First and foremost, hello. Welcome to the Investor and Analyst session here at Config 2026. My name's Kate DeLeo, for those who I don't know. I lead investor relations here at Figma. For those of you in the room and online, we want to first and foremost say thank you for joining us here today. We're so excited by the community who's gathered here in person and online, as well as the exciting product announcements that we made today. Today we'll be joined by Yuhki, who's our Chief Product Officer, Praveer, our Chief Financial Officer, and then Kris, our CTO, will be joining momentarily. He's running down from one of the other floors. Just to orient you, we're going to be spending about 40, 45 minutes on prepared remarks in this session, and then we'll leave time for Q&A at the end. I also see our GC here, so I have to read a couple things real quick before we get started. Here is our safe harbor statement, which I won't read in full, but the statements that we make today may include forward-looking statements that involve risks, uncertainty, and assumptions as of today. Our actual results could differ materially from those set forth in these statements. Please refer to the risk factors in our most recent filings with the SEC, which are available on our IR website, along with today's slides. We will get started, and I'll hand it over to Yuhki. All right. I think I'll use this. Is this okay? Hi, everyone. It's good to see all of you, and thank you for coming. Well, I have the honor of speaking to a lot of the things that hopefully many of you got to see earlier today and contextualize around why it's important and what problems we're trying to solve. All right. So I know most of you are familiar with Figma, but just to ground everyone for a second, we're a platform for designing and building digital products, and we're very lucky to be used by many of the world's best teams building world-class products. And why I mention this is, this affords us this unique point of view around how the process of creation, the process of building digital products is changing, and we get to learn from that and build for that. As all of you know, the way that creation is happening, the way digital products are getting created, is changing dramatically. We mentioned this a little bit in the keynote, but we no longer live in a world where everyone needs specialized skills to get things built. And every digital experience has become accessible in terms of how do you create it. People who never knew how to write code can build products now. People who never edited a video can now generate one. So it's interesting to see how it's become possible to go from a concept in your idea to code, all the way to even a campaign. But with that, it has brought its fair share of challenges. This has obviously led to an explosion of outputs and in particular products, but not necessarily of high quality things. I like to show this chart because it's interesting to observe that in the gray line, you'll see that this is showing the number of app releases on the iOS App Store monthly, and you see that exponentially growing, especially in that agentic era, starting from 2025 or so. At the same time, when you look at the light blue line, it's indicating that the apps that are used very frequently, very commonly, that's not growing proportionally. In fact, it's even dipping a little bit, which is interesting. What I take away from this graph is that it's not necessarily the case that as more and more things are getting produced, they're also being used, right? The data kind of makes sense because at the end of the day, you're competing for human attention. What we see is that the bar for differentiation is moving even higher. The problem isn't just the outputs, it's also the process. I have the chance to talk to a lot of leaders about what they're seeing and how AI has changed the process of creation. While it's absolutely true that from an individual perspective, working on things, everything has become faster, at the same time, when it comes to things like working together, working collaboratively, that has become much more difficult. Leaders talk about the tunnel vision that people get into as they're speaking to their agents and collaborating and everyone going really quickly, but in different directions. They talk about the siloing, right? Where everyone is working with their local machines, building really amazing prototypes, they actually have to share it, and sharing it and getting feedback is actually really inconvenient now. What's happened is that you have these artifacts that are scattered all over the place. Collective work has become much harder in some ways. We have a firm belief that the solution to these problems is the infinite canvas. Just to unpack that for a second, we think that infinite canvas is important because, first of all, it's built for broad exploration. You can explore a lot of ideas at once. You can see them side by side. You're able to bring people into the canvas, agents into the canvas, and get a diversity of different kinds of artifacts all in one place and keep everyone on the same page. It's also built for deep exploration and deep investigation. You can actually, in our canvas, be able to take an output and actually refine it further or make it precisely yours. These are the attributes of a canvas that we think is really powerful. Over a decade, we've been building one of the most powerful infinite canvases in the world, and one that's highly performant and professional grade. When we think about the next generation of workflows and teams, and the process of creating software or any digital experience for that matter, we've been asking ourselves, how does our canvas need to evolve? I think there are two ingredients. One is new creative materials. If you imagine building any digital experience with humans, with agents, what are all the raw materials that you might need to actually build that digital experience? Second is intelligent tools. You need tools to actually shape these things, tools that you can work with your team on to actually build something that can help you push the boundaries of digital creation. These together, this intelligent canvas for a new creative era, that's the theme of how we think about our launches. What I'm going to do is walk you through what you just saw through this lens. New creative materials. I want to first talk about the first material, code. It's true that code is increasingly becoming an implementation detail. It's very much the case that a lot of people who are writing code today aren't actually writing it by hand, they're doing it through an agent. At the same time, it's still the final material for digital products, it's incredibly dynamic, it's high fidelity, it's a material that isn't just about implementation though, it's also about exploration. Last year, for those of you who were able to tune in, we introduced code more natively to our platform via Figma Make. Figma Make is our product that takes you from prompt to a prototype. That brought code to our platform for the first time. Since then, we've been pushing a lot of updates to improve the quality of this, increase the capabilities, most recently, we've been able to make it so possible for you to actually use Figma Make to point it to your existing code base, you can actually directly push changes to your code base and not just build prototypes inside of Figma Make, which is really exciting. As we've made these investments, the question has still come up around, well, do you get started in design or do you get started in code? As Dylan mentioned, there's been this kind of debate around is design versus code for the longest of times. As he also mentioned, we really don't believe that it's a trade-off. We really think that code is actually a material for design. It's a material that you can actually explore just like any other material that is on the canvas. It just hasn't felt that way in the past with our current tools. That brings us to the first launch, which is Code Layers, code that lives natively on the canvas. As you may have seen today, that looks like a code back prototype that sits right on the canvas. One of the powerful things about it is you can bring a bunch of them into the same space, you can compare and contrast. You can bring your teammates into this so they can actually leave feedback or make their own riffs. You can bring your agents into this canvas and your teammate's agents so that they can also review these, critique these, or make iterations. one of the things that's really powerful about this relative to today's world is today's world is everyone is on their individual computers building these amazing prototypes, when it comes to sharing, they have to deploy it somewhere, or they have to record it as a video and share it somewhere. That's where they get the feedback, and the feedback gets scattered. If you imagine everything is in one place and in live, you get to have these really rich conversations and be able to explore the best ideas together. We think this is how code actually feels more like a creative medium by coming into this kind of environment. The important thing about this is it's not that you only have code or design. We actually make it possible to go between these two really quickly. You saw in the demo this morning the ability to go from a code back prototype, extract a design, or extract a sequence of screens even, so that you can either make some refined edits really quickly and more efficiently and more cheaply, in a more collaborative way. You can get the bird's eye view so that you're not just looking at one screen at a time, you're able to see the entire user journey. We think these are really powerful ways to mix the ability to have visual manipulation, seeing these kinds of artifacts spread out and get a bird's eye view while also enjoying the power of code. Code is just a start, we're expanding the set of materials that you can work to, by bringing to the canvas things like movement, texture, depth, and other generative workflows. I want to talk about a few of these. The first one is Motion. This is the last demo that we did and we're really excited about. I just want to contextualize how Motion worked before just so that you can understand why this investment was really important for us. In the past, if you wanted to create a Motion asset, oftentimes people did still use Figma to create the initial static assets, they had to take it to another tool. They had to take it to another tool that was often very difficult to use, so only specialized people could use it. It was a tool that was single player, it was a tool where once you created a Motion asset, it was hard to get something out of it that was easy to export back into code. We make that all possible by having it in one place. We've also done it in a way that's uniquely Figma with multiplayer, our infinite canvas. This is a Motion tool like no other from that perspective, it's also part of your design system. You saw how investments that you've made in your design system pay off, you can create Motion artifacts that are also part of your design system so that the other people who don't even know Motion can automatically get the benefits of these higher fidelity assets. Of course, Figma agents can actually generate these, get you started so you can refine them on your own. You can actually take these all the way to production via Dev Mode and MCP so that it's not just an artifact that lives inside a file. You can actually get it all the way into your user's hands. We're really excited about what Motion is going to do in terms of accessing much higher fidelity experiences that are just much higher craft and highly differentiated. I think it's really important because when you look, the reason our community is so excited about something like Motion is it's always felt aspirational. It's always felt like you have to be in motion design to be able to access this. We hope that today feels much more accessible to anyone who's interested. This is also powerful when you combine it with some of our other features. We've featured 3D transforms, the ability to bring perspective into the canvas. There are things that, like you're seeing here, makes sometimes the UI more intuitive or interesting. When in concert with Motion, creates a lot of amazing artifacts like you're seeing here. We also talked about shaders as a new material that we're bringing in, and you can think of it like light and texture that's really dynamic. Liquid Glass is probably the most famous shader that's been talked about in recent years. If you use an iPhone and that effect that you see, that's on everything that you use. Just for context, that was a feature that we had to build at the end of last year as introducing a new shader. What we've done is made it possible for you to create your own shader, your own version of that. If you have a material that you like velvet or whatever texture that might be interesting to you can just prompt it into existence. Not only that, you can actually have unique bespoke controls that are specialized just for your texture to get it just right. As you saw Rogers play with on stage, it's really important to be able to play with these things to get things and details just right. That's why those controls are really important. Let's talk about AI also as a creative material. That's really the idea behind Weave and why we got excited by it from an acquisition perspective to begin with. Weave gives designers the ability to be precise in terms of combining and shaping AI outputs by bringing different models together and stitching it together into workflows that other people can use. What we've done is bring Weave into our platform by making it possible to run those custom workflows right from within Figma Design. We're starting with some workflows that we've defined, and we're going to let our customers define their own workflows that their own employees can use as well. These are the new materials that are bringing the canvas to life. Across everything we've talked about, I think three things are really important. The first is we're just giving designers one place to express themselves in really powerful ways. As Dylan talked about raising the ceiling, giving people this expressive material is really important if you want to create differentiated experiences. The second thing is, not only that, we're actually making it so that it actually works really well with AI. We're supplying this material to AI as well. What's important is AI can create a lot of amazing things for sure, in today's world, you have no choice but to accept it or keep prompting to make improvements upon it. If these materials exist, you can go in and edit it yourself and have full creative control. That combination of power of AI and full creative control is really exciting. The third thing is that everything is shared, reusable, and in a collaborative environment. That's really exciting because you can reuse and remix and reimagine these. Materials are only half the story, and the other half is helping your team move faster together through intelligent tools. Earlier in the year, and actually this week, we rolled out to everyone agents on the canvas. These agents do everything from creating variations of your work, helping you clean up your work or reviewing your work. The way we thought about it is we wanted to make it feel like a teammate on the canvas. You can get a lot of them going in a highly visual way and in a way that you can actually see what your other teammates agents are doing as well because you're in a shared space. One thing that was maybe a little bit subtle in the keynote is we also made it possible for you to actually really view exactly what's going on in the canvas. For example, maybe my coworker was using the agent to synthesize feedback or prep for their meeting, for example. You can actually go into these conversations, see what your coworkers have put their agent, the ways in which they've had the agent work, and learn from that and remix that, which is really powerful. Of course, we give full control, so if you don't want people to see how you're working with the agent, you can make the conversation private. We think it's going to be very interesting as people today are wanting to learn from each other and understand ways to leverage this kind of technology. If you really take the metaphor further, if you think of an agent as a teammate, then you also want to be able to teach it skills. Today we talked a little bit about custom skills that you can actually build and also share with the rest of your team. An example that's shown here is this compliance review skill, right? Brendan in the back as you see, he and I used to work at Uber and oftentimes we'd need to have lawyers review the designs just in case there's some copy that suggests something that's problematic. Well, obviously, we still need lawyers, but it's really nice to be able to encode that as a skill so that you can run that skill against your design, against your prototypes, get some initial feedback, and really cut that feedback cycle. More recently, there's been skills like someone compiled all the feedback that Dylan has ever given to us into a skill and had it review designs so that you don't have to necessarily wait for him to get feedback to get an idea of what kinds of reactions he might have. Those are just some small examples of skills. We think it's really interesting to see what that can do in terms of transforming some of the workflows that exist today. We also talked about not just that, but the ability for agents to actually build tools for you. Figma has always been extensible, meaning that people are able to build custom plugins to make the canvas entirely their own. Oftentimes that's been very hard because it's required you to code to actually build these plugins. Today now agents can write it for you, and by doing that, they can create custom tools that your teammates can use that are bespoke to that given use case at hand. That, for example, maybe you want a tool that extracts colors and turns it into your design system. Well, you can do that once using the agent, or you can build a tool so that everyone else can use that tool as well. Taken together, this creates a canvas that gets more intelligent the more your team uses it, and your canvas starts to understand your organization's processes and adapt to the way you work. The canvas is built for you and your team. To sum it up, we're bringing more creative materials like code, motion, 3D transforms, shaders to the canvas, and we're also making it more intelligent, and more useful the more you use it through these intelligent tools. Just to step back at the end, Figma was one of the first to introduce a canvas as a shared space for designing and building software. We think the next evolution is an intelligent canvas with all sorts of materials that you can create yourself as well, so that you can go from concept to production or concept to even a marketing campaign. That's what we're building, and we couldn't be more excited. With that, I'm going to hand it over to Praveer. All right. Oops. All right. I clicked too quickly. Thanks, Yuhki. Great to see so many of you guys here at Config, and thank you for those that are tuning in virtually. It's one of my favorite moments of the year. It gives us an opportunity to share the stage with so many of the Figmates that work on a lot of the features behind the stage. It also gives me a lot of opportunity to really appreciate the energy of the community. Lastly, I love the opportunity here to share a conversation, and a real conversation at that, on where we're going. As Yuhki just walked through a lot of the product innovations and the strategy behind them, I really wanted to zoom in and talk a little bit about why we believe this is important, why we believe this is important to the business, and also a little bit more of why now in the current market backdrop. We've been saying it for a little bit, and Yuhki brought up the same slide, that design will matter more as AI democratizes access to code. We're watching this thesis play out in real time. More software is being built than at any point in history. GitHub now hosts about 630 million repositories as of the end of last year with over 120 million repositories added or created in 2025 alone. The marginal cost of producing code is approaching near zero, and AI is absorbing a growing share of that work. Some of the leading labs publicly share that they're already seeing about 80% of their code being written by AI. As the supply of software continues to expand, competitive pressure continues to intensify. AI accelerates the path to a plausible answer, but when teams accept that first solution and explore fewer directions, work starts to converge towards the same patterns, ideas, and experiences. More code doesn't mean better products. In a world where we can ship faster, what separates great products from average ones, increasingly, is design. The more teams lean into AI, the more they realize why Design is so critical. As code becomes commoditized, the distance between an idea and shipped product is collapsing. Design is the layer where that value begins to accrue. Figma ran a 2026 AI report that we published this morning that cites that 90% of product builders say that Design is at least as important as it was before the introduction of AI, and 57% say it's even more important today. For developers, that number rises to 65%. With AI handling more of the production work and more collaborators participating in the process, designers say that they have more room to focus on shaping the direction earlier alongside PMs and developers. Product building has become less about delivering the actual artifact and more about defining what truly will differentiate the good products from the truly great ones. For Figma, we're genuinely excited about this. Putting the power of design in everyone's hands has always been our core mission and should lead to more innovation. We believe that the visual canvas is the best medium for expressing intent and for exploring ideas together through prompting and code. As we found in that same AI report that I was mentioning, more than 3/4 of all product builders say that at least half of their work ends up on the canvas, and 59% say most of it does. It's where teams can riff together and pressure test ideas side by side. Designers are on the canvas the most, but PMs are a close second. Even among them developers, one in five would rather start a project on the canvas than versus in code. The future is not code or canvas, it's code and canvas. Figma brings all the modalities together from prompting, coding, and designing visually in one collaborative space for every person on that product team. The value we have been bringing to our customers is showing up in our business. These stats are from our Q1 earnings announcement earlier this year. Our net dollar retention rate accelerated to 139% as of the end of Q1, our highest level in over two years. Among our paid customers spending more than $10,000 in ARR as of the end of Q1, over 60% added full seats compared to their prior renewal. Our MCP server, which lets agents read and write on the canvas, saw a 5x quarter-over-quarter increase as of the end of Q1 in MCP weekly active users in Figma Design. As we saw in Q1, customer usage and workflows on our platform are deepening. Across all tiers, we're seeing customers grow seats, consume AI credits, and expand into new functions and teams. As AI is increasingly embedded across the platform, our business model continues to evolve to capture more of the opportunity. At this time last year, Figma was a pure seat-based model and was just beginning to imagine what AI could do for our customers with the launch of Figma Make. With today's releases, we now have AI embedded across our entire platform, including Make and local code environments, Code Layers, which brings together prompting and agentic workflows directly on the design canvas, agents working across Design, Slides, and FigJam, agent skills and generative tools, and entirely new creative capabilities for image, video, and Motion. We're becoming something broader, a platform where seats and AI consumption can grow together. The AI credit model we implemented as of the end of March is the first step to that evolution. We were deliberate about pricing in a way that's fair to our customers and durable for the business. We're watching adoption closely, and we'll continue to adapt the model as we learn. Q2 will be our first full quarter here, post AI credit monetization. We've been investing in more features to improve the customer experience, including expanded admin controls, pay-as-you-go for our Pro customers, and more flexible contracting structures for larger enterprises. Our goal remains the same. Our monetization model should support growth rather than constrain it. Everything that you saw on stage today connects directly to how we grow. As AI continues to blur the traditional boundaries across roles in product teams, more people are participating in the product development process across design, development, and product individuals. We've created first-class experiences and added good reasons for more individuals to hold a full seat on the platform. Our latest capabilities make that value proposition even stronger. Code Layers in Figma Make bring the best of code and design to the Figma canvas, while new creative capabilities such as Motion, shaders, 3D transforms, and image and video generation through Weave expand the type of work that can happen on Figma. Together, these products allow us to serve workflows that previously lived across multiple tools and bring new users onto the platform. As more of the product development life cycle moves on to Figma, we accumulate more context that no other tool has across design, code, prompting, and creative work, all within a connected platform. The context powers agent teammates that collaborate alongside users, which helps organizations standardize workflows through agent skills, connectors, and generative plugins, and extends into the broader ecosystem through our MCP server. The more work that happens on Figma, the more powerful that platform becomes, creating a flywheel that improves both the user experience and the intelligence that's built on top of it. We believe that all of this drives credit consumption and user expansion. We expect that more users, more AI-powered surfaces, and more workflows happening on our platform will increase opportunities for both seat upgrades while driving AI credit usage across the platform. Looking ahead, we're investing to win a growing market opportunity, but we also plan to be disciplined to do it responsibly. This means continuing to deepen our investments in our AI-native workflows and go to market while also driving operating efficiencies. AI will continue to transform our business and our goal is clear: Invest to win over the long term in order to drive long-term profitable and sustainable growth. We're excited about what's ahead and really want to thank you all for coming here again today. I'm actually going to invite the three of us back up here. Not invite us, we're already here, and we'll be able to spend a little bit of time running through a few questions. If you have a question, just raise your hand. She, of course. Are these on? Okay, perfect. Amazing. Thanks so much for taking my question. Great session and great conference so far. Looking forward to tomorrow as well. Maybe two from my side. Number one, I understand the vision as you've been laying it out today, and it makes a lot of sense, of having a collaborative area with AI agents, with humans, having that back and forth iterative loop. What have you seen within your enterprise customers about how they themselves have been adapting to that workflow? Is that something that you've actually been able to maybe even hold their hand and help them navigate that journey? Because it feels like a lot of companies are still trying to figure out exactly what that looks like. I'd love to get your inside perspective there. Praveer, just thinking through the numbers where we're at and investing, there's clearly a huge opportunity ahead, probably TAM expansive. How do you think about the balance right now of wanting to continue to focus on growth opportunities and incremental TAM expansion, while not meaningfully bringing down margins too quickly in the near term? Because obviously it's a tough balance, especially for a public company. Thanks so much. Yeah, I can take the first one. We've been talking to a lot of our customers, and the first question that we always get is: what is the recommended workflow? Actually every company is experimenting with a lot of different ways of working. That experimentation has been good, but the problem that they now face is this incredible fragmentation of prototypes in different places on people's local machines. The conversation usually comes back to how do we actually gain these benefits as a team, not just as individuals. Earlier in the year, we proposed solutions like, through MCP, being able to bring context wherever you created back into the canvas, where it becomes a collaborative space and you can iterate together, and you can then send it back to wherever you started. That was a really interesting solution for a lot of people and continues to be. I hope that with today too, and we've gotten some signal with some early access program customers too, with things like Code Layers, the ability to bring all these different prototypes, whether they were created in Figma or not, in a place where they can compare and contrast really quickly. We hope we'll solve some of the feedback loop challenges, the scattering of artifacts that has happened. I will also admit that all of these workflows are still very nascent. We're newly bringing collaboration to some of these really powerful workflows. We'll continue to talk to customer to see how they use it. At the same time, one of the things that we've started to see as well is a new kind of customer that's coming. For example, we're speaking more to product leaders and engineering leaders, who are interested to solve some of these organizational problems that uniquely AI has brought to us. I think your intuition is exactly right though, Rishi. We are finding that we have to be a little bit more consultative in our sales approach. We're complementing our AEs with individuals that can help, and sit side by side with customers to help them get into the happy states, that they can get the benefit of the end-to-end pieces of the platform. I think as we continue to have a larger platform that requires a little bit more handholding to get set up, the investments that we make there in order to support our customers, we see then the benefit in terms of increased retention, increased expansion opportunity will also then increase credit consumption. It's an area that's a no-brainer for me to invest in. The spirit of your second question on where do we sit on the investment cycle. I think we've been very vocal about it, even starting with Dylan's founder letter, that we believe that the opportunity ahead of us is much, much larger than we even could've imagined even a few years back. While we're holding ourselves accountable to the revenue goals and the op margin targets that we set out for the full year, having some freedom to invest and experiment is exactly what we need in order to make sure that we're not getting in front of ourselves, and trying to either over-monetize or under-monetize the user base on our way there. I've been pushing the org as well as the leaders across the company to be in this constant state of experimentation and learning. Obviously we want to make sure that we keep our word and hold our constraints, if we all believe that the opportunity we're going for is much, much larger even than what we initially anticipated, now's the time to keep our foot on the gas. Hi. Thanks so much for putting this on. Andrew Girard on behalf of Tyler Radke at Citi. I've got two quick questions too. To what extent, to use Dylan's words from the keynote, has AI lowered the floor for Figma Design within the organization or enterprise? What specific personas or people within the organization are coming onto Figma Design that weren't previously targeted? I have a quick one for Praveer afterwards. Great. Well, I can take that question. One of the things that, for example, when we introduced Figma Make into our product suite, that groups that we started engaging even more deeply than before is product managers and also just a bunch of non-designers who are interested in just starting to build some products or bring their ideas to life. While it's true that these personas have often always been on our platform, they've been collaborators, they've been leaving comments, viewing the work. For the first time, they felt empowered to actually contribute or visualize their idea. Too, as we've brought the agent into the canvas in the past, whereas Figma might have felt too intimidating for them, they have the ability to bring to life some concept and maybe that sharpens their feedback, or in some cases, it's allowed them to actually build something that they can ship directly. I think AI has in fact lowered the floor from the perspective of creating more, I guess editors on our platform or people who are actually contributing and creating ideas. Yeah. That's 100% what showed up in financials and if you look at our net dollar retention rate as at the end of Q1, you accelerate to the highest point over the last couple of years, in large part because we have been able to sell wider and deeper within these organizations. Got it. Thank you. Then, just one on as you guys are thinking about maybe the next few years in terms of just the net new ARR contribution, how are you guys just generally framework thinking about the different levers, whether it be contribution from seat expansion or rising ARPUs or consumption into the shape of net new ARR over the next three to five years? Yeah, I think all of them will contribute. It's a little bit of a non-answer there. I think we're both seeing the stat I was mentioning. We're seeing accounts where users over 10,000 continue to add full seats because we now are giving them more reason to pull up. You look at all the product launches that we have today, they provide our end customers with the ability to continue to consume AI credits alongside. We're just now lapping our pricing change that we made last year. We've also taken price along the way. I think if anything, we have way more levers today than we thought that we had even a few years ago. It's kind of fun to have your ability to tell the ops along the way. Perfect. Thank you for having us. I had two quick questions, one on the new products, or I guess maybe more so on the agents being available inside the canvas. How is the AI credit consumption intensity different between Make and the agents? Is it drastically different or not that much? Praveer, on just the new capabilities, again, that were launched, how do you think about that being a potential for you to raise price or give you a reason to raise price in the future versus maybe causing more users to upgrade to a full seat? Yeah, I can take the credit consumption question. What we're seeing is it's comparable. There are some nuances in that in the design agent, we are composing many different models, including some of our own. We have a lot more levers to pull there in terms of speed, performance, cost, efficiency, and we'll continue to experiment and iterate on that to bring a better product experience and also do better for the business. Yeah, I think comparable. At the same time, there are also opportunities for our customers to work faster when working on the different medium types. When working on static designs, oftentimes you can get a result faster with less credits and then use that as context to then produce an interactive result from that. It is changing the nature of how customers consume credits, but I think ultimately leading to more consumption. Yeah, Arjun, to answer the second part of your question there, we've got no plans for a price increase at the moment. I think that there's an opportunity here, again, for us to really deliver ever more value for our customers through to many of the launches that we just rolled out. For the users that are using all of these features, there's a ton of value there for the users that this flips them over to a full seat, that's also fantastic for us. For the additional consumption that we start to overlay, that's again gravy on top. We want to make sure that users feel like they're getting a fair shake, or users and our customers feel like they're getting a fair shake here. That's always been our intention, and that's always been the way that we've thought about pricing and packaging. We want to make sure that we allow for more use and more adoption and don't put pricing and packaging as a constraint to do that. Hey, guys. Thanks for having us. This is Ivan here from Wolfe Research. I wanted to ask on the first product that you released today, I think it was Code Layers, right? Which felt like the most flagship sort of release, at least to me. Can you maybe explain the value prop a little more clearly because it wasn't fully clear to me. Is this essentially bringing Make into the canvas? Is that kind of the idea? What is it supposed to unlock in terms of more usage, perhaps more monetization? Yeah, for sure. Well, let me first try to explain the product better. It is true that one way you can think about it is that you can bring a Make into the canvas, one of the things that we were really thoughtful about was to make sure that you can bring code that was authored anywhere, into the canvas. We have a GitHub integration. We have the ability to upload code from your local machine and wherever you prototyped, albeit these are web-based prototypes only, web-based code only, you can actually bring it into the canvas. We expect a lot of people to subsequently iterate using Make or even just bring in existing Makes into the canvas. The reason it's important is, I think it's important to maybe think about the counterfactual of what's happening today, with code-backed exploration, say. Obviously, more people are able to create code-backed prototypes because of today's technology. What ends up happening is a few things. First of all, maybe people are on their local machines, using their coding agents and their IDE, creating something really amazing, it's very hard to share. You actually have to share by either, we see behaviors like recording a video and sharing it in Slack or Teams, actually deploying it somewhere, actually checking it into some repository so someone else can actually check out that code. I've talked to leaders who've lamented that the new collaboration is Git merge, right? You don't get to actually see these iterations, these fast iterations, because it's going through these formal processes around code. The result of that is you have a lot of people creating amazing prototypes everywhere, but they're shared later in the process, or feedback around it ends up in all these different channels. Flip over to Code Layers. You can actually get everyone on the canvas, and if you don't like the prototype that you see, you can just copy-paste a different version and iterate on that yourself. Comment back, the original creator say, "Hey, how about something like this?" They can riff on your thing too. You can create these behaviors where people are sharing much earlier, or you can view things more side by side, or out of the gate you want to explore four or five different directions. Obviously you can do that with other products. You can create a bunch, 10 different tabs and try different approaches, it just doesn't feel the same in terms of being able to play it by side by side or even do it in a collaborative way. I think the value prop here is really treating code as the medium of exploration, that makes it really easy to collaborate around. That's what we're going for with Code Layers. Hi guys, Samik from JP Morgan. Maybe this might be a naive question, but, when we look at the product launches today, including a lot of focus on agent-led tools, how are you thinking about, is this a bigger driver of upsell with existing customers, or do you bring in more personas from the customer base itself? Just looking at some of these products at first glance, looks like somebody who's already pretty ingrained in the Figma ecosystem sort of up being sort of the usual customer there. Praveer for you, as you think about investing here to drive growth, i s that investing in terms of growing the pro pool of tier customers, or is that investing more at the entry-level tiers to be able to expand that TAM more materially faster? How should we think about when you're investing, is the strategy to invest more for the pro layer or get the entry-level customers in to upgrade over time pro? Yeah, I can take the first part. Maybe I'll start with some of the products like, for example, Figma Weave that we showed today. When you look at their average customer or the kinds of people who are buying, they're very different from the kinds of people who are buying Figma today. You have creative teams, people who are creating these rich assets in media, and using Weave to have much more control over it. Of course, there's some natural overlap because there are brand designers and people who are in marketing that have been in Figma from the past. We think that represents a net new opportunity. As for some of the other things that we showed, even Code Layers is a good example of being able to bring exploration in, making it much more accessible for someone to come in and contribute or just create their own riff. We hope that people who are lightweight users of these kinds of tools are now coming in and being able to be pulled in and try to make their own riffs. When you think about some of the more advanced things that we showed, actually what we're trying to do is also make it more approachable by having it in a more multiplayer space in the first place. For example, Dylan wasn't lying when he was saying that he's actually been in a Motion tool for the first time and riffing on some stuff that he saw. That's all possible because the best way to learn is just to be in a shared canvas with someone else and just being able to see something and play with it. In the past, it just wasn't possible at all because you had to download specialized software and go learn how to use the thing. I think it will invite more people into the creative process. Most importantly, it will also make it possible so if the unit is not just people, but the number of artifacts that are being created, number of assets that are being created, we think that will also explode with the expressiveness that we've offered which also means more use of AI. Yeah, to your question on who are we building for, the reality is we're building a little bit for everyone. Our focus has been on for folks that are ultimately building products that ultimately end up becoming ship product in the real world. Investing around product development teams and in serving around all the different users and personas in that team, those are areas that we continue to spend more time, effort, and energy. The broader creative personas, so brand and creative types that will come to the platform for products like Figma Weave or even Motion, or even now that we've exposed more generative tools capabilities, they now have the ability to assume higher craft. We're also seeing the benefit on the longer tail here. In Q1, one of the things that we shared was our new pro team conversions were up about 150% year-over-year in that quarter when you compare Q1 of this year to Q1 of last year. That's because we had introduced more services, more reasons, more opportunities for folks to be able to consume more credits and purchase on some of our higher price point plans. It's both been an acceleration across the lower tiers as well as high today. Hey, guys. This is Jack McShane from Stifel. Thanks for taking the questions today. I wanted to ask, we saw the slide twice on the explosion of apps being created. A lot of this is probably people building digital experiences for the first time. A lot of these, I would imagine, are weekend passion projects, things like that. A lot of these people are probably unfamiliar with Figma and design tools to begin with. Kind of a two-parter, but when you think about the ICP, how much does the long tail strategically matter? Then is there an investment that has to be done on the self-service side or the marketing front to capture this? I think the value of community is, if you're here at the conference, you're getting the sense of it. It's a really interesting profession in that they want to share their creative work. They want to work with others. They really want to hear each other's feedback on different ideas. I think in terms of how we continue to keep that flywheel moving, we need to make investments to ensure that we're always top of mind and continue to be tip of tongue so that as people are exploring those projects on their weekends, that Figma is the default tool so that they can do it. We've seen this in the past, that the tool that you use on the weekends, on the next weekends, ends up becoming the tool that you use for work. It's also the reason why we're deeply invested on the education side. Community is huge for us. Our ability to listen to our community for feedback and then continue to deliver has gotten us here today. We don't anticipate to de-invest there at all. Anything you guys want to add? I think we are seeing some of those new personas in Figma Make, in particular. I think our brand is strong enough that it goes beyond just traditional designers. Not everyone signing up for Figma Make is a designer by title, I think that opportunity continues to expand as we bring the design agent to our platform and make our products more accessible and easier to get started with as well. Yeah, I think proof of that is the new pro teams that we've been seeing too, where we're seeing more teams that are just using Make, for example, or Make as their first tool before they expand to other products. I think that's indicative of the fact that people, personas that have not contemplated Figma before are showing up and being interested. We're going to take a question from an analyst online, if that's okay. Great. This is from Billy Fitzsimmons. The question is, "Figma has been transparent around the credits different types of actions in Figma Make consume. As we think about the announcements today, whether the agent coding, et cetera, how do we think about the magnitude of some of the expected tokens consumed relative to some of the existing actions in Figma Make? I think, Kris, you touched on this already a little bit. Yeah, I think like prompt for prompt, we expect them to be roughly comparable. There's definitely differences, but it's totally a function of the complexity of the design and the complexity of the prompt as well. Then I think the opportunity here for us is that we're bringing the agent into the primary surface of our product suite, there's additional opportunities that come with that. Thanks. Jackson Ader at KeyBanc Capital Markets. Thanks for doing this. I'm just curious about maybe some of the history lesson that we can get as you guys in the past have expanded personas from. It sounds like you're expanding personas again. What has worked in prior iterations that it's like, "Well, we know we can do that again, this worked, this worked, this worked." What are some of the learnings where it's like, "Eh, that didn't really work. We won't be doing that in the future," whether it's on go-to-market product, those kind of investments. Thanks. I'll start, and then if you guys have ideas. Yeah, I think Dev Mode is the example case that I point to where there were a number of users that were already on platform, that they were being shared Figma files by their designer or PM or whomever counterpart, and then were then being tasked to convert that asset into their front-end code. What we realized is that those developers weren't necessarily greeted with a first-class experience when they landed in the canvas. They were treated with this infinite canvas. They didn't really know what had changed. They didn't really know what was ready for development, and they were hacking around the system to say, "Okay, designers have circled this thing, so they built this thing, it's ready," or, "Don't touch that," that sort of thing. With Dev Mode, what we realized is if we intentionally aimed at improving the ergonomics around their use case, we could then take to them a first-class experience that we believed we could monetize on the other side of it. That's kind of the intuition that we have when we think about some of these other surface areas. If you think about Make, Yuhki was talking about how Make ends up being this tool that PMs can be that much more empowered to use. What we were studying was how a PM was previously communicating with a developer and designer as their hand. They were effectively describing an idea, trying to communicate the intent, and then realizing that, like, "Okay, well, I can only take it so far, and I need to lean on my peers in order to take it the full way." When you then arm them with a product that allows them to go near end to end, it then is a tool that's extremely interesting for them to whip up their hand and pay for. I think studying the why, studying and developing intuition around the user pain there helps then us facilitate where do we take the product and direction. I think the other part of your question was, what are the things that we learned along the way? Do you guys have anything? I can share one. I think that this is more an opportunity. As we build all these different products, just keeping in mind the connectivity of all these products. We've taken an approach, like for example, last year where we launched a bunch of new products, and it was by intention that we wanted to get those out to market, learn, and really find product market fit first. It was definitely the case that some of these products felt a little bit disconnected from each other. With today's announcement, for example, with Code Layers, it's a really great opportunity to blur the line between Figma Make and Figma Design, for example. For people to truly understand how these artifacts can connect or really introduce new ways of collaborating. Maybe now a PM can be in their immersive view that looks like Figma Make and just be focused on their work there, but maybe someone else can be in the canvas comparing and contrasting. We open up a lot of these opportunities when we can blur the lines between these products, and we're really excited for the capacity of Agent to do that as well, to not make those walls felt as much. I think that's we internally talk about interoperability, that's something that we'll continue to invest in to make it easier to build more flexible workflows. Thanks. I wanted to ask a stack question. I think over the last, I don't know, six months, 12 months, there's been a lot of back and forth between the model providers and the competition amongst themselves and who has the better model. Now it seems like cost is starting to come into the equation, especially with open source models also becoming a little bit more advanced. I'm curious if you've changed any part of your model stack and which frontier models you're using and how you think about open source as a potential use in your tech stack. Yeah. We just completed a pretty big rewrite of the core platform that backs all of our agentic features across our product suite so that we are really well positioned to capitalize on the latest and greatest models very easily, so that we're always benefiting from advancements in the models themselves. We also get to deploy, experiment with, and post-train open weight models as well as they continue to improve. We feel like we're really well positioned to capitalize on any new model advancements and also to continue to invest in the post-training of these open weight models to make them best in class for our users. That's been a really core priority for us, really happy to have landed it, and it's one of the reasons why we've been able to launch all these new things around it. With that, are there any further questions? Great. Again, thank you all so much for making it here with us today or joining us online. We really appreciate it. We hope that you're able to join some of the other sessions and get to talk to some of the customers here today as well. We'll be around for a few more minutes too in case there's any further questions. Again, thank you so much, and thank you to the panelists for joining today. Thank you. Thank you, everyone. Thanks, everyone.