Feed

A UX teardown of Walmart's signed-out deals flow: a savings hub and an under-$50 tech filter that lead to a $39.99 game with no reference price, a third-party seller named at the bottom of the buy box, and review filters that take 150 ratings down to three verified four-star reviews.

What the business model canvas is, its nine blocks and where a product team finds the evidence for each, how to trace a product change across the canvas, and how it relates to the Lean Canvas and the Value Proposition Canvas.

A UX teardown of Prime Video's signed-out browse flow: rankings and Most Liked labels get a white tag on every tile, while the cost of watching gets one small yellow bag that means Prime, Paramount+ or a rental, and a Free to me filter that comes back empty without saying why.

Maslow's hierarchy of needs as he actually wrote it in 1943, where the pyramid came from (not Maslow), what later research kept and dropped, and how product teams use the model: Aarron Walter's hierarchy of user needs, a level-by-level product review, and the motivation models that fit product work better.

A UX teardown of HBO Max's signed-out browse flow in Brazil: the catalog is easy to evaluate without an account, but the 'plans from R$22.90/month' line on both catalog heroes is the twelve-installment annual price, and the monthly price only appears after a toggle at the bottom of a title page.

A UX teardown of The Guardian's signed-out reading flow: kicker colors that follow the type of story, a $12 reader-support ask above every page we opened, and a header where the only visible 'Search' is a jobs link.

Functional fixedness is the habit of seeing an object only as the thing it was introduced as. Where it came from (Duncker's candle problem), how it hides features from the people using a product, how it narrows what a design team sketches, and the one technique with evidence for loosening it.

A UX teardown of the signed-out path from FigJam's page into Figma's template gallery and Wireframe Kits page: twelve menu entries with no product labels, three different start buttons on one page, and ten screens in which the word FigJam never appears.

A cross sectional study measures many people once; a longitudinal study measures the same people more than once. Which one you need depends on whether your question is about how things are or how they change — with UX examples, the traps in each, and a decision rule.

A confounding variable is a third factor tied to both the thing you changed and the result you measured, so it distorts the difference you see between groups. Where confounders hide in UX experiments, launches and usability tests, and the designs that remove them.

How the WCAG contrast ratio is calculated, which thresholds apply to text, large text and UI components at AA, and the cases a color contrast checker can't see on its own: transparency, text on images, hover states and dark mode.

Automotive UX for in-car interfaces: glance budgets, HMI patterns that hold up while driving, and the research methods used to measure driver distraction.

A UX teardown of The Atlantic's signed-out reading flow: section fronts that open below a fixed advertising band, a 665-pixel reading column, and a subscription gate that offers two equal red buttons without a price in view.

The gestalt principles describe how people group what they see: proximity, similarity, common region, closure, continuity, figure-ground, common fate. On a real screen several apply at once, and the layout bugs come from two of them giving different answers.

WCAG 2.2 added nine success criteria and removed one. What each new criterion asks for, which ones count toward AA, what stayed the same on contrast, and how to update a 2.1 audit.

Visual design principles are hierarchy, contrast, proximity, alignment, repetition, scale, and whitespace. For a product team they are most useful as checks you can run on a screenshot in a review, so a disagreement about a screen ends with a measurement instead of a vote.

What a SWOT analysis is, how to run one for a product decision rather than a whole company, a worked example for a feature bet, the TOWS step that turns the grid into options, and where the method misleads.

A mood board is a collection of images, color, type, and texture used to agree on how a product should look and feel before anyone designs screens. The five decisions it can settle, what it can't, how it differs from a style tile, and how to finish one with a written decision.

A UX teardown of Duolingo's signed-out web path from course catalog to Super: a research hub with no sign-up button, a plan table with no prices, and a trial timeline that names the day you're charged but not the amount.

A UX teardown of DoorDash's signed-out web order flow, from the city directory to the pickup cart: an address that returns a launching-soon message in Manhattan, a pickup map that finds the restaurants delivery couldn't, and a cart that puts Continue above the order it confirms.

How to start a design system in a product that already ships: what to audit first, which tokens and components come before the rest, how much documentation you need on day one, and which tools to pick.

A UX teardown of ChatGPT's logged-out web chat: a composer you can use with no account, a plus menu that marks locked tools where you look for them, and a ChatGPT dropdown that opens a login card instead of options.

A UX teardown of Best Buy's signed-out headphone search: facets that cut 449 results to 6 in two clicks, a store location the shopper never chose, a shortlist where every visible card is unavailable, and one Add to cart button on the page, for a $399.99 sponsored pair.

A UX teardown of 1Password's signed-out path from Features to pricing: one tab bar that switches between a per-month promo, a flat price set beside a per-user one, and a page with no prices at all, with the billing toggle renamed along the way.

Retention and coercion move the same number. A test that tells them apart, the mechanisms that pass it, the ones that fail, and how to measure retention so the difference shows up before your support queue finds it.

A UX teardown of TechCrunch's read flow: one switch turns the feed into a dense scan queue and carries the preference into the next section — then removes the timestamp and the content type the returning reader came for.

Sentence case has become the default across Microsoft, GOV.UK and Material 3, while Apple still capitalises its buttons. A look at what each convention specifies, why title case is expensive to enforce, and how to write the rule so a team stops re-deciding it.

A UX teardown of Reuters' signed-out read flow: dependable section fronts, an article page that gives the reader its figures early, and a search that ranks 14,934 matches by the clock, then leaves a Business story at the top of a Technology filter.

A UX teardown of Mobbin's signed-out browse flow: a platform switcher that leaves the model alone, a topic that gets three names in three hops, and a flow gallery that hides the one annotation worth paying for.

A UX teardown of Magic Patterns' signed-out evaluation flow: a hero prototype drawn in grey boxes, the fullest product description filed inside a closed FAQ, plans priced in credits, and a design-system promise that reappears on the Enterprise card.

Password, magic link, one-time code and federated sign-in are four different bets about what your users can remember. A walk through the four login page patterns, the field-level decisions inside each, and the WCAG 2.2 rule that now rules some of them out.

How to run a heuristic evaluation that produces fixes instead of a checklist: independent passes before any group discussion, severity rated on what a problem costs, and the three habits that turn the method into theatre.

A UX teardown of Axios' signed-out read flow: search results priced to the minute, section fronts that give their lead story nothing but a headline and a face, and a newsletter gate that states its price out loud.

A WCAG scan can come back clean on a flow nobody can finish with a keyboard. Seven accessibility heuristics for the barriers success criteria pass over, and how to run them as a review.

What a control group is, how it differs from the experimental group and from a control variable, and how to set one up for a UX experiment or A/B test without quietly breaking it.

Voice user interface design without the hype: which tasks suit speech, how to write prompts people can hold in their heads, how to confirm and recover from errors, and what to measure once it ships.

A UX teardown of Stripe's signed-out documentation flow: a sandbox key in the hero, an AI block that buries the code example, a sample that breaks its own rule on line 5, and an API reference that changes every rule without saying so.

How stratified sampling works in UX surveys: choosing strata, proportional vs disproportionate allocation, weighting results back, and when quota sampling is the honest name for what you did.

A quick links module is a shortcut to the handful of destinations people actually came for. How to choose what goes in it, how many slots to give it, where it sits, and when to take it away.

The mobile app design patterns that hold up in daily use: bottom navigation, thumb-sized targets, bottom sheets, honest loading states, permission timing and forms that respect a phone keyboard.

A UX teardown of GitHub's signed-out repository browse flow: an exact repository name that still returns 4.8k results, a latest-commit row at every level of the tree, a folder listed twice, and an edit button on a generated file.

A UX teardown of Etsy's signed-out search flow: one chip row that holds applied and suggested filters side by side, a result count that doesn't move when you filter, and a five-star card for a mug nobody has reviewed yet, sitting second under Top reviews.

A UX teardown of Airbnb's signed-out search flow: a filter button that counts homes before you commit, one all-fees price carried from histogram to Reserve, hotels at the top of a Homes search, and a header that forgets the trip on the listing page.

Fidelity isn't a quality ladder you climb. What a wireframe is supposed to settle, what lo-fi, mid-fi, and hi-fi each buy you, when a mood board is the right artifact instead, and the failure mode that costs teams a sprint.

A UX teardown of WIRED's category pages: the same template runs Business, Science and Reviews, and what each one puts in its first rail gives away what the section is actually for.

A UX teardown of Webflow's public pages: the homepage argues entirely in revenue, the product page won't let you self-serve, the marketplace shows no price at all. Then the pricing page hides its real variable in a two-word footnote.

A UX teardown of The Washington Post's signed-out reading flow: the subscription card publishes both prices and the renewal terms, then centres itself over three lines of the headline it's selling.

A UX teardown of Vercel's site: the homepage now sells 'Agentic Infrastructure' while the product pages underneath still sell people working together. What happens when the top of a funnel changes audience and the middle doesn't.

A method is the thing you ran. A methodology is the chain of reasoning that makes its output count as evidence. What the three research methodology families actually commit you to, the five decisions every methodology makes, and where they fail quietly.

Most personas are a stock photo and a fake quote. What personas are actually for, which attributes earn their place, how to derive them from behavioral axes instead of demographics, and why a set without a designated primary decides nothing.