Innovation within Design Systems

#Schema2025 just introduced major updates that will change how we build and scale design systems in Figma. The new features are exactly what large teams have been asking for more control, more flexibility, and less overhead. We’re especially excited about what this means for teams trying to scale design systems without slowing down their workflow. → You can now organize variables into collections Perfect for managing themes, brands, or localization without making a mess. → Components now support slots This gives teams flexibility without relying on overrides or hacks. → Figma added a native design linter So consistency is no longer a manual process—it happens by default. → Dev Mode is maturing fast Specs, handoffs, updates—all in one place, right inside Figma. They directly impact how fast teams can move while staying aligned. By combining design tokens, component logic, and Figma Make prototypes with MCP-based development workflows, we’re seeing measurable gains: → Prototypes built faster → Components reused across teams → Less design-developer back and forth → Reduced rework and decision fatigue If your team is developing or expanding a design system and you're interested in how to integrate this into your workflow or product team, feel free to leave a comment or reacht out. We're always happy to share what we've learned.

Service design and futures practice are converging. Here's what I think that means. Something has been quietly shifting in our field. A few years ago, mentioning horizon scanning or scenario planning in a service design context would get polite nods and a quick return to the journey map. Foresight belonged to strategists and policy teams. Service designers improved experiences. The two rarely sat in the same room. That is changing. Design schools are building bridges between both practices. OCAD University, the Royal College of Art, and RMIT have all established design futures programs. Practitioners trained in foresight are showing up inside design and innovation teams. Service designers are quietly picking up foresight methods and asking what they might do with them. This isn't just an academic trend. It's a response to something real. A shifting context that asks for more. The environment that services operate in is changing faster than the tools we use to design them. Climate pressures, demographic shifts, geopolitical instability, and technological change are no longer distant considerations. They are reshaping the conditions under which services function, often faster than organizations can redesign their way out of problems. A service that works brilliantly today can become fragile within a few years when the assumptions underneath it shift. Many of the organizations I've worked with are starting to feel that it's not an abstract risk, but something operational. Reactive redesign. Costly rework. Systems that made sense when they were built, but no longer fit the context they're operating in. The convergence of service design and foresight feels like a field-level response to that problem. It changes what good research looks like. It changes the artifacts we produce. And it changes who we need to collaborate with, bringing foresight practitioners, systems thinkers, and policy specialists into conversations that used to be led by designers alone. None of this means abandoning what service design does well. Improving present experiences still matters enormously. But I think we're entering a period where the most interesting and important design work will sit at the intersection of these two practices helping organizations not just improve what they have, but prepare for what's coming. There is a strong case for Anticipatory Service Design as a practice. Not speculative design but true anticipatory practice to help build resilient services and service organizations. I look forward to sharing more on this over the next few weeks. Happy Monday!  #ServiceDesign #FuturesThinking #StrategicForesight #AnticipatoryDesign #DesignFutures #Foresight #FuturesLiteracy #Futures #ThreeHorizons #Innovation #OrganisationalResilience #BusinessDesign #TransformationDesign

Design system teams are dying 🫠 Over the past few months, I've been asking founders, PMs, and designers a simple question: . "Is your design system serving your product, or is your product serving your design system?" Most people pause. The truth is, this question explains why some companies ship fast and innovate, while others get stuck debating design token names. Many companies have design systems. They even have dedicated teams. So what makes them product-led while traditional enterprises remain product-enabled? When innovative companies need to ship something new, they ask, "What's the best solution for users?"  They experiment with new components and patterns, ship them to learn, and then decide whether to incorporate them into the design system. When slow-moving companies need to ship, they ask, "What can we build with existing components?" Same tool, opposite outcomes. ➡️ The 3 types of companies 🚀 Product-Led (Agile, constant iterations) Lovable, Anthropic, Stripe, Notion, Linear, Replit, Figma, Miro "Ship to learn, iterate to win" "Breaking" the design system for user value is encouraged. ⚙️ Product-Enabled (Top-Down Hierarchy) Salesforce, Adobe, IBM, SAP, Cloudflare, Microsoft, Meta "Let's check with the design system team." Breaking the design system requires committees. 🐌 Product-Supported (Slow-paced environment) Banks, Government, Traditional Retail "We need a plan, a roadmap, ..." ➡️ When AI can ensure consistency in milliseconds, when components generate on demand, when you can create documentation with prompting, the question isn't whether you need a design system, but who's in charge? ✨ Prediction: In 2 years, design system teams as we know them will be gone. The winning teams will merge system expertise directly into product development, creating intelligent systems that maintain what matters (brand, accessibility, quality) while enabling what counts (speed, innovation, differentiation). #designsystem #productdesign #productdevelopment #AI

When I worked in #automotive, after many years the same ideas kept recurring. They reappeared like perennials (2-stroke engines, fuel cells, electrification, new ignition systems, novel engines, novel transmissions). Most of those ideas 'flamed out' after burning huge amounts of R&D and investor capital. After looking at it for a while, the only pattern I could see that could shed insight into success or failure (beyond the human team pushing it) was if the innovation was based on one or more of the following things changing. If these things were present, a concept could become viable enough to displace an existing solution. Once example is the addition of microprocessors and fuel injectors to replace carburetors on engines. Another was the recent success of battery electric vehicles after almost a century as a failed competitor to the gas engine. The four things that drive innovation, enable real progress, and make disruption possible are: 1) New #materials that improve things like strength, heat resistance, conductivity, cost, manufacturability, etc. ➡ For example, carbon fiber allow lighter and more fuel efficient aircraft.[1] 2) New #manufacturing capabilities that improve things like tolerances, production rates, scrap rates, etc. ➡ Additive manufacturing (#AM) is an example of this transforming entire industries. [2] 3) New #simulation and #analysis capability to be able to predict and optimize design performance in a virtual setting that make it possible to iterate a design much, much faster and see cause and effect that is impossible in the real world. Some things are practically impossible to develop without simulation because the improvements and learning from physical prototypes is too expensive and slow. ➡ I doubt an iPhone or most any other electronics could be engineered without simulation. [3] This also applies to many mechanical products like the substitution of plastics for metals in many applications. 4) Finally, new #control, #communication, and embedded #intelligence in products transforms their operation and performance in fundamental and often unseen ways. ➡ Virtual power plants [4] are an example of a service that could not exist without new control, communication, and embedded intelligence. Additionally sometimes the market, regulations, or commercial ecosystem changes, so none of the above are needed to make a system a better fit. Existing capabilities can be recombined to improve a product. Am I missing anything? #deeptech #hardtech #spacetech #design #manufacturing

A strong #informatics team doesn't just follow the playbook - we write new ones. The best innovations often come from deep system knowledge meeting real-world problems. When you truly understand the functionality and features across modules and applications, you start to see connections and possibilities that weren't in the original design. Sometimes the informaticist spots it. Sometimes it's a #clinician or #nurse asking "could we use this for...?" Either way, these moments of creative problem-solving - where we discover use cases beyond what was intended - are where real transformation happens. This year, we're preparing to submit several presentations for Epic XGM that showcase exactly that kind of thinking (hopefully they are accepted!): - #Beacon in #Imaging: #Radiotherapeutics and #Theranostics We used Epic's Beacon module in a completely unintended way - to manage radiotherapeutics workflows. It wasn't recommended by Epic at the time, but it worked so well that they're now recommending it to others. This is a story about creativity and the courage to innovate within constraints. It's also proof that deep cross-module knowledge can unlock solutions hiding in plain sight. - #AI generated patient messaging + Prompting: Driving 46% Sustained Draft Usage Nearly half of all AI-generated drafts seen by our clinicians are being used. We'll share how we achieved and sustained that level of engagement through thoughtful prompting design, change management, and workflow integration. We see systems and opportunity where others see features We engineer creative solutions when the standard path doesn't exist We turn "it wasn't built for that" into "but it could be" #Innovation happens when informatics teams are empowered to experiment and collaborate - building a bridge between today's workflows and tomorrow's possibilities. What innovations has your informatics team brought to life?

𝐓𝐡𝐞 𝐔𝐧𝐝𝐞𝐫𝐚𝐩𝐩𝐫𝐞𝐜𝐢𝐚𝐭𝐞𝐝 𝐀𝐫𝐭 𝐨𝐟 "𝐈𝐧𝐧𝐨𝐯𝐚𝐭𝐢𝐧𝐠 𝐖𝐢𝐭𝐡𝐢𝐧" #3 in the series of essays on engineering culture from Luiz Barroso (supplementary material to “The Data Center as a Computer” book at https://lnkd.in/gKAvmnkF +Urs Hölzle). When a system feels old or limited, the engineering instinct is often to "tear it down and start over." In "Innovate Within," (https://lnkd.in/gwhzDRS6) Luiz Barroso challenges this "disruptive innovation" bias. He points to the story of x86 vs. Itanium. While Itanium was a bold, from-scratch reinvention, it failed. Meanwhile, x86 survived and thrived because teams at Intel and AMD "innovated within"—systematically addressing limitations while allowing a massive existing ecosystem to benefit immediately. Building from scratch is often a tax on developers and users. True innovation often means finding clever ways to evolve your most successful, mature systems without disrupting the value they already provide.

After working with numerous #DesignSystem teams, I feel that one underperformed activity is the actual collaboration with product teams and conducting product inventory workshops. The Design System Inventory Workshop is a practical approach aimed at improving how design system and product teams work together. The workshop involves analyzing a product comprehensively by mapping the sitemap, taking screenshots of its various pages, and then reviewing them together. The main goal is to identify and catalog the patterns and components seen across the product's flows and features. This process helps determine which components can be reused or should be replaced with standardized options from the design system. This method benefits both teams. Product teams learn about the design system's structure and the reasoning behind it, while design system teams get to work closely with product teams. This direct interaction is invaluable for understanding how components are actually used in the product, discovering any variations that were not considered, and identifying new components that could be standardized and added to the design system. Moreover, this workshop is useful regardless of whether a design system is just starting out or already in use and needing wider adoption. It goes beyond simple metric analysis, which only shows how components are currently used. This more in-depth review reveals new patterns and opportunities for the design system, leading to better prioritization and decision-making. Most importantly, it strengthens the working relationship between the teams, laying the groundwork for effective collaboration and shared understanding 👋 In my upcoming series of articles, I'll be sharing a guide on conducting these workshops, complete with practical tips and examples. Your feedback is important—drop a comment if you'd like access to the Design System Inventory Workshop materials

What if AI coding tools could speak the same language as your design system? That’s what Waqar Ali and the team at Typeform set out to explore. In their latest post, they shared how they built a Design System MCP server to bridge the gap between Figma and code, making their Echo Design System machine-consumable by tools like Claude Code, Cursor, and Cline. Here’s what stood out to me: • They used their Storybook docs as the foundation for an intelligent system interface • The MCP server helps AI tools identify component boundaries, usage, and context • It's not about generating code from scratch — it’s about giving AI the right constraints • They’re aligning design tokens, props, and behaviors to work across humans and LLMs • And they’re doing it all while keeping their system stable, documented, and scalable. But it also raises some deeper questions: 🔹 Is your current documentation format AI-friendly by design? 🔹 How far can we push the idea of “design-to-code” when LLMs get involved? 🔹 What kind of governance or structure will this require from design systems teams? You can read the full story here: https://lnkd.in/dEtMAhwC Kudos to Waqar Ali for driving this and breaking it down so clearly. Is anyone else exploring something similar with AI and design systems? Drop your thoughts or experiments below 👇 This is a space that’s evolving fast, and we can all learn from each other. #AI #DesignSystems #designsystem #DesignToCode #LLM #Figma #Storybook #Typeform #DesignOps #UXEngineering #uxdesign #uidesign #uxui #uiux

“Where am I?” As our design system matures, things move around. We’re trying to invest in ways that allow feature designers and engineers to locate where they are within a design system, and where to find what they need. → Multiple design libraries can be hard to track and become confusing when designers and engineers aren’t aware of what does and doesn’t exist in each library. → Every component and pattern has a varying degree of specificity from generic to a specific business context or content pattern and deciding what goes where is confusing. → Theming in Figma is great, but... the current nature of variables can make managing multiple modes across multiple brands challenging. Here’s how we’ve been trying to get our head around reducing complexity for our team and consumers 🏗 Figma file architecture - Having a clear plan for structuring your design system libraries is critical for folks to find stuff. Our early attempts at a highly modular system with numerous smaller libraries were too unwieldy. We’ve landed into a Tokens, Components, Patterns, and Features structure that streamlines the process for everyone. It was super cool to stumble on these Figma docs that deep dive into various strategies for organizing your files https://lnkd.in/dGyN-SQw 🚦 Wayfinding elements - Once designers and engineers are in the files themselves, it's still easy to get lost. We’ve implemented consistent color-coded “file covers” distinguishing design system files from feature files, “getting started” pages outlining file dependencies, and consistent page naming conventions indicating iterations from elements that are ready for development. Thanks to Vitaly Friedman for the reference to Saurav Rastogi's post on Figma file organization which is packed with insights. https://lnkd.in/giBs_Jgn 🎨 Multi-brand variables - Our team has worked hard to figure out the most intuitive way to allow for white-labeling our design system for custom theming. We currently leverage modes at multiple layers of the design system based on their specificity from global to semantic to component to pattern. Romina Kavcic has endless resources for teams who are looking to build out their token strategy. https://lnkd.in/deiMGCe9 💎 Wayfinding within design systems is hard but a couple of things that help folks understand where they are and what they can do → Partner with engineers on an architecture that maps somewhat to their mental model making things more intuitive → Setup wayfinding documentation within your Figma files that orient users to where they are within the system → Map your tokens and variables across your system so that it is clear how to layer modes to achieve custom themes #designsystems #figma #uidesign #tokens #components #patterns

Real-Time UI: https://lnkd.in/dNbmGcX8 "A prototype is worth 1000 meetings." But what if the meeting _is_ the prototype? That’s the spirit of an idea I’m calling “Real-time UI” (the name of which I gave next-to-no thought, so forgive me). The tools and technologies now exist to generate UI in realtime, making it possible to convert a conversation into a working digital thing. In this video, I introduce the concept to TJ Pitre and Ian Frost , and we talk about the possibilities and ramifications of generating UI in realtime, as well as speaking to the infinite creative potential of using AI & design systems together, as we are covering in our course: https://lnkd.in/eG5h8uaP https://lnkd.in/dubfHuCn As I see it, real-time UI can help accomplish a number of things: ◉ Visualize UI components in real-time – surfacing design system components immediately as they’re referenced in conversation (design systems are a shared language!) ◉ Visualize product design in real-time. Make abstract ideas real as soon as the words exit your mouth, and use the working prototype as a wet ball of clay the team can sculpt together over the course of a conversation. ◉ Wield your design system’s infrastructure to make realistic things. The spirit is to have the conversation and infrastructure tuned to your specific team’s context. Create prototypes that are built using your organization’s best practices rather than whatever AI decides to randomly generate. ◉ Minimize the friction involved in making prototypes ◉ A visual accompaniment to a conversation can help teams unlock new ideas, expose weak spots, explore opportunities, and iterate collaboratively ◉ Open the door to a more participatory design process. Diversity is critical to success, and it’s so important to make sure that digital products represent the best thinking from different disciplines & perspectives at a company. Historically, the design process was prohibitive to people who weren’t skilled in the mechanical aspects of creating designs & code. This is no longer the case. Of course professional designers or developers are still necessary (now more than ever!) to produce great results, but there’s now an opportunity to create more democratic, collaborative, participatory design workflows. If you're interested in exploring the future of using design systems and AI together, we'd love it if you joined our community by preordering our AI & Design Systems course! #ai #designsystems #ux #uxdesign #frontend #prototyping #design #process #workflow #collaboration