The experience
Picture a checkout page with a ticking timer. A badge murmuring that stock is nearly gone. A subscription that took one tap to start and now seems to have no exit at all. If those examples feel familiar, that is rather the point: design tricks that nudge you into buying more or sharing more than you meant to are common enough that you have almost certainly met them: fake urgency timers, "only 2 left!" messages, or subscriptions that are easy to start and hard to stop. When researchers crawled about 11,000 shopping websites, they found 1,818 instances of these "dark patterns" on roughly 11% of the sites, and the more popular a site was, the more likely it was to use them.Supported, with limits The walkthrough names each trick through generic examples, so you can study its shape without any real shop in the room.
What is happening
The tricks have a name and a family tree. The term "dark patterns" was introduced in 2010 by UX specialist Harry Brignull, who began cataloguing them at darkpatterns.org (now renamed deceptive.design, with the preferred term "deceptive patterns"). Researchers later sorted practitioner-collected examples into five recurring strategies you can learn to spot: nagging (repeated interruptions), obstruction (making a task needlessly hard, like cancelling), sneaking (hiding relevant information or costs), interface interference (visually privileging the choice the company wants), and forced action (making you do something extra to get what you came for).Historical / descriptive Those five strategies are the spine of the walkthrough: the five sketches above are those five strategies, one apiece, each wearing a different costume.
Naming things has consequences beyond tidiness. Regulators on both sides of the Atlantic now treat many of these designs as unlawful rather than merely annoying. A 2022 US Federal Trade Commission staff report, "Bringing Dark Patterns to Light," details four common tactics — ads disguised as independent content, hard-to-cancel subscriptions, buried terms and junk fees, and tricks that steer people into giving up personal data — and the EU's Digital Services Act (Article 25) prohibits online platforms from designing interfaces that deceive or manipulate users or materially distort their ability to make free and informed decisions.Historical / descriptive
What this shows
It would be comforting to believe these designs merely annoy — that a fake timer fools no one and a pre-weighted button changes nothing. The evidence says otherwise. These designs are not just irritating — controlled experiments show they measurably change what people do. In a randomized experiment with a census-weighted US sample of about 1,900 adults, only 11.3% of a control group accepted a dubious paid identity-theft-protection service, versus 25.8% exposed to mild dark patterns and 41.9% exposed to aggressive ones; hiding information and trick questions were among the most effective single techniques, and less-educated participants were significantly more susceptible to the mild versions.Supported, with limits The techniques that did the most work in those experiments — hidden information, trick wording — are precisely the ones the explanation slows down and names, so you can inspect them at your own pace.
What this does not show
None of this means every persuasive design is manipulation, or that these tricks work reliably on everyone. Researchers themselves note the umbrella term lumps together ethically quite different practices, prevalence figures depend heavily on the definitions and detection methods used, and experiments find that heavy-handed dark patterns can backfire — in one study the aggressive versions provoked visible anger and far more people simply quitting (65 dropouts versus 9 in the mild condition), which is partly why the subtler patterns are considered the more insidious problem.Supported, with limits Two further things this entry does not claim. The research gathered here does not tell us how much recognition practice actually protects anyone, so treat the walkthrough as a vocabulary lesson, not a vaccine. And nothing on this page diagnoses any particular real interface as manipulative — the sketches are abstractions, and judging a live product is a job for regulators and courts, not for a walkthrough.
Research and limitations
The evidence behind this entry comes from a young field — the name itself only dates to 2010 — and rests on two main instruments, both introduced above: large automated crawls that count patterns in the wild, and randomized experiments that measure whether patterns change behaviour. Each inherits the limits already flagged: crawl-based prevalence figures move with the definitions and detectors chosen, and the headline experiments used US samples facing one dubious paid service, so how far those exact numbers travel — across countries, product categories and years — is genuinely open. The regulatory picture is a moving target too: the documents cited here describe the rules as written, not how enforcement and courts will ultimately read them. Where the research is silent — on long-term effects, or on whether learning to spot these patterns durably changes what people do — this entry stays silent as well.