I Thought I Already Knew How to Do Research

A reflection on the difference between collecting findings and actually synthesizing them into insights. The key shift is building a traceable chain from evidence to patterns, interpretations, and product decisions

NNG: Do Participants Say Less to AI Than to Human Moderators?

In a controlled study, participants spoke 45% less with AI moderators, while AI-led interviews were 37% shorter and much more moderator-dominated. The result suggests AI moderation can reduce participant elaboration, although more talking does not automatically mean better insights

Prototyping: We’re Trying Hard to Design the Right Solution, But Did We Design the Right Path?

A product can be built for the right audience while the path to its value still ignores their needs. The author uses Inflow’s 37-step onboarding for people with ADHD to show why reducing friction and demonstrating value before collecting sensitive data can matter

AI: Vibe coding alone

Vibe coding let the author turn a decade-old budgeting spreadsheet into a working tool without a developer — and made her question what gets lost when everyone starts replacing colleagues with AI. The punchline comes later: the tool wipes her data, and Claude can’t help her recover it

Opinion: Why Researchers Need to Build

Researchers lose context when insights pass through reports, designers, and developers before becoming a product. Learning to prototype and build lets them turn research directly into testable solutions without becoming full-time engineers

Design: If You Don’t Get Lost in a City, a Designer Probably Did Their Job Well

Copenhagen’s public transport shows how wayfinding, consistency, and progressive disclosure can make a complex system feel effortless. Good design gives people the right information at the right moment and often works best when users barely notice it

Basics: Low Fidelity - A Concept First Approach to Design

Low-fidelity prototypes help teams test the concept before visual polish makes a weak idea look convincing. Their value is not speed itself, but exposing gaps, inviting critique, and making problems cheaper to fix early

@uxdigest

From Research to Recommendations: Closing Out My LFX Mentorship

User interviews and competitive analysis shaped a new information architecture for the OpenTelemetry Ecosystem Explorer. The project shows how research findings can become reusable product principles instead of ending as a one-off report

NNG: The 3 Roles of Context for AI Agents

AI agents work better when context is split into three roles: global rules and preferences, local task-specific information, and ambient streams like email or meeting transcripts. Managing these layers well can make detailed prompting less important because the agent already understands the user, the project, and the surrounding work context

AI: What Happens When a Screen Has More Than One Future?

What happens when an interface stops following a predefined flow and starts generating its next state in real time? Using Runway’s Solaris as a starting point, the article explores generative UI where the same action can lead to different outcomes. The tricky part isn’t generating the interface — it’s deciding which of its hundred possible futures is actually allowed to become real

Experience: I’m starting to hate UX research and I don’t know if it’s the job or the career

A solo researcher describes juggling six product areas, doing everything from recruiting to synthesis, and watching research repeatedly get ignored. The discussion quickly shifts from burnout itself to a bigger question: whether the problem is UX research as a career or working in organizations with low research maturity

Case Study: Designing for focus in a product designed for engagement

A UX case study exploring a Focus Mode for YouTube that reduces Shorts and unrelated recommendations during study sessions without blocking access. Research with students shaped the concept around user intent, content filtering, and lightweight reminders instead of hard restrictions

Design: Beyond Aesthetics - 3 Surprising Ways Color Theory Shapes Your Reality

Color is more than decoration: it shapes emotion, trust, attention, and behavior in interfaces. The article focuses on three practical ideas — psychology, accessibility through contrast, and choosing colors intentionally to guide users

@uxdigest

The MUSe Framework: How to Measure the Functional Scale of a Digital Product

MUSe is a lightweight framework for estimating product scale through modules, structural depth, unique components, and key user scenarios. It helps teams compare B2B products without confusing the number of screens with actual functional complexity

NNG: How AI Works and How Users Think About It - Study Guide

A curated guide to NN/g’s research on how AI works and how users actually understand and interact with it. It covers prompting patterns, trust and anthropomorphism, AI search behavior, mental models, adoption, and AI literacy

Prototyping: A lil guide to optional fields - why “optional” cannot be optional if the page disagrees

An “optional” field is misleading if skipping it changes what the user gets, breaks functionality, or reduces personalization. The author suggests either explaining the consequence, making the field required, or asking for the data only at the moment it becomes necessary

AI: How I’m Using AI for User Research

A practical look at using AI to analyze thousands of reviews, vibe-code research tools, and build a local RAG research repository. The author also stresses privacy, source traceability, and the need to stay critical of confident AI outputs

Case Study: Improving bus booking experience on Nuego

The case study redesigns NueGo’s booking flow around clearer pickup and drop-off selection, better trip review, and a new round-trip option. The main changes use distance, maps, location details, and stronger visual hierarchy to reduce confusion and booking errors

Experience: The Journey Map of Two Design Students in Advocacy Spaces

Two design students reflect on how advocacy work challenged their usual design-thinking process, especially when real people could not easily articulate the problems they faced. By changing their research methods and iterating in the field, they learned that meaningful change often starts with helping people notice and express issues before trying to solve them

Opinion: Successful UX Never Ends

UX is not a one-time phase that finishes at launch but a continuous cycle of observing, learning, and improving the product. The article argues that mature teams treat research, feedback, and iteration as an ongoing operating model rather than occasional project work

Basics: Understanding Website Analysis - Exploring Cognitive Biases and UX Laws

The article explores how cognitive biases and common UX laws can be used to analyze websites. It focuses on understanding why certain interface patterns influence user attention, behavior, and decision-making

@uxdigest

Using Synthetic Data Responsibly: A Practical Framework for Researchers

Synthetic data can help researchers explore ideas and narrow hypotheses, but it should not be treated as evidence from real users. The framework focuses on bias, validation, privacy, documentation, and combining synthetic responses with real participant data

NNG: When Should You Disclose AI Use? The PACED Framework

AI disclosure doesn’t consistently increase or reduce trust — reactions depend on the audience, context, and how much AI contributed. The PACED framework suggests deciding based on five factors: policy, audience, context, expectations, and degree of AI use

Prototyping: The Frictionless Design Trap - Why the Right UX Decision in High-Stakes Software Is to Slow People Down

In high-stakes products, removing every bit of friction can increase the chance of costly mistakes. The article argues for intentional pauses, confirmations, and extra steps when they help users notice risk and make more deliberate decisions

AI: I Ran the Same Usability Study Twice — Once with AI, Once as a Human Moderator. Here’s What I Learned

AI moderation works well for opinions, naming, and broad discovery, but breaks down in complex task-based usability studies. The main risk is that participant confusion gets mistaken for product problems, turning bad sessions into misleading research data

Opinion: UX Is Moving Beyond Screens - What Google Research on Self-Evolving Interfaces Means for Designers

Google’s research on adaptive generative interfaces suggests software could reshape itself around user intent instead of forcing users through fixed navigation. That shifts UX work from designing screens toward defining the rules, constraints, and systems that let interfaces safely generate themselves

SEO: Boost Your SEO - How User Experience Can Make or Break Your Rankings

UX affects SEO through factors such as page speed, mobile responsiveness, navigation, content quality, and user engagement. Improving the experience can help visitors stay longer, explore more pages, and make the site more useful to both users and search engines

Basics: Good UX vs. Poor UX - The Difference Users Feel But Can’t Always Explain

Good UX is less about visual polish and more about reducing uncertainty, effort, and unnecessary thinking. The article breaks this down through navigation, forms, feedback, errors, accessibility, consistency, performance, and visual hierarchy

@uxdigest

Research Beyond Research

UX research should be optimized not around studies and methods, but around reducing uncertainty in product decisions. AI makes generating hypotheses cheaper, which increases the value of researcher judgment about when existing evidence is enough and when reality still needs to be tested with users

NNG: Empathy Mapping - The First Step in Design Thinking

Empathy maps organize what users say, think, do, and feel to create a shared understanding and expose gaps in research. NN/g also warns against filling them with assumptions or letting AI invent evidence that cannot be traced back to real user data

AI: ChatGPT vs Humans for Coding Open-Ended Responses, Three Years Later

ChatGPT-5.6 matched human coders on open-ended survey responses and showed near-perfect consistency across repeated runs. Compared with GPT-4, it also produced more granular themes while maintaining strong agreement with human researchers

Design: The Missing Data in Design

Product teams are getting faster at designing and building, but not always better at understanding whether users actually need what they create. The article argues for grounding product decisions in research, analytics, and observed behavior instead of assumptions, stakeholder opinions, and competitor copying

Basics: The Tiny UX Detail That Changed How I Saw the Entire Product

A tiny “Continue” button exposed a much bigger UX problem: users often hesitate not because of the control itself, but because they don’t understand what will happen next. The article reframes UX as designing a sequence of decisions, where each step gives users enough context, feedback, and reassurance to keep moving

@uxdigest

The Obsession with Being Data-Driven

Being data-driven can become a trap when teams mistake metrics for truth and use numbers to avoid judgment, context, and real conversations with users. The article argues for combining quantitative data with qualitative insight, treating metrics as evidence rather than the final decision-maker

Standard Accessibility Is Not Truly Inclusive

Standard accessibility checklists often stop at interface-level fixes and miss cognitive load, executive function, emotional safety, and cultural context. The author argues that inclusive design should treat accessibility standards as a baseline, not a complete solution

NNG: The New Big Ball of Mud - Why Agentic AI Systems Turn Fragile

AI makes it so cheap to build and extend agentic systems that users often overengineer them into fragile “big balls of mud” they no longer fully understand. NN/g argues that the key skill is not keeping everything perfectly organized, but periodically reviewing what each part does and knowing when it’s time to rebuild

AI: What the Bible got right about AI

The article uses biblical stories about creation, idolatry, and human ambition to frame today’s relationship with AI. Its central argument is that AI is less a new kind of mind than a mirror of humanity — carrying our knowledge, biases, and tendency to hand power to things we created

Prototyping: Data Visualisation – Stop Bad Design Ruining Good Data

Choosing the right chart starts with the question the user needs to answer, not with visual preference or a dashboard template. The article explains how different encodings suit comparison, trends, outliers, composition, and flow — and why strong data UX often combines several complementary visualisations

Case Study: From Looking at Art to Connecting With It

A UX case study on why young museum visitors can be interested in art yet still lose engagement during a visit because they lack context, storytelling, and guidance. The team designed and tested Museoko, a personalized museum experience built around tailored routes, richer context, and helping visitors connect artworks to their own interests

@uxdigest

Hick’s Law: Why Netflix Never Shows You Everything

Hick’s Law explains why more options increase decision time, making hierarchy and grouping as important as the number of choices itself. Netflix applies this by showing a small personalized slice of its catalog, grouping titles into rows, and helping users judge options faster

NNG: Context Libraries - How AI Power Users Manage Context Engineering

Power users spend more effort managing context libraries than crafting prompts, but those libraries quickly become stale, messy, and hard to navigate. NN/g recommends letting agents maintain context, organizing it for findability, and loading only the information needed for each task

Prototyping: A Practical Guide To Naming Things

A practical guide to naming UI components, colors, design tokens, variables, features, and products consistently. The core idea is to use short, meaningful names rooted in shared team and user language rather than visual properties or internal jargon

Experience: Privacy Is Not Always a Door You Can Close

Research with 199 women in India, Pakistan, and Bangladesh shows that privacy on shared family phones often has to remain invisible to avoid suspicion. Good privacy design should therefore support subtle controls like easy deletion and hidden content, not just obvious locks and “private” modes

Opinion: Recording everything made research harder

Recording and transcription became effortless, but that created more material to review without making important insights easier to find. The author argues for letting researchers mark meaningful moments during interviews while AI handles transcription, organization, and search

Interesting: How Physical Space Proves UX Isn’t Just Digital - The Existential Crisis of the Metro’s EPID Panel

A Delhi Metro advertising banner blocking essential passenger information becomes a case study in why UX extends far beyond screens. The author argues that accessibility, environmental context, and users’ real goals should take priority over marketing visibility in public spaces

AI:

Synthetic users are making me rethink “imperfect research”

Synthetic users can turn assumptions into convincing “findings” and create false confidence, even when their answers look close to real-user data. The author argues they’re useful for rehearsal and preparation, but shouldn’t replace research with actual people

Are synthetic users (really) the future of UX research?

Synthetic users can speed up early research, help form hypotheses, and reduce recruiting overhead, but their answers tend to be generic and overly agreeable. The author argues they should only extend research with real people, never replace it as the primary source of evidence

Building AI Evals - Who Gets to Decide What’s Good?

DoorDash argues that evaluating AI starts with deciding whose judgment should define “good”: the individual user, experts, or the affected community. Their research shows that poorly designed evaluation criteria can misread real user preferences, so human judgment has to shape the rubric before AI can reliably score against it

The hardest part of AI UX? Knowing when to trust it

Good AI UX is not about maximizing trust, but helping users understand when AI is useful, what context it lacks, and when they should step in. The article argues for transparency, human control, and using AI only where it genuinely reduces effort instead of creating more work through validation and correction

@uxdigest

The job is moving from designing screens to designing how a capable system understands, acts for and stays accountable to a person

AI is shifting product design from fixed screens and flows toward systems that assemble interfaces and act on a user’s behalf. That makes control, memory, predictability, accountability, and outcome-based metrics increasingly important design problems

NNG: The New Siri’s Biggest Strength Is Not Intelligence

A useful AI assistant needs relevant personal context as much as raw intelligence. Good AI UX also preserves that context across tasks, makes its limits visible, and lets users verify or correct actions

Prototyping: How I built five new UX trends into my own SaaS product

The author tests five predicted UX trends for 2027 in his own AI product: intent-first interaction, trust and control, AI behavior design, lower cognitive load, and continuous research. His main conclusion is that as AI becomes more capable, UX shifts from designing screens to deciding what the system should do, when it should act, and how humans stay in control

Experience: Fortnightly weeknotes | 9 October

A service designer at Public Health Wales shares updates on weight-management service design, research incentives, recruitment, and working in the open. One useful example shows why teams should focus on user needs rather than requirements: a requested download feature turned out to be unnecessary once users could view the needed information directly

Opinion: The "Average" User is a Myth - Why Designing for the Extremes is Your New UX Superpower

Designing for the “average” user hides real barriers caused by differences in ability, language, context, devices, and environment. Designing for extremes often improves the experience for everyone through clearer language, larger targets, stronger contrast, and more resilient flows

AI:

Building AI Evals - Who Are The “Experts”?

DoorDash shows that AI evaluation depends on choosing the right humans to define what “good” means — individual users, domain experts, or a broader community. For local recommendations, narrowing research to neighborhood-level experts revealed where real consensus exists and where AI should present variety instead of forcing a single “best” answer

If AI can do end-to-end UX research, where is our value?

AI can already plan studies, analyze transcripts, and draft findings, but researchers still decide what is worth pursuing, what context is missing, and what evidence can be trusted. The author argues that our value increasingly lies in judgment between the steps: steering AI, checking its work, and forming an independent interpretation before accepting its conclusions

From Activity to Intent: Generating User Journeys with LLMs

Pinterest replaced a multi-stage clustering pipeline with a 4B LLM that reads user activity and generates ranked, human-readable journeys such as “Modern Farmhouse Kitchen.” The new approach reduces fragmented and generic recommendations, supports multiple languages, and turns long-term intent into a more useful personalization signal

@uxdigest

UX and NPS Benchmarks of Third Party Automotive Websites (2026)

MeasuringU surveyed 561 users across 11 automotive platforms, revealing a striking gap between usability (62nd percentile) and trust (34th percentile), with Kelley Blue Book leading in NPS (+34) and Carvana trailing (−23). The research shows that inaccurate listings, intrusive ads, aggressive lead generation, and privacy concerns undermine customer confidence, suggesting that trustworthy information matters more than polished interfaces when buying a car online

Salesforce paid about 67 times revenue for an agent that runs user interviews, and the research calendar is the next thing to leave the PM’s job

Salesforce is acquiring Listen Labs, an AI platform that automates user recruitment, interviews, and research synthesis, in a deal reportedly worth $2 billion — roughly 67 times the startup’s annualized revenue. The author argues that as AI takes over research operations, product managers must retain responsibility for three things: asking the right questions, ensuring sample quality, and making decisions based on evidence rather than simulated user behavior

Everyone Loved the App. So Why Did They Leave?

A smooth interface and positive usability feedback don't guarantee retention: users may still return to familiar products if the new experience offers no compelling reason to switch. Through a hypothetical grocery app example, the author explains why product teams should investigate user motivations, habits, and actual value before adding features, redesigning screens, or trying to boost engagement metrics

NNG: Synthetic Users vs. Human Respondents - Same Conclusion, Less Variation

Nielsen Norman Group compared responses from 120 synthetic users and 604 real participants: while both groups reached the same overall conclusion, AI generated only 12 distinct answers versus more than 50 from humans, demonstrating a phenomenon known as mode collapse. The study shows why synthetic users may help generate hypotheses but cannot replace real research when uncommon perspectives, audience differences, and unexpected insights matter

Case Study: Saral - Designing clinic software for the doctors India’s software forgot

Three weeks of field research with 13 participants and 854 affinity notes overturned the original product hypothesis: the key user of clinic software wasn't the doctor but the receptionist managing queues, payments, and patient records. The author shows how these findings shaped Saral, a working prototype built around existing habits, WhatsApp, offline access, and minimal disruption to clinical workflows, with usability testing revealing further improvements and unresolved challenges

AI: The Executive Builder - What Product Leadership Actually Looks Like Now

AI-native product leaders increasingly build things themselves, manage agents alongside people, and automate routine work instead of scaling through headcount alone. As execution gets cheaper, product judgment becomes the differentiator: talk to users, understand why you can win, prototype fast, and kill most ideas before they reach production

Prototyping: The Interface Is Where Problems Become Visible. It’s Not Always Where They Begin

Mounika Konidala explains why UX problems that appear in an interface often originate elsewhere — in broken workflows, fragmented information, or organizational processes. Using an example of managers struggling with an internal tool, she argues that designers should investigate the root causes of friction before creating wireframes, because a better screen doesn't necessarily solve the underlying problem

@uxdigest

Redesign or UX Audit? How to Decide Before You Spend

Before investing in a full redesign, teams should conduct a UX audit to identify usability problems, validate them through research, and prioritize improvements by business impact. Lollypop Design shares a practical audit framework, a 30-day client case, and ten diagnostic questions to help decide when targeted fixes are enough and when a complete redesign is justified

What I told the CFO when he asked why design matters

SAP’s Chief Design Officer Arin Bhowmick shares five ways to explain design’s business value to a CFO: return on investment, conversion growth, reduced development risks, design system efficiency, and pricing power. Drawing on research from Forrester, McKinsey, and Baymard Institute, he argues that design leaders should translate their work into financial metrics that executives already understand and trust

NNG: UX Maps - Glossary

Nielsen Norman Group presents a practical glossary of 31 UX mapping methods, from journey maps and service blueprints to ecosystem maps, user flows, and stakeholder maps. Each entry includes a definition, visual example where available, and links to related resources, helping researchers and designers distinguish similar methods and choose the right visualization for their goals

AI: The Realities of Getting a Design System Ready for AI

Henry Daggett shares how Société Générale built an AI interface generator using its existing design system, explaining why model fine-tuning, RAG, and DESIGN.md files failed to deliver reliable results. The team instead restructured component documentation and packaged it into reusable AI Skills with strict generation rules, enabling different AI models to produce consistent interfaces that have already been used in production

Prototyping: The Cost of a Checkbox

A missed checkbox during an online appointment booking created duplicate patient records, wasted staff time, and undermined years of customer trust. Scott Crawford argues that when multiple users make the same mistake, teams should investigate the design rather than blame users, drawing on insights from both customers and frontline employees

Opinion: I Had 10 Minutes on an NVIDIA Livestream. I Treated It Like a UX Problem

Midori Chen explains how she applied UX research principles to prepare a 10-minute NVIDIA livestream presentation about vision-language models, starting with audience needs, prioritizing tangible outcomes, and treating confidentiality as a design constraint. She also describes how rehearsals became a form of usability testing, helping her simplify technical explanations and make the presentation accessible to both technical and non-technical viewers

@uxdigest

Pixels and Psychology: Why Good UI Design is Just Good Empathy

Good interface design goes beyond aesthetics by applying psychology to reduce cognitive load, understand user behavior, and make navigation more intuitive. The author explores three principles — Hick’s Law, empathy-driven research, and recognition over recall — to explain why designing for the human mind matters more than designing beautiful screens

Jakob’s Law: Why Users Want Your App to Work Like Every Other App

Afroza Riju explains Jakob’s Law: users expect new interfaces to behave like products they already know, and breaking familiar patterns can increase cognitive load and frustration. Using examples from Amazon, Snapchat, and YouTube, the article shows why designers should preserve established interaction conventions, introduce major changes gradually, and reserve innovation for improvements that genuinely benefit users

NNG: Prevent Last Minute Design Requests - A Guide

Nielsen Norman Group offers practical guidance on preventing last-minute design requests by helping stakeholders better understand UX work and the effort behind deliverables. The video emphasizes more intentional requests and clearer communication to reduce unexpected deadlines and improve collaboration between designers and stakeholders

AI: Beyond Asimov - Six Laws for Enterprise Agentic AI

Mike Silva, Head of User Research at SAS, proposes six design principles for enterprise AI agents: understand the system, respect permissions, rely on trusted data, follow governance rules, never assume user intent, and explain their actions. Inspired by Asimov’s Three Laws of Robotics, he argues that the future of agentic AI depends not on maximizing autonomy but on building trust through transparency, accountability, and meaningful user control

Instrument: Dovetail - When Customer Evidence Stays Attached to the Answer

A hands-on evaluation of Dovetail shows how AI can synthesize customer interviews, support tickets, and surveys while preserving links to the original evidence, even across different languages. The experiment also reveals an important limitation: citations make AI-generated insights easier to verify, but they don't guarantee accurate interpretations or protect against unsupported conclusions

Prototyping: Calm Commute — An Attention-Aware HMI for a Swappable-Battery Scooter

Yuga Walimbe presents an HMI concept for electric scooters that adapts information and notifications to the rider’s attention, prioritizing safety alerts while postponing less important content until the next stop. Based on rider interviews, observations, and prototype testing, the case explores battery-swap navigation, offline resilience, and physical controls designed to minimize distraction

Basics: Who Owns the Risk You Didn’t Test?

Kristine Breum argues that decisions to skip UX testing should be framed as conscious risk management rather than negotiations over deadlines and budgets. She proposes a practical approach: identify critical assumptions, assess the consequences of being wrong, document who accepts the remaining risks, and establish how the team will monitor them after release

@uxdigest

Test your way to personalisation at scale: How we built a high-impact experimentation engine

Zalando ran 35 experiments, but only eight delivered measurable business value, helping the team exceed its ambitious growth target. The case shows how rapid testing and research-driven prioritisation can outperform intuition

Digital Twins Systematically Overestimated Business Software SUS Scores

MeasuringU compared 289 real users with their AI digital twins across six products: AI inflated SUS scores by 10.5 points on average. Twins ranked products fairly well but proved unreliable for measuring actual usability

NNG: AI in UX - Work Is Faster, But Practitioners Have Mixed Feelings

NN/g surveyed 521 UX professionals: AI speeds up work, but unreliable outputs and privacy concerns limit its benefits. Researchers are more worried about their future roles than designers

AI: AI Is Not Creating a Writing Crisis in Science. It Is Creating an Evaluation Crisis

AI makes scientific papers easier to produce, but evaluating their quality remains costly. The author argues that transparent evidence and verifiable reasoning matter more than who wrote the paper

Design: Why design (still) matters, Taste is not a methodology, Borrowed slopes

UX Collective curates articles on design’s business value, the limits of taste in the AI era, and the importance of listening. The roundup also covers career growth and useful design tools

Opinion: Language Is the Interface

Shangyanyan Li explains how AI agents can analyze online discussions to uncover users’ natural language, recurring problems, and workarounds. The approach helps design better conversational interfaces, but researchers must verify the findings

Interesting: When Two Introverts Fall in Love - A Psychology of Deep Connection and Silent Distance

Two introverts can build deep emotional intimacy through mutual understanding and comfortable silence. But avoiding difficult conversations can turn that same silence into distance and misunderstanding

Prototyping:

Redesigning Grindr: Exploring Trust, Authenticity and Safety Through UX Research

A team of five design students explores trust and safety in Grindr through user journey mapping, app review analysis, literature research, and three user interviews, turning their findings into a redesign focused on profile authenticity. The resulting prototype introduces identity verification and a behavioral trust score, while highlighting a critical UX challenge: making users feel safer without compromising privacy or mistaking verification for actual trustworthiness

Jakob’s Law - Why Users Want Your App to Work Like Every Other App

Afroza Riju explains Jakob’s Law: users expect new interfaces to behave like products they already know, and breaking familiar patterns can increase cognitive load and frustration. Using examples from Amazon, Snapchat, and YouTube, the article shows why designers should preserve established interaction conventions, introduce major changes gradually, and reserve innovation for improvements that genuinely benefit users

@uxdigest

How To Make Your UX Research Hard To Ignore

Researchers gain influence by joining product decisions early, challenging assumptions, and sharing responsibility for risks. This User Weekly issue also highlights practical resources on AI in UX research and communicating insights

My First Content Audit: Webpages, One Long Spreadsheet, and A LOT of Clicking

A student shares lessons from auditing 174 nonprofit webpages, uncovering broken links, duplicate content, and navigation problems. Comparing content quality with traffic data revealed that discoverability was often a bigger issue than the content itself

Fundamentals of UX We Forget: You Are Not the User

Product expertise doesn't make designers or stakeholders representative of real users. The author explains why internal assumptions must be challenged through user research rather than treated as evidence

NNG: UX Disciplines - Glossary

NN/g explains the differences between UX, UI, CX, research, service design, and emerging Agent Experience. The glossary clarifies overlapping responsibilities in a field where one survey identified 134 distinct job titles

Prototyping: Deel vs. Rippling for Contractor Management - Which Platform Gets the User Experience Right?

Nick Babich compares two contractor management platforms across onboarding, payments, automation, compliance, and pricing. Deel suits international freelancers, while Rippling excels at integrating contractors into broader company workflows

Opinion: Sycophancy - Where are the UX designers?

AI's tendency to agree with users is a UX problem, not just a model flaw. Designers should shape feedback, behavior rules, and success metrics to reward accuracy over approval

AI: AI can remember more than ever. But memory isn’t continuity

Navid Borjian argues that better AI memory doesn't guarantee continuity in long-term projects: systems must understand which past decisions, constraints, and assumptions remain relevant and which should be reconsidered. He proposes a practical UX approach that makes AI's working context visible and editable, allowing users to inspect the reasoning behind decisions, correct outdated information, and maintain coherence across tasks and handoffs

Basics: UX designers are not designing interfaces, they design personalities

Bas Wallet argues that metrics don't just measure behavior — they shape how people perceive themselves. Through Fitbit, Uber, and Jira, he explores why designers must consider the human consequences of quantifying everything

@uxdigest

Peak-End Rule: Why People Remember the Worst Moment of Your App

People remember an experience largely by its most intense moment and how it ends. The author shows why fixing painful interactions and designing clear endings can matter more than polishing the entire journey

10 Rules for Creating User Flows

Dominic Rogers shares ten practical rules for creating clear, useful user flows. The key is to capture system responses, decision points, and edge cases—not just the happy path

NNG: Accordion Icons - Which Work Best?

Vitaly Friedman reviews NN/g research on which icons best communicate expandable menus. Carets prove the safest choice, while ambiguous arrows and split menus often confuse users

AI: AI-generated UI, What the AI Era Owes You, Amazing Glass, Jev for UX

UX Collective’s weekly roundup explores the difference between AI-generated interfaces and genuine UX design, the future of design careers, and the limitations of synthetic users in UX research. The selection also covers AI trust, ethical gamification, interactive interfaces, and practical resources for designers, including the Jev decision model

Opinion: UX history - the many phases of the (digital) button

From skeuomorphism to invisible controls, the article traces how digital buttons evolved alongside user expectations. With AI agents taking over tasks, buttons may become checkpoints for human approval rather than primary controls

Design: The Illusion of Transparency in Design and How to Overcome It

Designers often assume their interfaces are obvious because they know the product too well. A real user case shows why regular usability testing and contextual guidance matter, even for experienced users

Prototyping:

Why Reusable Patterns Matter More Than Reusable Components

Jason Han argues that reusable UI components alone cannot ensure a consistent user experience — design systems also need patterns that define behavior, context, user intent, and edge cases. Through examples of onboarding flows, dashboards, and usability studies with designers, he shows how documenting interaction patterns can reduce cognitive load, accelerate design work, and improve consistency across products

Dark Mode for Medium

Medium finally introduced dark mode after ten years of user requests. The design team reflects on how even valuable UX improvements get deprioritized—and why the AI era leaves fewer excuses for neglecting product quality

@uxhorn

Anchoring Bias: Why Your Users Judge Everything by What They See First

Lakshya Thakur explores how anchoring bias shapes user decisions through pricing, dashboards, and AI recommendations. Seven practical UX patterns show how reference points can improve clarity—or manipulate users

The Missing Strategy in Banking Transformation isn’t the App

UXDA argues that even excellent banking apps fail when customer journeys, communications, and internal operations remain fragmented. The Bank-as-a-Product approach proposes treating the entire bank as one coherent experience, not a collection of separate interfaces

NNG: Experience Design - Dropping the “User” from “User Experience”

NN/g argues that AI agents are expanding UX beyond direct human interaction with interfaces. Experience design must now address human-agent collaboration, control, and handoffs while keeping human goals at its center

AI: AI-generated UI, What the AI Era Owes You, Amazing Glass, Jev for UX

UX Collective’s weekly roundup explores the difference between AI-generated interfaces and genuine UX design, the future of design careers, and the limitations of synthetic users in UX research. The selection also covers AI trust, ethical gamification, interactive interfaces, and practical resources for designers, including the Jev decision model

Case Study: Willing to help, afraid to act - Pawse UX case study

Pranshi Verma shares a UX case study of an app concept connecting people who find injured street animals with local rescuers. User research and usability testing helped simplify emergency reporting, clarify actions, and provide reassuring status updates

Prototyping:

Redesigning Grindr: Exploring Trust, Authenticity and Safety Through UX Research

A team of five design students explores trust and safety in Grindr through user journey mapping, app review analysis, literature research, and three user interviews, turning their findings into a redesign focused on profile authenticity. The resulting prototype introduces identity verification and a behavioral trust score, while highlighting a critical UX challenge: making users feel safer without compromising privacy or mistaking verification for actual trustworthiness

Jakob’s Law - Why Users Want Your App to Work Like Every Other App

Afroza Riju explains Jakob’s Law: users expect new interfaces to behave like products they already know, and breaking familiar patterns can increase cognitive load and frustration. Using examples from Amazon, Snapchat, and YouTube, the article shows why designers should preserve established interaction conventions, introduce major changes gradually, and reserve innovation for improvements that genuinely benefit users

@uxdigest

Tesler’s Law: Why Complexity Never Disappears, It Just Moves

Afroza Riju explains why good UX shifts complexity from users to the system, using checkout forms, payments, and autofill as examples. Automation should reduce effort without hiding important decisions or taking away user control

NNG: The Role of UX in Agentic Software Factories

NN/g warns that AI-driven development risks reducing UX to documentation for agents, leaving user needs and usability testing overlooked. Researchers must help decide what to build, evaluate AI-generated products, and make research insights accessible to agents

Prototyping: Why Reusable Patterns Matter More Than Reusable Components

Jason Han argues that reusable UI components alone cannot ensure a consistent user experience — design systems also need patterns that define behavior, context, user intent, and edge cases. Through examples of onboarding flows, dashboards, and usability studies with designers, he shows how documenting interaction patterns can reduce cognitive load, accelerate design work, and improve consistency across products

AI: AI can remember more than ever. But memory isn’t continuity

Navid Borjian argues that better AI memory doesn't guarantee continuity in long-term projects: systems must understand which past decisions, constraints, and assumptions remain relevant and which should be reconsidered. He proposes a practical UX approach that makes AI's working context visible and editable, allowing users to inspect the reasoning behind decisions, correct outdated information, and maintain coherence across tasks and handoffs

Case Study: Mobile POS Usage Evaluation - A UX Research Case Study for an Outdoor Retail Company

Field research in two retail stores revealed why mobile checkout succeeded in one location but went unused in another. The key barriers were staffing, store layout, and workflows rather than interface usability alone

Experience: How to survive 25 years in UX design

Rob Gillham reflects on 25 years in UX, arguing that tools and trends change, but core design skills and user research remain essential. His biggest lesson: continuous learning and strong teams matter more than chasing AI hype or becoming a one-person design department

Opinion: Good Product Design Is About Decisions, Not Screens

Helena Borges explains why teams can meet their individual KPIs while the overall user experience still fails. AI accelerates interface production, but understanding user needs and aligning decisions across the entire journey remain essential

Interesting: Does Apple’s HomeView reduce effort, or ask for our attention?

Heenesh Patel questions whether Apple's new smart home screen solves real user problems or simply demands more attention. Drawing on Calm Technology principles, he argues that good automation should fade into the background

@uxdigest