, maybe you know his name—but the guy who is the founder of Teenage Engineering was there and he gave a keynote. He reminds me of a star architect or an auteur, right? Or a famous musician or something like that. And he’s just doing what makes sense to him, and then people are buying it, right?
My perspective is that as the materiality of the software becomes easier to compose, we will start to see Slack alternatives that I prefer over Slack, even though I love Slack. There’ll be some, and Slack’s probably a bad example because Slack has advantages of the platform, right? There are certain systems where the graph of the application enables its incumbency and value to be created because the more people participate in it, the more valuable it is to the people who participate in it. I’m not saying that. But on certain kinds of applications, there’s an opportunity for us to have diversity of things. And then I think that gives an opportunity for designers to create stuff.
So I don’t know. I don’t know the answer. I definitely believe in the upstream conversation. I think there’s a great opportunity for designers to be much more involved in the decision-making around what products to build. And that means learning some new skills around business and go-to-market and how to talk to customers. Those are things that some designers don’t want to do. But if you do learn how to do them, I think there’s opportunity.
Skills for Future Designers
Jorge: Could we spend a few minutes narrowing down on that? Because I wanted to ask you, if someone is getting into design, I’ve been teaching new designers for the last six years and it is a time of change, right? Could we enumerate what those skills might be that designers might need in the future?
Greg: Yeah, it’s a great question, by the way. I went to Config with my daughter, which was an amazing experience, right? She’s a senior studying design at Cal Poly University. And it was her first design or tech conference. It was really great to see it through her eyes. She is, I would call, an AI native. We have this term “digital natives.” I’m not an AI native. I’m learning AI. I’m using AI in all kinds of interesting ways. She uses it for everything and she’s super clever about how she uses it, right? She’s using juxtapositions of different tools to do different things, to solve different problems.
So I think skill number one is just, don’t avoid these tools; be conversant in them, because they’re here and they’re not going away. And by the way, they’re great partners if you know how to use them well. I’ve started my own Substack and I’ve been doing some writing, but I’ve been writing with Andros Claude, and that’s been great for my process. Like, I have a couple of ideas. It tightens them up. I keep working on it, and I find it useful for me, right? So I think that’s one space.
The second thing I think is, many people in our community lambasted the design thinking canon as non-designers entering the designer space. I think they missed what it was. What design thinking in my mind is, is a set of activities that allow you to stay close to your customer. So there’s the user research part of it, but there’s also co-creation activities, testing of ideas, and the art of synthesis, right? So really trying to understand the signal from the people that you’re trying to solve a problem for. And that still matters. So I think it’s really important for people to be there.
I’ve never been an orthodox person around roles, so I think designers should do research. Researchers are amazing, and researchers who only do research are great because they have deeper skill sets and more tools, but I think designers need to be great researchers too. So they need to be great at research.
I think you need to understand something around the business outcome for the companies you work for. I think you have to understand how the company makes money. At the end of the day, if you work for an organization, it has a set of objectives and goals, and it has to pay for them somehow, including your own salary. You have to be aware of that, and you have to be aware of how leadership makes decisions when stressed and when not stressed around the financial prospects of an organization. Because if you understand that, it’ll help you navigate and understand things. Sometimes you won’t like the answers, but you’ll have empathy for the decisions that people have to make because business leadership is hard. There are a lot of decisions that people have to make that are confounding sometimes.
So I think those are a variety of things. And then at the same time, still be an avid maker. I think making is the path to showing where we’re at. There’s one thing I learned for myself, which is, different disciplines come to problem solving in different ways. At leadership levels, you often find yourself in a meeting where they want to make a decision coming out of the meeting.
They’re going to make a decision with the information that they have. And usually for designers, that’s really hard because we usually need to make something to understand it before we have an informed opinion that we feel confident voicing our perspective on. So we can be in that meeting and one of two things can happen. We can be on the sidelines because we’re like, “Shit, this is moving too fast and I don’t understand it.” Or, “I don’t think I have something to offer.” Or we could be saying, “Whoa, slow down. We don’t have all the insight. We need to figure this out.” That gets interpreted as a negative from your peers. They’re like, “Hey, we just, we need to keep momentum moving here. We need to walk out of here with a decision.”
So one of the hacks for me is I usually spend time before the meeting making something if I know the meeting is really important. This way, I am able to operate at the level in the room and say, “I think this is a good idea,” or, “I don’t think… here’s what I would do or here’s my perspective,” so that we can leave the meeting with a decision. Or, in the case where the decision is being made in a way where it’s a one-way door and we haven’t really articulated the impact, I’ll ask for 24 hours or 48 hours or some period of time, but not an infinite amount of time. You make that something that you’ve done before, and so people understand it and they understand what you mean by it. You say, “Look, we need to do an exercise to validate this assumption, and it would be highly valuable for us. It will take us this amount of time. We’ll do it with you. Are you willing to do that? Are you willing to postpone the decision for two days to do that?”
Jorge: I love this list of things. You started the conversation by describing yourself as a possibilitarian, and I love that these are guidelines that open up possibilities for folks, right? If you don’t get wrapped around the axle with things like job titles and descriptions. So thank you for that.
Closing
Jorge: Where can folks follow up with you?
Greg: I’m on LinkedIn. You can connect with me on LinkedIn. I almost say yes to everybody except for financial planners and folks who want something from me. I am on Twitter very rarely these days. I’m on Threads, @gpetroff. I just started a Substack space called Improbable Futures, so I’m starting to do some writing there. If you want to follow me there, that’d be fun too.
Jorge: Fantastic. I’m going to include links to all those in the show notes, and of course, I’m going to follow myself, as I always do when you post or share presentations. Thank you, Greg, for being with us and sharing your knowledge, insights, and expertise with us.
Greg: Jorge, I appreciate it. It was great hanging out with you today.