User Experience

Empathy Isn’t Missing — It’s Misframed I’ve watched this video countless times. Every time, I don’t see generosity. I see design. I used to believe people ignore the truth because they don’t care. Now I realize it’s because they don’t see what I see. Empathy isn’t a lack of compassion — it’s a lack of perspective. And perspective can be designed. The words didn’t change the man’s story — they changed our frame of perception. When language shifts from description to contrast, it activates awareness. That’s the mechanism behind empathy — it’s not emotional contagion, it’s cognitive reframing. → We respond to difference, not repetition. → We act when a message bridges our world with someone else’s. → We feel when language turns distance into proximity. Here’s how I try to apply that lesson in my own work: ✅ Reveal contrast, not condition. Don’t describe pain — expose the gap between what is and what could be. ✅ Design for awareness before emotion. Help people notice first; feeling follows naturally. ✅ Make others participants, not observers. Use framing that transfers perspective, not pity. ✅ Use silence strategically. Leave room for the reader to complete the meaning. Because empathy doesn’t start with emotion — it starts with architecture. The right words don’t tell people what to feel. They help them feel what was already true. 💭 The Question 👉 When you communicate — are you trying to make people care, or helping them notice what they’ve been blind to all along? #LeadershipDesign #FramingEffect #CommunicationStrategy #CognitiveEmpathy #BehavioralPsychology #PerceptionDesign Video credits: Dr. Marcell Vollmer

Back in 2007, Nobel Prize-winning psychologist Daniel Kahneman taught a private master class to tech founders including Larry Page and Jeff Bezos. The following year, Elon Musk joined. Among the topics: priming, where subtle cues shape our decisions without us realizing it. In that room, Musk pressed on subliminal versus explicit persuasion: “Does the hidden beat the obvious?” Kahneman's answer: "There are many situations in which subliminal effects are stronger than superliminal effects." Translation: Hidden influences shape behavior more than obvious ones. You can't resist what you don't notice. Later after that session, Bezos connected the dots: “You can choose your choice architect.” You either design the decision environment, or it designs you. Amazon designed theirs. One-click purchasing removes the pause where doubt lives. Every additional step is an exit ramp. They chose zero exits. Google designed theirs. That empty white homepage isn't minimal by accident. No portals, no distractions. Just one thought: search. Most companies let chaos choose. Cluttered onboarding. Buried CTAs. Friction everywhere. They're not architects. They're accidents. So how do you become the architect instead of the accident? 1. Choose your pricing architect: Sell your core product for $99/month. Then offer a bundle with two add-ons for $119. The bundle makes the core feel essential. 2. Choose your onboarding architect: When users first sign up, make their first action create immediate value - a report generated, first customer added, dashboard live. Success in 30 seconds primes confidence in everything that follows. In contrast, when you make the frame obvious, you lose it. Slap "Most Popular!" on everything and watch trust erode. The moment users detect manipulation, they create their own frame - one where you're untrustworthy. Kahneman warned Musk about this directly. Covert cues work precisely because they're not noticed. Priming is architecture, not decoration. By the time logic kicks in, the frame has already decided. Because you’re already an architect. The only question is whether you know what you're building.

Invisible UX is coming 🔥 And it’s going to change how we design products, forever. For decades, UX design has been about guiding users through an experience. We’ve done that with visible interfaces: Menus. Buttons. Cards. Sliders. We’ve obsessed over layouts, states, and transitions. But with AI, a new kind of interface is emerging: One that’s invisible. One that’s driven by intent, not interaction. Think about it: You used to: → Open Spotify → Scroll through genres → Click into “Focus” → Pick a playlist Now you just say: “Play deep focus music.” No menus. No tapping. No UI. Just intent → output. You used to: → Search on Airbnb → Pick dates, guests, filters → Scroll through 50+ listings Now we’re entering a world where you guide with words: “Find me a cabin near Oslo with a sauna, available next weekend.” So the best UX becomes barely visible. Why does this matter? Because traditional UX gives users options. AI-native UX gives users outcomes. Old UX: “Here are 12 ways to get what you want.” New UX: “Just tell me what you want & we’ll handle the rest.” And this goes way beyond voice or chat. It’s about reducing friction. Designing systems that understand intent. Respond instantly. And get out of the way. The UI isn’t disappearing. It’s mainly dissolving into the background. So what should designers do? Rethink your role. Going forward you’ll not just lay out screens. You’ll design interactions without interfaces. That means: → Understanding how people express goals → Guiding model behavior through prompt architecture → Creating invisible guardrails for trust, speed, and clarity You are basically designing for understanding. The future of UX won’t be seen. It will be felt. Welcome to the age of invisible UX. Ready for it?

🌎 Designing Cross-Cultural And Multi-Lingual UX. Guidelines on how to stress test our designs, how to define a localization strategy and how to deal with currencies, dates, word order, pluralization, colors and gender pronouns. ⦿ Translation: “We adapt our message to resonate in other markets”. ⦿ Localization: “We adapt user experience to local expectations”. ⦿ Internationalization: “We adapt our codebase to work in other markets”. ✅ English-language users make up about 26% of users. ✅ Top written languages: Chinese, Spanish, Arabic, Portuguese. ✅ Most users prefer content in their native language(s). ✅ French texts are on average 20% longer than English ones. ✅ Japanese texts are on average 30–60% shorter. 🚫 Flags aren’t languages: avoid them for language selection. 🚫 Language direction ≠ design direction (“F” vs. Zig-Zag pattern). 🚫 Not everybody has first/middle names: “Full name” is better. ✅ Always reserve at least 30% room for longer translations. ✅ Stress test your UI for translation with pseudolocalization. ✅ Plan for line wrap, truncation, very short and very long labels. ✅ Adjust numbers, dates, times, formats, units, addresses. ✅ Adjust currency, spelling, input masks, placeholders. ✅ Always conduct UX research with local users. When localizing an interface, we need to work beyond translation. We need to be respectful of cultural differences. E.g. in Arabic we would often need to increase the spacing between lines. For Chinese market, we need to increase the density of information. German sites require a vast amount of detail to communicate that a topic is well-thought-out. Stress test your design. Avoid assumptions. Work with local content designers. Spend time in the country to better understand the market. Have local help on the ground. And test repeatedly with local users as an ongoing part of the design process. You’ll be surprised by some findings, but you’ll also learn to adapt and scale to be effective — whatever market is going to come up next. Useful resources: UX Design Across Different Cultures, by Jenny Shen https://lnkd.in/eNiyVqiH UX Localization Handbook, by Phrase https://lnkd.in/eKN7usSA A Complete Guide To UX Localization, by Michal Kessel Shitrit 🎗️ https://lnkd.in/eaQJt-bU Designing Multi-Lingual UX, by yours truly https://lnkd.in/eR3GnwXQ Flags Are Not Languages, by James Offer https://lnkd.in/eaySNFGa IBM Globalization Checklists https://lnkd.in/ewNzysqv Books: ⦿ Cross-Cultural Design (https://lnkd.in/e8KswErf) by Senongo Akpem ⦿ The Culture Map (https://lnkd.in/edfyMqhN) by Erin Meyer ⦿ UX Writing & Microcopy (https://lnkd.in/e_ZFu374) by Kinneret Yifrah

Data is everything in product design. Without data, we open ourselves up to: - Biases - Opinions - Confusion - Misalignment When we are data-informed and that data is accurate, we can truly make educated product decisions. I like to think of data in two layers: a) What’s happening and b) Why it’s happening. Let’s break it down. What’s happening: - Business data tells us how the business is doing - Marketing/sales data tells us where our customers come from - Retention data tells us when and why customers are leaving us - Engagement data tells us how customers are using our product Why it’s happening: - User research gives us rich insight into why something is happening - Voice of the customer data shows us how customers talk about our product - Usability scores show us how people perceive our product or feature experience in a measurable way - Product market fit & satisfaction scores give us a simple and actionable metric to track and improve over time In terms of accessing that data, methodologies vary, but generally speaking, I always advise the following: 1. Get access to growth and retention data through business dashboards. 2. Get access to product data through your product analytics tool. 3. Set up a cadence to gather customer reviews & comments, either manually or via automated tools. 4. Set up a cadence to speak to your users continuously to answer the why. 5. Set up a recurring survey to track satisfaction and usability. If you don’t have the data structure for any of the above, speak to your product and data team to see if you can change that. If not, rely on the data that you can actually get. PS: The list of metrics is indicative: Actual metrics will differ greatly from one company to another and largely depend on the industry, niche, as well as your data infrastructure and setup. — If you found this useful, consider reposting ♻️ How are you collecting and using data in your design process? What else are you tracking?

New! We’ve published a new set of automated evaluations and benchmarks for RAG - a critical component of Gen AI used by most successful customers today. Sweet. Retrieval-Augmented Generation lets you take general-purpose foundation models - like those from Anthropic, Meta, and Mistral - and “ground” their responses in specific target areas or domains using information which the models haven’t seen before (maybe confidential, private info, new or real-time data, etc). This lets gen AI apps generate responses which are targeted to that domain with better accuracy, context, reasoning, and depth of knowledge than the model provides off the shelf. In this new paper, we describe a way to evaluate task-specific RAG approaches such that they can be benchmarked and compared against real-world uses, automatically. It’s an entirely novel approach, and one we think will help customers tune and improve their AI apps much more quickly, and efficiently. Driving up accuracy, while driving down the time it takes to build a reliable, coherent system. 🔎 The evaluation is tailored to a particular knowledge domain or subject area. For example, the paper describes tasks related to DevOps troubleshooting, scientific research (ArXiv abstracts), technical Q&A (StackExchange), and financial reporting (SEC filings). 📝 Each task is defined by a specific corpus of documents relevant to that domain. The evaluation questions are generated from and grounded in this corpus. 📊 The evaluation assesses the RAG system's ability to perform specific functions within that domain, such as answering questions, solving problems, or providing relevant information based on the given corpus. 🌎 The tasks are designed to mirror real-world scenarios and questions that might be encountered when using a RAG system in practical applications within that domain. 🔬 Unlike general language model benchmarks, these task-specific evaluations focus on the RAG system's performance in retrieving and applying information from the given corpus to answer domain-specific questions. ✍️ The approach allows for creating evaluations for any task that can be defined by a corpus of relevant documents, making it adaptable to a wide range of specific use cases and industries. Really interesting work from the Amazon science team, and a new totem of evaluation for customers choosing and tuning their RAG systems. Very cool. Paper linked below.

Hi everyone! Today I want to share a real, hands-on experience that shows just how transformative vibe coding can be for project managers and PMOs. 💯 I recently discovered a no-code AI platform called Lovable — and no, I’m not an investor, just genuinely impressed. With zero coding skills, I used it to build a fully functional PMO dashboard for my organization in just minutes. All I did was describe what I needed: a colorful, modern dashboard showing key project portfolio performance indicators, plus an admin area. I typed a simple prompt… and Lovable did the rest — designing, coding, deploying — all while I sat back and watched. I could ask it to fix errors, tweak the layout, or add new features using natural language. When a link returned a 404 error, I just told the AI to fix it — and it did, in seconds. What usually takes weeks of planning, coding, and iteration… was done in minutes. This is vibe coding in action — a fundamental shift in how we build digital tools, and a powerful glimpse of how AI is transforming project delivery, decision-making, and speed-to-value. For project professionals like us, this isn’t just about coding. It’s about unlocking new ways to prototype ideas, iterate fast, and bring visions to life — all without technical barriers. This is just the beginning. Let me know what you think — and if you’ve tried vibe coding yourself! 👀 Ricardo #AI #ProjectManagement #Innovation #VibeCoding #DigitalTransformation #FutureOfWork #NoCode #AIForPMOs #AIInAction #Leadership #EmergingTech #TechForGood #PMO #Agile #ArtificialIntelligence #5MinPodcast #AIAgents

Building software today doesn’t look the same as 2 years ago ! Some teams write every line by hand. Some build alongside AI. Others ship products without touching code at all. What changed isn’t technology - it’s how fast ideas move from thought to product. This visual breaks down the three modern ways of building 👇 𝗖𝗼𝗱𝗶𝗻𝗴 (𝗧𝗿𝗮𝗱𝗶𝘁𝗶𝗼𝗻𝗮𝗹 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁) This is full-control engineering. You design architectures, write logic, manage infrastructure, and integrate complex systems. It’s best when you need performance, deep customization, scalable backends, and production-grade applications - but it demands strong technical skills and longer build cycles. 𝗩𝗶𝗯𝗲-𝗖𝗼𝗱𝗶𝗻𝗴 (𝗔𝗜-𝗔𝘀𝘀𝗶𝘀𝘁𝗲𝗱 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁) Here, developers work with AI copilots to move faster. You still write code, but tools help generate snippets, suggest fixes, speed up debugging, and accelerate prototyping. It’s ideal for rapid iteration and smarter development workflows while keeping technical control. 𝗡𝗼-𝗖𝗼𝗱𝗶𝗻𝗴 (𝗩𝗶𝘀𝘂𝗮𝗹 & 𝗔𝘂𝘁𝗼𝗺𝗮𝘁𝗶𝗼𝗻 𝗣𝗹𝗮𝘁𝗳𝗼𝗿𝗺𝘀) This is building with blocks instead of syntax. Drag-and-drop tools handle logic, integrations, and workflows so non-engineers can ship MVPs, automate processes, and launch apps quickly. It trades deep customization for speed and accessibility. The real takeaway: These aren’t competing approaches, they’re complementary. Traditional coding powers complex platforms. Vibe-coding accelerates developers. No-code empowers builders. The best teams mix all three, choosing the right approach based on speed, scale, and complexity - not ideology. Build with what fits the problem. That’s how modern products ship.

Over the last year, I’ve seen many people fall into the same trap: They launch an AI-powered agent (chatbot, assistant, support tool, etc.)… But only track surface-level KPIs — like response time or number of users. That’s not enough. To create AI systems that actually deliver value, we need 𝗵𝗼𝗹𝗶𝘀𝘁𝗶𝗰, 𝗵𝘂𝗺𝗮𝗻-𝗰𝗲𝗻𝘁𝗿𝗶𝗰 𝗺𝗲𝘁𝗿𝗶𝗰𝘀 that reflect: • User trust • Task success • Business impact • Experience quality    This infographic highlights 15 𝘦𝘴𝘴𝘦𝘯𝘵𝘪𝘢𝘭 dimensions to consider: ↳ 𝗥𝗲𝘀𝗽𝗼𝗻𝘀𝗲 𝗔𝗰𝗰𝘂𝗿𝗮𝗰𝘆 — Are your AI answers actually useful and correct? ↳ 𝗧𝗮𝘀𝗸 𝗖𝗼𝗺𝗽𝗹𝗲𝘁𝗶𝗼𝗻 𝗥𝗮𝘁𝗲 — Can the agent complete full workflows, not just answer trivia? ↳ 𝗟𝗮𝘁𝗲𝗻𝗰𝘆 — Response speed still matters, especially in production. ↳ 𝗨𝘀𝗲𝗿 𝗘𝗻𝗴𝗮𝗴𝗲𝗺𝗲𝗻𝘁 — How often are users returning or interacting meaningfully? ↳ 𝗦𝘂𝗰𝗰𝗲𝘀𝘀 𝗥𝗮𝘁𝗲 — Did the user achieve their goal? This is your north star. ↳ 𝗘𝗿𝗿𝗼𝗿 𝗥𝗮𝘁𝗲 — Irrelevant or wrong responses? That’s friction. ↳ 𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝗗𝘂𝗿𝗮𝘁𝗶𝗼𝗻 — Longer isn’t always better — it depends on the goal. ↳ 𝗨𝘀𝗲𝗿 𝗥𝗲𝘁𝗲𝗻𝘁𝗶𝗼𝗻 — Are users coming back 𝘢𝘧𝘵𝘦𝘳 the first experience? ↳ 𝗖𝗼𝘀𝘁 𝗽𝗲𝗿 𝗜𝗻𝘁𝗲𝗿𝗮𝗰𝘁𝗶𝗼𝗻 — Especially critical at scale. Budget-wise agents win. ↳ 𝗖𝗼𝗻𝘃𝗲𝗿𝘀𝗮𝘁𝗶𝗼𝗻 𝗗𝗲𝗽𝘁𝗵 — Can the agent handle follow-ups and multi-turn dialogue? ↳ 𝗨𝘀𝗲𝗿 𝗦𝗮𝘁𝗶𝘀𝗳𝗮𝗰𝘁𝗶𝗼𝗻 𝗦𝗰𝗼𝗿𝗲 — Feedback from actual users is gold. ↳ 𝗖𝗼𝗻𝘁𝗲𝘅𝘁𝘂𝗮𝗹 𝗨𝗻𝗱𝗲𝗿𝘀𝘁𝗮𝗻𝗱𝗶𝗻𝗴 — Can your AI 𝘳𝘦𝘮𝘦𝘮𝘣𝘦𝘳 𝘢𝘯𝘥 𝘳𝘦𝘧𝘦𝘳 to earlier inputs? ↳ 𝗦𝗰𝗮𝗹𝗮𝗯𝗶𝗹𝗶𝘁𝘆 — Can it handle volume 𝘸𝘪𝘵𝘩𝘰𝘶𝘵 degrading performance? ↳ 𝗞𝗻𝗼𝘄𝗹𝗲𝗱𝗴𝗲 𝗥𝗲𝘁𝗿𝗶𝗲𝘃𝗮𝗹 𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝗰𝘆 — This is key for RAG-based agents. ↳ 𝗔𝗱𝗮𝗽𝘁𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝗦𝗰𝗼𝗿𝗲 — Is your AI learning and improving over time? If you're building or managing AI agents — bookmark this. Whether it's a support bot, GenAI assistant, or a multi-agent system — these are the metrics that will shape real-world success. 𝗗𝗶𝗱 𝗜 𝗺𝗶𝘀𝘀 𝗮𝗻𝘆 𝗰𝗿𝗶𝘁𝗶𝗰𝗮𝗹 𝗼𝗻𝗲𝘀 𝘆𝗼𝘂 𝘂𝘀𝗲 𝗶𝗻 𝘆𝗼𝘂𝗿 𝗽𝗿𝗼𝗷𝗲𝗰𝘁𝘀? Let’s make this list even stronger — drop your thoughts 👇

Last week, I described four design patterns for AI agentic workflows that I believe will drive significant progress: Reflection, Tool use, Planning and Multi-agent collaboration. Instead of having an LLM generate its final output directly, an agentic workflow prompts the LLM multiple times, giving it opportunities to build step by step to higher-quality output. Here, I'd like to discuss Reflection. It's relatively quick to implement, and I've seen it lead to surprising performance gains. You may have had the experience of prompting ChatGPT/Claude/Gemini, receiving unsatisfactory output, delivering critical feedback to help the LLM improve its response, and then getting a better response. What if you automate the step of delivering critical feedback, so the model automatically criticizes its own output and improves its response? This is the crux of Reflection. Take the task of asking an LLM to write code. We can prompt it to generate the desired code directly to carry out some task X. Then, we can prompt it to reflect on its own output, perhaps as follows: Here’s code intended for task X: [previously generated code] Check the code carefully for correctness, style, and efficiency, and give constructive criticism for how to improve it. Sometimes this causes the LLM to spot problems and come up with constructive suggestions. Next, we can prompt the LLM with context including (i) the previously generated code and (ii) the constructive feedback, and ask it to use the feedback to rewrite the code. This can lead to a better response. Repeating the criticism/rewrite process might yield further improvements. This self-reflection process allows the LLM to spot gaps and improve its output on a variety of tasks including producing code, writing text, and answering questions. And we can go beyond self-reflection by giving the LLM tools that help evaluate its output; for example, running its code through a few unit tests to check whether it generates correct results on test cases or searching the web to double-check text output. Then it can reflect on any errors it found and come up with ideas for improvement. Further, we can implement Reflection using a multi-agent framework. I've found it convenient to create two agents, one prompted to generate good outputs and the other prompted to give constructive criticism of the first agent's output. The resulting discussion between the two agents leads to improved responses. Reflection is a relatively basic type of agentic workflow, but I've been delighted by how much it improved my applications’ results. If you’re interested in learning more about reflection, I recommend: - Self-Refine: Iterative Refinement with Self-Feedback, by Madaan et al. (2023) - Reflexion: Language Agents with Verbal Reinforcement Learning, by Shinn et al. (2023) - CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing, by Gou et al. (2024) [Original text: https://lnkd.in/g4bTuWtU ]