UX Digest - mejores posts

KPIs Are Not the Problem: Why Solving the Right UX Issues Improves Performance KPIs are symptoms, not causes. Teams skip diagnosis and jump to A/B tests. Framework: problem unclear → research; solution clear → test. Users need two answers: "Why should I?" (copy) and "Can I easily?" (design). Example: removing login before checkout increased conversion 45%. Research creates understanding, experimentation creates proof. KPIs lag experience quality. Fix the experience, not the metric Research: 2026 Emerging Technology Trends from J.P. Morgan Four predictions: 1) Context-driven architectures (Physical AI, knowledge graphs, MCP, RL environments). 2) Inference demand drives AI buildout. 3) Intent replaces app switching (agentic browsers, AI-native workspaces). 4) AI simulation enhances testing (synthetic users). Core theme: AI success depends on agents securely accessing relevant data and tools. Governance must evolve with adoption NNG: Boost Design Autonomy with an Information Pipeline A four-step framework for building influence over product direction by closing the information gaps that large, complex organizations create Prototyping: We Don’t Want Menus. We Want Conversations People don't want to navigate menus—they want to state their problem once and get it resolved. Traditional systems force users into predefined categories, but users think in stories, systems think in labels. Shift from screen-first to intent-first design: ask what users need, not where to go. People don't wake up wanting to navigate interfaces—they wake up wanting problems solved. The best experience begins with "Here's what I need" AI: Is your AI research giving you a False Negative? AI can miss important insights in qualitative data because LLMs rely on frequency—if a user says something critical once or uses subtle language, AI may ignore it. The fix: treat AI as a junior analyst. Manually code some data first, use multi-layer prompting, and maintain a "chain of custody" log. If you hand off data blindly to AI and it misses a pain point, you'll never know it was there Case Study: We thought we knew our users. Then we watched A case study on field observation for a palm-scanning payment device. At a food market, people walked away or refused for religious reasons ("the mark of the beast"). At a corporate office, trust was higher. Key insight: moderated sessions can't capture real-world reactions—field observation reveals a more honest picture of users @uxdigest

Design impact: outcomes over output Design impact is often measured by activity (screens, components, research sessions)—describing what was done, not what changed. Focus on three levels: experience quality (task success, error rate), product outcomes (conversion, retention), and organizational impact (faster delivery, less rework). Define expected outcomes upfront, combine quantitative and qualitative data, and speak business language: "We reduced drop-off" beats "We improved the UI." Design impact is about what changes, not what we create 🎥 NNG: Analyzing Good Designs - Figma’s Shortcut In Figma’s Shortcut, typography and other elements are aligned to a grid, a clear visual hierarchy is established, and various design elements are used consistently in the design Case Study: Improving the Experience of Visiting Public Hospitals A UX case study focused on hospital visitors—an overlooked user group facing disorientation and stress. The solution: a mobile web tool where visitors scan a QR code to register, find patient rooms, and get step-by-step navigation guidance Prototyping: Dark Mode Design Systems - A Complete Guide to Patterns, Tokens, and Hierarchy Dark mode needs a design system foundation, not an afterthought. Key principles: 4 surface elevation levels with luminance stepping (not shadows), semantic tokens, and perceptual color mapping (preserve hue, adjust luminance). Design dark-first, use mode-based organization, and export via CSS variables. Avoid pure black (causes eye strain), respect system preference, and offer a manual toggle AI: AI Adoption in UX - Identify Your Level and Understand Where You Stand A five-stage maturity framework: Awareness → Embracing → Experimentation → Scaling → Transformation. True AI value balances efficiency, user impact, and business impact—not just speed. Progress isn't about using more tools but closing gaps in skills, workflows, or alignment. The key question: "Why aren't we seeing better outcomes yet?" Use AI wisely and purposefully, not just more Book: The Best Books on UX Research — Book 1. It’s Our Research Key lesson: interview your stakeholders before your participants. Five questions to ask PMs, engineers, and designers: What are we building and why now? What unknowns keep you up at night? What assumptions need verification? Who are the users? What are the priorities and timeline? Don't skip this step Opinion: The Invisible Impact of Design Decisions We Rarely Talk About User-centered design alone isn't enough—it ignores broader consequences. Designers must consider non-users, future users, and the larger system. Practical steps: ask "what happens next?", challenge default metrics, and design with restraint. Good design isn't just about making things work—it's about understanding ripple effects Interesting: The Unstable Shelf - Rising to the Tabletop A critique of the _Red Rising_ board game as a case study in how passion for source material can hurt design. Slavish loyalty to the books (112 unique cards, character pair bonuses) created an overcomplicated, messy experience. Faithfulness came at the expense of player experience. Passion can lead to worse products @uxdigest

How To Improve UX In Legacy Systems A guide to improving UX in legacy systems—slow, decade-old "black boxes" critical to daily operations. One broken legacy step makes the entire product feel broken. Start by mapping workflows and dependencies. Choose a strategy: incremental migration, parallel migration (beta alongside legacy), or legacy UI upgrade + public beta. Build stakeholder trust, report progress. Revamping legacy is tough, but the impact is enormous BOOK EXCERPT: The Crisis Worth Using Crisis engineering uses organizational crises as windows for rapid change. Five indicators: fundamental surprise, sensemaking failure, core process degradation, high visibility, rigid deadline. When these align, crises create opportunities to build something better. The question isn't if a crisis will reshape you—it's whether you'll be ready to direct it What Resume Inflation Is Really Telling Us Resume inflation is a symptom of a broken system. Companies post unrealistic job descriptions, so candidates rationally stretch the truth to compete. Honest candidates get punished. The solution: honest job descriptions and honest resumes. The resumes aren't the disease—they're the fever. Fix the system NNG: Handmade Designs - The New Trust Signal In an era of AI-generated-everything, AI-fatigued users want designs that look like they were made by a person AI: The AI Trap for Designers in 2026 - Why Constantly Learning New Tools Is a Dead End — and How to Become a Truly AI-Powered UX Designer Designers who chase every new AI tool are mistaking technical proficiency for real growth. Instead, focus on three rules: design for user "intent" (not just clicks), obsess over the final 5% of execution (edge cases, micro-interactions), and use AI as a sparring partner (simulate personas, get strategic advice) rather than a content generator. The core message: AI-amplified designers will replace tool-chasers, but value lies in strategic thinking, not mastering every plugin Prototyping: Lean UX Research - Validating an MVP Quickly and Cheaply A guide to validating an MVP with minimal budget (~$100, two weeks). Combine a lightweight survey (direct messaging for responses, not just posting links) with an unmoderated field study using the Experience Sampling Method (ESM)—an automated diary study with daily check-ins to capture real-time behavior, not memory. Turn insights into testable hypotheses (e.g., daily goal-alignment tasks). Key takeaway: even a short survey or mini-ESM beats designing in isolation Experience: We didn’t mean to build this— engagement at any cost How well-meaning designers become complicit in broken systems. Success metrics focused on engagement ignore human costs. When flawed briefs pass to AI agents, each step multiplies harm without accountability. Ethical frameworks exist but are ignored because they hurt profit. Profits are chosen over people. Good intentions aren't enough—designers must learn to refuse Opinion: The Entropy Offset - Friction is the new Effort The classic Value/Effort ratio is obsolete because AI has reduced implementation effort. Replace it with Value/Friction, where friction = user cognitive load (discovery + adoption). Prioritize High Value/Low Friction first. User attention is now the bottleneck, not development time. Ask "Should we do this?" not "What order?" Metrics: Behavioral Loops and the Architecture of Retention Three loop types: risk reduction (Slack), artificial (Candy Crush), hybrid (TikTok). Key insights: transition friction breaks momentum; internalization (deliberate → automatic) is the milestone. Metrics: loop depth, return elasticity, engagement amplitude (flat = fading). Optimizing individual features misses the point—a loop only works as a whole @uxdigest

Credible vs. Confidence Intervals: Different Meanings but Similar Decisions Confidence intervals are hard to interpret correctly (no "95% probability" of containing the true value). Credible intervals do allow that natural interpretation. But both methods produce nearly identical numerical ranges. The difference is in what we can say, not the numbers. Use either, focus on clear communication. If endpoints lead to the same decision, you have enough data Personalization vs. Customization: Crafting Tailored Experiences in UX Personalization (system adapts for you) and customization (you configure for yourself) solve different problems. Personalization reduces effort but risks trapping users in past preferences. Best approach: combine both—personalization for a smart start, customization for ongoing control. Add exploration modes to break the loop. Designers shape choices, not just interfaces NNG: AI Agents as Users AI agents now interact with digital interfaces alongside humans. Designing for both requires rethinking what "user" means and prioritizing accessibility AI: AI moderated interviews - methodological error amplified AI-moderated interviews collapse qualitative discovery and quantitative measurement into one flawed pass, committing "acontextual counting"—treating all responses as equally weighted. Scale (80,000 interviews) doesn't fix this: you can't count "it" before you know what "it" is. A classic mixed-methods design would work better Prototyping: How I Built an Enterprise Design System for 50+ Insurance Apps — Without a Design Team A solo designer built a design system for 50+ insurance apps by starting with design tokens (colors, spacing, typography) before components, enabling multi-brand theming without duplicate work. Then built 60+ accessible components, prioritized adoption by speaking engineers' language. Results: 80% less inconsistency, 40% faster handoff. Start with tokens, document as you build Opinion: The trust-latency gap - why the future of UX is intentionally slower As AI speeds up decisions, trust decreases. The "trust-latency gap" is the distance between execution speed and the time humans need to feel confident. For high-stakes actions, "strategic friction" (intentional delays like confirmation steps) builds trust. The key question: not "how fast?" but "how fast should it feel?" Visual: Make the user to look where you want them to look - the guide on guiding attention Guide to directing attention in dashboards using the mantra: overview first → details on demand. Techniques: layout (left-to-right), size (big numbers first), color (highlight key sections), arrows, text hints, icons, and interactivity. Design is not about beauty—it's about guiding attention. Consistency is key Basics: From Research to Design - How UX Turns User Behavior into Real Solutions A structured process: affinity mapping → thematic clustering (trust precedes action, fear is a UX constraint) → behavioral model (explore→verify→act, not search→select→book) → design principles (reassurance before action, reduce cognitive load). Core insight: users are not slow—they are careful. In high-stakes scenarios, UX is about making users feel certain enough to act Interesting: Turns Out, Everyone Does UX. They Just Don’t Know It Yet A UX designer started a podcast and discovered that people in other fields (architects, artists) already do UX thinking—observing behavior and solving for users—they just don't have a name for it. The podcast itself became a practice in asking good questions and listening without steering @uxdigest

A Practical Guide To Design Principles The article breaks down core design principles like balance, contrast, hierarchy, and repetition into practical steps for immediate application in digital projects, with real examples and checklists. It stresses using them as decision-making frameworks during iteration, not rigid rules, to create intuitive, cohesive user experiences that scale across devices The Earpiece Isn’t Ready Yet. Neither Are We The article argues that we're not psychologically or socially ready for seamless AR earpieces that overlay digital info on reality, as they risk overwhelming attention, eroding real human connection, and creating dependency on filtered experiences. Author warns current prototypes ignore deeper behavioral readiness, urging designers to prioritize cognitive limits and interpersonal authenticity over technical dazzle NNG: AI Interviewers AI interviewers can conduct user interviews on your behalf, but they come with real limitations. Learn how they work, how well they perform, and the best use cases for adding them to your research toolkit Prototyping: Stop Making These 7 UX Mistakes (Fix Them Like a Pro) The article lists 7 common UX mistakes like designing for yourself instead of users, poor navigation, cluttered layouts, unlabeled icons, weak CTAs, ignoring mobile responsiveness, and skipping user feedback. It offers pro fixes such as user research upfront, clear information hierarchy, progressive disclosure, icon labeling, and regular usability testing to prioritize actual needs over assumptions Process: How to Turn User Research Into UX Design – The Four-Hour Design Sprint The four-hour design sprint compresses user research insights into rapid UX outcomes through structured phases: synthesize findings (1h), ideate solutions (1h), decide & storyboard (1h), and prototype key flows (1h). It prioritizes high-impact problems, fosters cross-team alignment, and delivers testable wireframes—proving research-to-design translation doesn't need days of workshops AI: Designing with AI - Moving Beyond Tools to Orchestrate Meaningful Experiences Designers should use AI not just as a tool for generating assets, but as a partner to orchestrate adaptive, context-aware experiences that feel meaningful and human-centered. The article emphasizes moving beyond efficiency gains to focus on emotional resonance, ethical personalization, and new interaction paradigms enabled by AI's understanding of user intent @uxdigest

From Benchmark to Decision The article explains how UX benchmarking goes beyond raw metrics (like SUS scores or task success rates) to drive actionable product decisions. It emphasizes prioritizing business-aligned KPIs, tracking changes over iterations, and using competitor/industry baselines to justify redesigns and demonstrate ROI NNG: AI Can Help with Survey Writing, But It Still Requires Human Expertise AI can produce polished survey drafts quickly, but experienced human review is still needed to catch subtle survey-design flaws that weaken data quality Prototyping: Lost in transactions - designing a human-readable activity for crypto wallet The article describes redesigning crypto wallet transaction history from raw blockchain data into human-readable stories that show real-world context like "bought coffee" or "received salary," reducing confusion and cognitive overload. Key approach: pattern matching, natural language summaries, and visual timelines to make complex technical activity feel intuitive and trustworthy AI: Conversational UX is not the shift. Judgment is Conversational UX isn't just about chat interfaces—the real shift lies in building human judgment into AI systems for context-aware, adaptive responses. The article argues designers must focus on intent recognition, error recovery, and ethical decision-making rather than surface-level dialogue flows Experience: What My Grad Thesis Taught Me The graduate thesis taught the author key lessons in research rigor, iteration through failures, and balancing deep focus with broader career skills like communication and resilience. Ultimately, it showed that academic work builds not just knowledge, but adaptability and storytelling for real-world impact beyond graduation Design: The dark side of dark patterns — and how to design ethically Dark patterns trick users into unintended actions like hidden subscriptions or fake scarcity to boost short-term metrics at the cost of trust. The article urges ethical alternatives like transparent choices, progressive disclosure, and frictionless exits to build genuine loyalty instead of manipulative compliance @uxdigest

Bayes’ Law in UX Research: The Power and Perils of Priors Bayesian reasoning in UX research uses prior beliefs (like historical benchmarks) combined with new data to update confidence in hypotheses about user behavior, such as task completion rates. The article warns that subjective priors can heavily sway results—strong priors need robust justification, while weak data demands caution to avoid overconfident conclusions NNG: A Concrete Definition of an AI Agent An AI agent pursues a goal by iteratively taking actions, evaluating progress, and deciding next steps. Useful agents must be reliable, adaptive, and accurate Prototyping: Great products are built on the opportunities surrounding them Great products succeed by solving the unmet needs and opportunities around core problems, not just the problems themselves—like how Uber didn't just create rides but built an entire ecosystem around urban mobility gaps. The article emphasizes mapping adjacent opportunities through user journeys to create defensible, expansive product ecosystems that drive retention and network effects AI: The rise of synthetic users Synthetic users—AI-generated personas for UX testing—are rising as fast, scalable alternatives to real participants for early validation, hypothesis testing, and catching obvious friction points. While they excel at speed and cost-efficiency for prototypes and edge cases, the article cautions they're no replacement for human emotional depth, biases, and unpredictable behaviors in final research Experience: I Thought I Knew My Users. Then I Visited Their Homes Visiting users' homes revealed how much richer their real-life context, frustrations, and workarounds were compared to remote interviews or analytics alone. The experience transformed the designer's assumptions into empathy-driven features that better solved actual pain points in daily routines Basics: Unmoderated usability testing — some mistakes to avoid The article outlines common pitfalls in unmoderated usability testing, such as vague task instructions, poor participant recruitment, and skipping pilot tests. Key advice includes writing crystal-clear scenarios, over-recruiting participants to account for dropouts, and allowing ample time since there's no live moderator to guide users Interesting: How LinkedIn Uses UX to Keep You Coming Back (Even When You Don’t Need a Job) LinkedIn keeps people coming back by making the product useful even when they are not job hunting: feed content, networking prompts, profile completion nudges, and social proof all create a habit loop. The piece argues that the real UX goal is retention through relevance, identity, and low-friction re-engagement, not just job search functionality @uxdigest

UX Doesn’t Stop at the Platform. Neither Should Research UX research shouldn't stop at the platform—real user behavior often happens outside system metrics, like drivers gaming algorithms to maximize income. The article urges researchers to look beyond dashboards and study the full context: environments, workarounds, and hidden incentives that shape decisions. True insight comes from questioning the system, not just validating it NNG: Outcome-Oriented Design - The Era of AI Design Outcome-oriented design shifts how we approach UX in the AI era. Instead of designing single interfaces, designers now define adaptive frameworks that respond to individual user goals rather than optimizing for average user needs Prototyping: Stop Designing Screens. Start Designing Outcomes The article argues designers should stop optimizing screens and start designing for user outcomes—what people actually want to achieve. With AI and "invisible UX," the best experiences minimize steps and anticipate intent, not just look polished. Shift focus: measure success by how fast users get results, not how pretty the flow is AI: The Physics of Great UX - Making Digital Interfaces Feel Real Great UX feels real by applying physics principles—gravity, momentum, elasticity, resistance—to digital motion, making interfaces intuitive. Interactive animations tied to user actions (via tools like Lottie's State Machines) create tactile, responsive experiences. Build a consistent motion system, not just isolated effects, so your product feels cohesive and predictable Experience: I don’t collect teapots - What happened when we fixed feedback in our design team The article describes how a design team replaced vague, performative feedback ("I collect teapots") with specific, actionable critiques, transforming their collaboration and output. The key fix: train teams to give concrete, behavior-focused feedback tied to user goals, not personal taste. Result: faster iterations, clearer communication, and designs that actually work for users @uxdigest

How to Use Banner Tables to Present Survey Results Banner tables compress complex survey data into a single, scannable view, making it easier to compare metrics across multiple demographic segments. Though standard in market research, they are underused in UX but highly effective for large-scale studies requiring segmentation analysis. The article demonstrates how to build them using R, enabling researchers to efficiently present weighted and unweighted results side by side NNG: GenUI vs. Vibe Coding: Who’s Designing? With generative UI, the AI system decides to generate an interactive element or entire product in response to a user need. Vibe coding is when users request the AI to build it Prototyping: AR glasses are here, but what about accessibility? As AR glasses emerge, the article urges designers to prioritize accessibility early—using multi-sensory features like haptics and audio—to support users with disabilities. These universal design choices, such as speech-to-text or enhanced sound cues, ultimately improve the experience for everyone. The core message: inclusive AR isn't an add-on, but a foundation for better tech AI: UX Research in the Era of AI AI isn't the main force changing UX research—organizational power dynamics are. The enduring value of researchers lies in providing ground truth and wielding influence, which AI cannot replicate. To stay impactful, focus on understanding power structures, adapting to change, and communicating clearly Basics: I Bombed a User Interview. Here’s What I Learned The author reflects on bombing a user interview with a pro user, learning to avoid closed questions and surface-level answers. Key fix: adapt questions to experts by focusing on team workflows and real pain points. Success comes from active listening and flexibility, not a rigid script @uxdigest