Kill the Org Chart: Building Skill-Based Design Teams at Notion
Checking session availability…
Hang tight while we load the latest updates.
When growth accelerates, hierarchies slow you down. Randy Hunt shares how Notion’s design organization engineered itself around skills and behaviors rather than titles and functions, a shift he calls a return to the fundamentals of “making and building.” As adoption soared from personal users to global enterprises, this model helped the team stay fluid, creative, and connected to the craft of design. The talk reveals how a skill-based culture fosters adaptability, ownership, and speed when predictability disappears.
Takeaways:
- How to structure design teams around skills, not roles
- Preserving creative DNA through growth, unpredictability, and complexity.
- Why rediscovering fundamentals — curiosity, craft, and shared behaviors, enables sustainable scaling.
Kill the Org Chart: Building Skill-Based Design Teams at Notion
Randy Hunt at UXDX USA. Video: https://youtu.be/q_0EOrCYeNE
Readable transcript: edited from the recording's captions for readability (fillers and false starts removed, punctuation and section headings added). Wording is the speaker's own. Timestamps are positions in the video. Names marked [?] could not be verified against the audio.
Introduction and the Notion design team
[00:00:08] Randy: Hi everybody. Well, thanks for having me here. This looks aggressive. I like it.
[00:00:17] Host: I'm going to take your stools away.
[00:00:18] Randy: Oh, cool. Should we leave one, just to have a comedy effect here? Okay.
[00:00:23] Host: Yeah, yeah, sounds good.
[00:00:25] Randy: That's the funniest thing I'll say in the next 20 minutes. So, well, this looks aggressive. Maybe I'll just tell you about myself for just two seconds. I support the design team at Notion, which is a small, mighty, talent-dense team, which is maybe some connecting bridge between the panel discussion you just had and some of what I'm going to share with you. I also chair the graduate design program, the MFA design program at the School of Visual Arts here in New York.
[00:01:00] And also, when I wake up in the morning, I grab my little rescue mutt dog by the face and pull her face really close and tell her she's beautiful and special. I'm sharing that with you to put in context that I promise I'm a nice person, even if you don't like some of the things that I'm about to say now.
[00:01:18] This is the Notion design team, or most of it. I think there are maybe five people that didn't make it into this picture. This is from maybe two or three weeks ago, a team offsite. We do these a couple of times a year. That's it. That's pretty much the whole team. For me that has an interesting relationship to a lot of the teams I've led and built in the past, and to the opportunity I have, a couple of years into Notion here, to really align, I would say, my personal values and motivations with a lot of my learned experience, with the opportunity ahead of us.
What does team size have to do with anything?
[00:01:55] In the professional dialogue, and I'm sure it's happening in between sessions here, you're asking each other questions about your professional environment. It's been 20 years now of that happening for me, and if I rewind 20 years, I feel like a lot of the questions we would ask each other at that time, probably because of the macro conditions more than anything else, but maybe also because there were a lot of people new to some of these kinds of roles, were things like, "How big is your team?" Or, "Are you hiring?" And then the follow-up question would be, "How many people?"
[00:02:32] You might even talk to a recruiter or someone somewhere else, and they would triangulate your qualification for doing something, maybe delivering value to their customers, though they might not be seeing it through that lens. They'd triangulate that, or use some proxy, from how many people you managed, or how many people were in your org, or how many layers there were.
[00:02:55] I'm very grateful for a lot of the experiences I've had. Some were scaling teams very quickly, some scaling them very methodically over a longer arc of time, inheriting teams and making them smaller, inheriting teams and making them bigger, hopefully in every case making them better. From orgs of 150-ish, larger, that were very multi-functional, to, at one point a few years ago, going back to a super IC role for about a year, for the fun of it. I formed some perspectives and opinions, and I'm putting those into action now, I think to great effect. So let's talk about that a little bit.
[00:03:43] What does team size have to do with anything? That's really the question I started to come to. I was like, this just seems like a bad proxy for lots of stuff. In fact, in my grumpiest form, I was like, when we ask this question, we're basically starting with: how much communication overhead do you have? How expensive is your team? All of the, let's say, burden or cost or responsibilities. Of course there are other sides of this. At a personal level, it's awesome to create opportunities for people and employ people. And sometimes the solution to the problem in the moment is just to put people on it. But it's an expensive, risky thing.
[00:04:27] I noticed that even in the biggest, most successful teams, or the biggest teams we built, the most successful pockets of those teams functioned just like a smaller team that had only a handful of people on it. So I've come around to this different perspective, which is that team size matters a whole lot, but for different reasons than I used to think.
The further you are from the work, the harder it is to learn
[00:04:48] We can call this Randy's Excellence in Text-Based Data Visualization TM chart over here. The further you are from the work, the harder it is to learn. And I mean learn in almost every way: learning about your customers, learning about the effectiveness of the choices you make, learning about your practice or craft, learning about the people around you that you work with, or really your organization more broadly. It's very, very difficult. I'm sure some of us in this room, or maybe many of us, have occupied some or all of these at many points in our careers.
[00:05:26] They're often aspirational. They're often put at the top of diagrams. I have such tremendous and profound respect for individual contributors, and all of us down here are overpaid and unnecessary. That's an extreme statement, but I think if I had to pick, I would say that. So the story points a lens at myself just as it does at everybody else.
[00:05:55] So why does this matter, that the further you are from the work, the harder it is to learn? It matters because this moment demands learning. Now, we could say that about any time, maybe, but when the ground is shifting underneath you, there's technology change, macroeconomic conditions, stuff is happening, stuff outside of our control, and then there's stuff we can control that's happening. We've got to learn. I want to be closest to the learning opportunities, and I want everyone on our team to be closest to the learning opportunities.
[00:06:27] When growth accelerates, predictability disappears. When change accelerates, predictability disappears. So you have to have learning opportunities, tight learning cycles, and the ability to apply those learnings. There's a mindset part of this. We've probably seen some things go around that say the future favors the curious. It's probably always true, but it works well for this time.
Welcome to the great blurring
[00:06:53] Welcome to the great blurring. I'm sure yesterday and today there were several parts of this dialogue, and in the panel discussion we just had: what does it look like when multiple roles are getting fuzzy with each other? Tool expertise is becoming less relevant, ish. Of course, expertise in emergent tools becomes a powerful differentiating factor that I think we underappreciate in the moment it's happening. Everything gets displaced and evolves. This maybe felt stable for a decade or something.
[00:07:27] What I think is interesting about this is that we could have said the same thing 10 years ago. We could have said the same thing 20 years ago. Now, there are conditions that are different today, but most of the great products that I admire and respect, and most of the people I look up to, from a distance I would say, but also those that I've learned the most from directly through collaboration, behave like all of this is blurry all the time anyway. Regardless of job descriptions or professional aspirations or anything like that, it's just how they choose to show up in the world, because they're interested in something else beyond this.
[00:08:13] For the people who have that mindset and are motivated by a particular set of things, this is what we've been waiting for. This is actually the moment that empowers and reflects the way we've been trying to work, or the way we've worked at peak moments: the moments of deep satisfaction or great joy, where we see value created, where we surprise ourselves and have a whole bunch of fun. I think it's happening again right now. At least for a subset. I wouldn't stand here and tell you how I think you should feel, but I'll tell you how I feel, and hopefully transfer the reasons why that's true.
[00:09:01] If you're the kind of builder who, how would we say, fell in love with making stuff on the internet, I think we didn't have the back button, yeah, you ended up blurring a bunch of those things without being aware of it. In fact, before you learn the vocabulary, before you learn the disciplines, before you learn the stages of work in professionalized ways, or the different tools and stuff, if you just start making stuff, you never learn that those boundaries should exist anyway. It's blurry in your bones.
[00:09:41] I think there's something really beautiful about that. There's something really human about that. It's why metaphors like garden tending, home renovation, landscape architecture, there's a bunch of these metaphors we use to help us understand software design practice, resonate with us. In part, I think, they resonate because we can sense the wholeness or fullness of those things in a different way, maybe when we're a little distant from them.
Smaller, flatter, blurrier teams
[00:10:17] I think the new team, and this is the spicy version, by putting the word "new" in there, because I would say the optimal team, for as long as I've been doing professional work, or even just caring about putting stuff on the internet, really, was smaller, flatter and more blurry than we tend to think is true, or than we tend to operate by, or tend to prefer or orient around.
[00:10:44] This has a bunch of challenges in it, because blurry can make people uncomfortable. Of course, there's work we can do to add precision within things, and nuance. But when we say, "I can do a little bit of product. You've got to own the strategy. Yeah, you do the design. You're doing a bunch of front-end stuff. Who does the writing? Oh, the designer does," people get really uncomfortable with that. It's sort of like, "What's your job?" Because titles, and even teams, are almost like wayfinding inside an organization.
[00:11:19] But the bigger your organization is, the more wayfinding you need. Yuck. You know what it feels like, even when there's good wayfinding? How many of you flew into New York to come to the event? Did you land in JFK Terminal 4? Who took Delta? Any of the B terminal? Even to the B gates at Terminal 4, it's like a 25-minute walk. You're like, "I'm pretty sure this is how long the plane was in the air between here and Detroit." It's long. And even though the wayfinding is quite good, you come to feel that the scale doesn't feel right. It's almost too much. I'd rather have less terminal and less wayfinding.
[00:12:00] We're seeing this in practice at Notion. How would I say this? I joined Notion because I had a strong conviction that these principles were already true, and ready and willing to be practiced with enthusiasm, I'd say. So the story I have, or the experience I have at Notion, isn't one of radical change from big team to small team. It's one of focus and intentionality and conviction about a particular way of doing things that we believe is highly effective. It happens to be aligned with what I believe is highly effective, and so those two things together become quite enjoyable.
Band of misfits, mature daily makers, lovers of software
[00:12:53] When I think about the people we invite onto Notion's design team, and how we look at the spirit of the team, think about these three attributes. One is a band of misfits. What I mean there is that people can come from anywhere and have any background. For the thing we hope they do with us, we only need to know that it's possible and see evidence that they can do it. It doesn't matter if they studied it. It doesn't matter if their previous job title reflects it. It doesn't matter where they sat in the organizational chart. Those things can be signals for stuff. They can be helpful, but they're not the reason we would do anything or not.
[00:13:35] Meaning, we've hired engineers onto the design team, people whose title is engineer, and you look at their resume and it's engineer. But you look at the output of their work, the way they think and behave, and their desire for what they want to do, and really what they want to do is make software. They also happen to have great taste, and can deliver on that taste. Sounds like a designer to me.
[00:13:57] The other is that, and this is why small and talent-dense can work, these all add up. Nothing here is extraneous or erroneous. They're mature daily makers. What I mean is that they wake up every day and have a practice and ritual of making stuff happen or exist. They're compelled to make things. It's not because the job says I must do these things today. It's in your heart. Which can sound a little soft, but I think it's so important.
[00:14:31] How do you get momentum and enthusiasm from a group of people who work extremely hard? We're intense. It feels like pro sports, or F1 racing or something. It's like, better get a good night's sleep, because tomorrow we're going to design. That kind of energy can only be sustained, or people only have an interest in it, if you're compelled and enriched to do the thing, whether or not someone else asked or wanted you to do it. And that exists, I would say, for the vast majority of the team, in and out of software, or in and out of design. They bring that spirit to most things in their life.
[00:15:17] The other, and this is an important distinction, I'm going to give you a spicy-take sub-story here, I think: they love building software. And I mean this truly. We think software is cool. We think it is fun. We are delighted by the magic and power of working with software, building software, shipping software to users. It's awesome. Because we care about it that way, even the hardest things feel like play. Even the hardest things feel like intellectual stimulation. It gives you a great sense of pride, because we're doing something we care about.
[00:15:52] There's a distinction here, and here I think is the spicy part. And maybe there's no one way to do things; there are different ways. But our way, I would say, doesn't feel like a comfortable fit for someone who is anchored in other parts of software making, and unwilling, or uncurious, to push beyond the blurry bounds. What I mean is, no one on the team, well, the coming-through part isn't important, no one on the team holds an explicit and rigid process in their mind. They don't think there's a right way to do UX or a human-centered design practice. There's not a right way to incorporate research, or to interface with customers. There's not a right way to continuously deploy and develop the software. We're very, very interested in almost the culture of software building.
[00:17:06] I think over the last maybe 10 or 15 years of the tech industry, we had some incredible ramp-ups. Software ate the world. Macro conditions were very favorable: super low interest rates, blah blah blah, lots of VC money. I think anyone who was practical and reasonable about increasing their earning potential or other things, and saw where the job opportunities were, came into part of the software-making world, for reasons I don't judge and I think are perfectly great. But I think there was a gravity, or magnetism, towards the work of product making and building for reasons other than loving building software. Again, those are completely valid reasons. I don't challenge those, but the spirit of it is different.
[00:17:57] It is maybe more professional and professionalized. Again, not a problem. Maybe more careerist. I don't judge that. We all want to feel like we're making progress. People like to earn more money. They want to feel job security, blah blah blah. All normal, reasonable life things. It's just different from doing the thing because you love it so much and would do it even if that wasn't the case. So important. So important.
How we recognize these attributes
[00:18:24] So, we know it when we see it, isn't it? Well, that's fine. How do we know people have these attributes when we meet them? One is that they've shipped ideas, many of them. For us, it's somewhat arbitrary, or doesn't matter, whether these are in a professional or commercial context or they're hobbies and side projects. You've probably heard this narrative, this kind of identifier or qualifier, many times over the last couple of decades, because I think it holds. It's actually not a new idea.
[00:18:55] But people who have ideas, go after them, develop them, get them out into the world, possess an entrepreneurial spirit, an ability to solve problems and make things happen on scarce resources, the grit to follow through and make things exist, because it's often not smooth or easy, and then do it several times over: that's just a strong indicator of the ability to get inside, for us, a blurry environment and figure stuff out. And that's okay and intentional. It's not blurry because we didn't create structure. We didn't create structure because we wanted it to be blurry.
[00:19:34] The second is that there's really strong evidence of intrinsic curiosity, the motivation to learn, and incorporating that learning into the work. This can happen at the customer insights level. It can happen at the tooling and process and methodology level. It can happen at the cultural and strategic level. It can happen at the audience or segment level. And of course you see a bunch of those things happening across the body of work. It's quite impressive. This is really where you end up with people who are hyper-generalists who also have depth in several areas. And it supports this idea that people can come from anywhere or any background, because the origin doesn't matter so much. It's the collection of experiences over time which you stack up through this mindset.
[00:20:22] Then this is another tell, I think. When we're talking with people, we look for enough vocabulary for people to talk with enthusiasm about the things we care about. It means they really know about it, so you can riff, you can jam, you can improvise. It's part of the environment of getting the work done. But free of hype. When everything shifts, it's very easy to even unintentionally get into the hype cycle. It would be very possible for any of us to go out into the hallway and have a conversation about the headlines of LinkedIn posts we've seen over the last 10 days. Yuck. Ooh. And yet it happens.
[00:21:01] Then something floats up into whatever Fast Company article, and then it's a bigger distribution. Ah, it's a cultural theme. All of this stuff's going to happen, and we talk about it, but I think the tacit knowledge of doing the work itself, having direct learned experience and forming opinions about it, those are the conversations you're going to have. When I see that people have enough of the vocabulary, and they've earned the language to talk about the thing by doing it, it's a very strong indicator that they're a good fit for this way of working, or the spirit of this team.
Three elements of executional craft
[00:21:36] So I've talked about the vibes, maybe, the feeling or spirit of it. But the other part that's really important here is executional craft. This was maybe the topic du jour last year, before everybody freaked out about AI. My definitions, I don't know if they match yours. That part doesn't matter so much, but it's important because they tie into everything; I told you, nothing extraneous here. I think there are three elements of craft that we look at and intentionally pay attention to, try to nurture and develop and support.
[00:22:14] One is that the people in our talent-dense team know the difference between good and great, and act on it. They act on closing that gap: in their own work, in other people's work, in what they see, in what customers are experiencing, in what they might sense from the zeitgeist of the market. Some of it learned, some teachable, some mix, and we can debate that. Two, their skill and technique itself. That allows you to do the acting on it well: the mastery of the tools, the methods, et cetera. And it is, again, earned through doing it.
[00:22:53] That gets you to the third part, which is the habits and rituals. This is doing it every day. This is the mature daily maker who loves doing the things. They're going to cycle through it over and over again, because it's fun to do that. I don't know, there's a dopamine hit when you try some stuff and you see the results of it. It used to be, whatever, scratching through some HTML and refreshing your browser in 1998. And now it's some carefully crafted prompts and some agent interaction, and in some ways I also just refresh my browser and see what happened. That part doesn't feel that different. What's old is new again.
[00:23:35] But doing this over and over again is what feeds into the other two. The skills and technique develop, and again, you develop the taste. It allows the team to also operate in a way where, how do we say it, highly skilled people like to be surrounded by highly skilled people. And so this too allows the team to stay dense and tight, because there's a shared appreciation for the level of work. So the communication overhead, and the need to manage that work into that level of precision, is also lower. It still exists. There are still disagreements, there are still explorations, but a bunch of these things are done to reduce overhead and bandwidth, and all these things allow the team to stay small.
Momentum, empowerment and ownership
[00:24:25] This is important because momentum creates progress, and progress is what matters to the business, or to your user. For us, we think much more about the user. If it matters to the user, it matters to the business, but I think it's safe to make these interchangeable. Smaller and flatter teams enable momentum, and momentum drives progress.
[00:24:51] I don't know if you have read The Mythical Man-Month. I don't know how much people pay attention to this book anymore. It's a software engineering management book from, is it Fred Brooks? Is that his name? I think so. Basically, the premise of the book is: if you've got a big, complicated project and it's moving too slow and you want it to go faster, take people off, don't add people. There's a lot more to the book to read; that's not a fair summary, but that key point aligns.
[00:25:16] None of this works unless people are empowered. This is where the leadership behaviors, and really the cultural alignment, happen. If there's trust and access, people can do their best work without asking for permission. Small team, empowered, go tackle things. And the other side of this, because that all sounds great and good, is that when you have that, people have to have great responsibility for the work, too. When you're playing many roles, and you're effectively the product manager of the work that came out of the design opportunity you saw, you've got to own that all the way, not just to delivery, but then to its improvement, to its maintenance. We love building software, so that's okay, we do that, but that's super real. So you've got to care about it enough. It's not about getting the agency with no exchange. It's agency with ownership.
[00:26:17] So I'll leave you with these three thoughts. Empower teams around skills and behaviors first, not titles or functions. You'll notice I didn't really talk about titles or functions on my blurring slide. Smaller, flatter teams stay closer to the work. That means they learn faster and they're able to adapt when predictability disappears. And finally, curiosity, craft, and these shared habits and culture are the foundation of this skills-based approach. So thank you so much.
Q&A
[00:26:45] Host: Thank you, Randy. I definitely identify with the blurry part, and I think everybody out here, some of us are designers who do product work, product managers who have an aesthetic, engineers who lead products. I've been blurring things for a good amount of time, and I think certainly the ability to actually do these jobs makes the blurring even more strange, as opposed to intense shoulder surfing.
[00:27:16] All right, so let's get to the first question. Okay, so we get the blurring of roles. Where does accountability lie, without the finger pointing in all directions? As the roles blur, you still have to get things done. And where do you investigate when things don't get done well?
[00:27:39] Randy: In a really practical way, I'd say most projects, most bodies of work, end up with one owner, in whatever industry, like a PIC or something like that. That's pretty typical and normal. It's so interesting, though. Communication overhead is so low, because teams are small, that actually hashing it out works most of the time. There's, again, nothing new; it's all old, it's just optimized, meaning sometimes people escalate stuff, it just happens so rarely. And I think the investigation part is very rarely necessary, because of something I didn't mention here: the legibility of the work, the documenting, the universal awareness, almost everything being public in Notion. That allows us to have a high degree of visibility at various elevations of the work. So there's more ongoing awareness than need for investigation.
[00:28:45] Host: Okay. We have a couple left. Do designers at Notion pair with engineers?
[00:28:50] Randy: Sure, yes. But also they don't; this is the blurriness. I didn't really mention this here: every designer on our team ships production code. Again, I would almost call them software makers.
[00:29:03] Host: Got it.
[00:29:05] Randy: Software builders, software designers. To varying degrees, but the vast majority of the team did that long before coding agents were a part of the daily practice. That's where I get back to: if you think and work that way, then this is your moment. Also, because you really understand how software is built, you can very effectively work with the new tools and understand maybe some of the tradeoffs or assumptions you're making when you do that.
