Alexandre Benitah

I have reviewed hundreds of UX case studies in my career. Most of them suffer from: - Lack of polish - Poor scannability - No engaging narrative - Failing to show the impact - Focusing too much on the process The solution is quite simple: 1. Create a story 2. Lead with the impact 3. Share your process at a high level 4. Keep it short and polished; optimize it for scanning 5. Share it in a format that is easily accessible and shareable The following cheatsheet will help you structure your case studies in a concise, engaging, and impactful manner. If you found this useful, consider reposting to your network ♻️ -- Bonus tip: You can't always measure the impact of an initiative, and that's totally ok. Here is what you can do instead: - Show the metrics you expected to move, even if you don't have the numbers - Did it have a strategic impact for the team/business? - Did you test the new flow/feature? Show the results #productdesign #uxdesign #uiux

929

40 commentaren

Dublin UX MEETUP TAKEAWAYS. It took place in the Workday office, which hosted around 200 people for the event. Interestingly, the same Workday office hosted a couple of other events at the same time. Some offices are quiet and empty, while others are full of energy, driven by a hunger for learning, innovation, and networking. 1️⃣ Humans must ultimately take ownership of the final output. AI can support, suggest, and accelerate, but the responsibility stays with us. Human judgment is the final step, not an optional layer. 2️⃣ To build trust, we must show the data’s path, not just the AI’s conclusion. Transparency is more than explainability. Let people see how the data flowed, what the model considered, and why the outcome makes sense. 3️⃣ AI challenges our established sense of self in the workplace. As AI takes on creative and strategic tasks, we’re all redefining what it means to contribute. That shift affects motivation and collaboration. 4️⃣ Use JTBD, mental model, and user type for research. If your research doesn’t reflect how people think and what they’re trying to achieve, your features will never feel intuitive. 5️⃣ If it causes friction, it’s UX That means you can improve it, so start looking for UX everywhere. 6️⃣ With AI amplifying data overload, we must architect clarity, not just deliver information. AI brings more signals than ever. Our job now is to architect clarity, not just push more information. 7️⃣ Reducing cognitive load is always important. If people have to stop and think, something’s wrong. Good design should feel obvious. 8️⃣ Inclusive design rests on accessibility, adaptability, and affinity. Can people access it? Can they shape it to their needs? Does it feel like it belongs to them? 9️⃣ Conversational interfaces lower barriers by making tech feel natural and easy to use. Natural language and human tone reduce intimidation. This matters for inclusion, especially in high stakes or unfamiliar contexts. 🔄 UX is a mindset, not a role. It’s how we see problems and how we choose to solve them. The best UX work often comes from unexpected places. Thanks to speakers Grace Hughes, John Conneely, Silvia Amaduzzi, Rachel Bradstreet. Thanks to Patrick M. and many more for the great job. #dublinux #workday #uxdesign #productdesign

43

5 commentaren

The "whiteboard" step in the hiring process for new designers @ Customer.io has been my pet project over the past year because, frankly, I hate traditional whiteboard sessions—both as a candidate and a facilitator. Here's why: In my opinion, what separates a great designer from a good one isn’t just their ability to generate solutions. It’s their ability to bridge context gaps with stakeholders and build alignment. Designers hold the key context needed to align people, but traditional whiteboard sessions fail to replicate this dynamic. Instead, they flip the power structure. The facilitator knows more than the candidate, turning the exercise into a quiz rather than a collaborative session. And let’s be honest—thinking on your feet with minimal context doesn’t showcase real design skills. It just reveals how good you are at interviewing. So, we reimagined the whiteboard session with a fresh goal: flipping the power dynamic to give candidates equal (or more) control, creating an environment that mimics a real-world co-design session. Here’s how it works now: 1. The candidate picks the product. We ask them to choose any B2B SaaS product with a self-serve signup flow they’re unfamiliar with. We provide suggestions, but the final choice is theirs—this helps reduce bias. 2. We co-design as teammates. During the session, we frame it as if the candidate has invited another IC designer (the facilitator) to collaborate. There are no constraints—the facilitator engages just like they would in a typical co-design session: asking questions, brainstorming, pushing back, and so on. It’s about two teammates working toward a shared outcome. 3. The outcome: journey over destination. In 60 minutes, the two designers aim to define the business’s goal, uncover problems impacting that goal, and brainstorm ways to address one of those problems. The focus isn’t on solving everything perfectly—it’s on seeing how they approach uncovering and framing problems that drive business outcomes and collaborate with others. My favorite part? Each session feels fresh and exciting for me as the facilitator. I come in with zero context or expectations and get to see where the conversation leads. Plus, for more introverted candidates, I can step into a driver role to ensure we stay on track while observing their approach. Is this process perfect? Definitely not—no process is. But we’re clearly learning in the right direction: one that moves away from high-pressure, artificial exercises and toward scenarios that reflect how designers actually work.

68

5 commentaren

These days I keep thinking.. should we use AI to build our skills? Or should we build skills to work with AI? There’s a big difference. Now anyone can type a few words and get a nice looking result. Designs, writing, ideas, all look good enough. But that’s the real problem. Good enough stops us from growing. We see someone post an AI output and go.. Wow, this looks great! Then we copy the same prompt. Soon, everyone’s work starts to look the same. That’s not skill. That’s just automation. AI made things faster, not better. Real skill starts after AI gives you something. When you: – Check it – Fix it – Add your ideas – Make it better That’s when you learn. AI doesn’t teach you design or writing. It only gives you a start. You grow when you think, change, and improve that output yourself. So yes, use AI. But don’t stop there. Use it to test your thinking. To improve your taste. To see what good really looks like. AI won’t replace skilled people. It’ll replace people who don’t know how to use it well. The future belongs to those who can:  - Work with AI  - Think beyond prompts  - Make outputs truly their own

45