Design Leadership is a Contact Sport

I’ve never believed design leadership is about standing at the front of the room with all the answers. If anything, the more senior I’ve become, the more I’ve realized my job is to create the conditions for other designers to do their best work. That means leading by example. Protecting quality. Staying close enough to the craft to know when something isn’t working. Coaching people through the messy parts of the job. And, perhaps most importantly, helping designers become more capable and confident than they were when they joined the team. Because great design teams don’t happen by accident. They’re built.
Lead by example, not by exception Leadership doesn’t mean you stop being a designer. I don’t believe in asking a team to do things I wouldn’t do myself. If we’re going to sweat the details, I need to sweat the details. If we’re going to push for better customer experiences, I need to be willing to challenge the easy answer. And if we’re going to work through ambiguity, I need to be comfortable sitting in it too. That doesn’t mean I’m the person doing the work for everyone. Quite the opposite. It means showing what good looks like. Sometimes that’s getting into Figma and working through an interaction. Sometimes it’s asking the uncomfortable question in a product review. Sometimes it’s helping a designer sharpen a story before they walk into an executive meeting. The goal isn’t to be the smartest designer in the room. The goal is to make the room better. Quality and craft still matter As organizations grow, it’s easy for design to become increasingly focused on process, roadmaps, frameworks, and operating models. Those things matter. I’ve spent plenty of time building them. But none of them matter if the actual experience isn’t good.

Design leadership has to protect craft

That means creating space for designers to explore multiple directions instead of immediately converging on the first reasonable solution. It means giving useful critique instead of vague feedback like “make it more intuitive.” It means asking whether the interaction actually makes sense, whether the hierarchy is doing its job, whether the system is coherent, and whether we’re solving the right problem in the first place. And yes, sometimes it means saying, “This isn’t quite there yet.” That’s not about perfectionism. It’s about having a team that cares enough about the work to keep making it better.

Coaching is not fixing

One of the biggest shifts in my leadership approach has been moving from solving problems for designers to helping designers solve problems themselves. It’s tempting, especially when you’re experienced, to jump in with the answer. I’ve learned that’s usually the easy way out. Instead, I try to understand what’s underneath the problem. Is the designer struggling with the interaction? The strategy? The stakeholder relationship? The confidence to push back? The ability to tell the story? The solution might be a design critique. It might be a new framework. It might be role-playing an executive conversation. Or it might simply be asking a few questions that help someone see the problem differently. The best coaching moments are the ones where the designer eventually says, “Oh. I see it now.” That’s much more valuable than me telling them what to do.

Build designers, not just deliverables

I also believe leadership has a responsibility to create intentional opportunities for growth. Training shouldn’t be something we dust off once a year during performance reviews. Growth happens through the work. Give someone a bigger problem than they’ve solved before. Let them lead the critique. Put them in the room with executives. Teach them how to navigate a difficult product partner. Help them understand the business, not just the pixels. Give them exposure to research, strategy, systems thinking, storytelling, facilitation, and all the other skills that turn a good designer into a strong design leader. And when someone is ready for more responsibility, give them room to stretch. They might stumble. That’s okay. If we only give people opportunities when we’re certain they’ll succeed, we’re not really developing them. We’re just giving them more of what they already know how to do.

Make the team better than you found it

For me, that’s ultimately the measure of design leadership. Not how many projects I personally shipped. Not how many decisions I made. Not whether my name is attached to the work. It’s whether the designers around me are doing work they’re proud of, whether they’re making better decisions independently, whether they’re challenging ideas instead of waiting for permission, and whether they’re becoming leaders themselves. The best leaders eventually make themselves less necessary. That’s not stepping away from the work. It’s scaling your impact through other people. And in a world where design is becoming increasingly complex—AI, platforms, systems, ecosystems, organizational change—the ability to build strong designers and strong design teams may be one of the most important pieces of the job. Because the real output of a design leader isn’t just great design. It’s a team that knows how to make great design happen.

Enterprise UX Should Be Better

We’ve somehow decided that enterprise software gets a free pass on being difficult. If it has a lot of data, it can be dense. If it has a lot of customers, it can be complicated. If it’s used by highly trained professionals, apparently we can throw a few hundred settings, seventeen tabs, and a cryptic error message at them and call it “powerful.” Meanwhile, those same people go home and use consumer products that somehow manage to make complex things feel obvious. That feels like a pretty low bar for enterprise UX. Enterprise users are people too. Shocking, I know. They use our products all day. Sometimes all night. They’re trying to get work done, make decisions, solve problems, and get home at a reasonable hour. They shouldn’t need a PhD in our information architecture to figure out where the thing they need is hiding. Enterprise UX should be just as easy, intuitive, and delightful as consumer UX. Maybe even more so.

The user doesn’t care that it’s enterprise

One of the biggest mistakes we make in enterprise design is treating complexity as an inevitable part of the category. It isn’t. The business may be complex. The workflows may be complex. The data model may be complex. The permissions model may be complex enough to make everyone question their life choices. The experience doesn’t have to expose all of that complexity to the user. That’s our job. Good UX absorbs complexity. It organizes it, prioritizes it, explains it, and gets it out of the user’s way whenever possible. The best enterprise experiences can make incredibly complicated systems feel surprisingly simple. Not because the underlying system is simple, but because someone did the hard work of making the complexity understandable. That is design craft. Your enterprise users are power users—and they still deserve joy There’s a misconception that enterprise users don’t need delightful experiences because they’re “just trying to get their work done.” Exactly. They’re trying to get their work done. That’s why the experience matters so much. If someone spends eight hours a day in your product, every little bit of friction compounds. One extra click doesn’t sound like much. Neither does one confusing label, one unnecessary modal, or one workflow that requires users to remember something the system should remember for them. Multiply that by hundreds of interactions a day, thousands of users, and years of usage—and suddenly “just one extra click” is a pretty expensive design decision.
Enterprise UX has an incredible opportunity to improve people’s daily lives at scale. We should take that seriously. Consumer UX raised the bar. Enterprise needs to catch up. People have become accustomed to products that anticipate what they need, provide useful feedback, remember their preferences, and make complicated tasks feel manageable. They swipe. They search. They personalize. They get recommendations. They undo things. They recover from mistakes without having to call someone named Steve in IT. Then they log into work and suddenly it’s 2007. That contrast is a problem. The expectations people bring from consumer products don’t disappear when they sit down at their work computer. If anything, they become more important because enterprise software is often where people spend the majority of their waking work hours. The bar should be the same: clear, fast, intuitive, accessible, and—yes—maybe even enjoyable. Delight isn’t decoration When I talk about delight in enterprise UX, I don’t mean adding confetti to a dashboard. Please don’t. Delight is often much quieter than that. It’s an interaction that behaves exactly the way you expected. A workflow that saves you ten minutes. A system that anticipates your next step. A message that actually tells you what went wrong—and what to do about it. It’s the feeling of, “Oh, that was easy.” That might be one of the most underrated outcomes in enterprise design.

Design for the person, not the org chart

Enterprise products often have a long list of stakeholders: administrators, executives, security teams, IT, procurement, compliance, developers, product managers, and approximately 47 people who need to approve something before a button can change color. All of those perspectives matter. But they aren’t the user experience. The person actually using the product still needs to be at the center. That means understanding their goals, context, expertise, pressures, and workflows—not just the requirements of the organization buying the software. The customer who signs the contract and the person who uses the product every day may have very different definitions of success. Great enterprise UX has to serve both.
Complexity is not an excuse I’ve spent much of my career working on complex platforms and enterprise products, and I’ve learned that complexity is often where the most interesting design problems live. You can’t always remove the complexity. But you can decide how much of it the user has to carry. That’s the opportunity. We should be designing enterprise products with the same care we bring to the consumer experiences we use every day. Not because enterprise software needs to become consumer software, but because enterprise users deserve the same level of thoughtfulness. They deserve products that respect their time. Products that help them feel capable instead of confused. Products that make hard work feel a little less hard. And maybe, occasionally, products that make them smile. Because if someone is going to spend eight hours a day in your product, the least we can do is make the experience worth showing up for.

UX in Agentic Enterprise

We’re entering the age of the agentic enterprise, which sounds very futuristic until you realize we’re still asking people to navigate the same seventeen tabs they’ve been navigating for the last decade. We can do better. AI and agents shouldn’t add another layer of complexity to enterprise software. They should remove it. The opportunity is simple: make complex enterprise products feel dramatically simpler than the systems underneath them actually are.

Less work, not more

Enterprise software has accumulated a lot of baggage: complex workflows, legacy architecture, data silos, specialized terminology, endless configuration, and enough navigation to make Google Maps jealous. Agentic technology gives UX another lever. Instead of asking users to learn the system, we can let them describe what they’re trying to accomplish and have the system figure out how to help. From “How do I use this product?” to “What are you trying to get done?”

Make the product quieter

There’s a temptation with AI to make the technology visible. Put an AI button here. Add an assistant there. Give it a glowing icon. Maybe add a little sparkle because apparently everything is AI now. But visibility isn’t value. The best agentic experiences may actually make the product feel quieter: understanding context, handling repetitive work, surfacing the right information, and asking for help only when it genuinely needs it. Traditional enterprise UX often says: Here are all the things the system can do. Figure out which one you need. Agentic UX can say: Tell me what you’re trying to accomplish. I’ll help figure out how to get there. That’s a pretty significant upgrade.

Give legacy products a second life

One of the biggest opportunities for enterprise AI isn’t creating entirely new products. It’s making old ones better. Most enterprises have decades of software, data, workflows, and institutional knowledge embedded in their systems. Replacing all of that isn’t realistic—and it may not be necessary. An agent can become a new interaction layer over that complexity—connecting systems, translating workflows into natural language, surfacing buried insights, and helping users accomplish tasks without understanding every underlying system. That means legacy technology doesn’t necessarily have to be the experience. It can become the infrastructure behind the experience. And that’s a huge distinction.

UX becomes the orchestrator

In an agentic enterprise, UX isn’t just designing a screen. It’s designing how work gets done across an ecosystem of intelligence—people, agents, data, workflows, permissions, and systems that were never designed to cooperate. Designers now need to think about: What should the agent do automatically? Where should the human stay in control? When should the system ask for confirmation? How does the user understand what happened? How do we make agent decisions explainable? What happens when the agent gets it wrong? How does a user recover, undo, or intervene? How do multiple agents work together without creating chaos? The interface is only one part of the experience. The real product is the relationship between the human and the system.

Keep humans in control

Just because an agent can do something doesn’t mean it should do it without you. Enterprise work involves real consequences, so users need to understand what the system is doing, what it plans to do, and where they can intervene. The goal isn’t to remove humans. It’s to remove the work humans never needed to be doing in the first place. The real goal: easy Good agentic UX isn’t about maximizing AI interaction. It’s about minimizing unnecessary human effort. We can move from designing interfaces that expose functionality to designing experiences that understand intent. From navigating systems to accomplishing goals. From teaching people how software works to making software work more like people expect. And perhaps most importantly, we can take complicated, deeply entrenched enterprise products and give them a radically better experience without throwing away everything underneath. Less complexity. Less busywork. Better decisions. More time for work that actually requires humans. The most impressive thing about an agentic enterprise product won’t be that it feels futuristic. It will be that it finally feels easy.