67 cognitive biases people fall for when thinking and deciding. Understanding them helps you make clearer judgments and design more responsibly.
Content adapted from the ByteDance Lark Design Team「Cognitive Biases Handbook」.
No matching biases. Try another keyword.
We tend to avoid the unknown — when deciding, people steer away from options with insufficient information. Adding clear details minimizes ambiguity and improves conversion.
Example: A CTA button should carry an informative label beside it to clear up any uncertainty.
When making decisions, people over-rely on the first piece of information they receive (called the anchor), even when it is irrelevant to the decision. People tend to use this fragment of information (the anchor) to reach a decision quickly.
Example: A bias often exploited in shopping scenarios — a before/after price contrast amplifies the sense of a discount and nudges users toward the decision we want.
Introducing a third option to strengthen the appeal of an existing one.
Example: Suppose there are products A and B. Here are the options a customer can choose:；· $100 — A；· $300 — B；· $300 — A + B；Here, option two is the "decoy," used to steer customers toward the "target option," three.
When comparison options are present, we become sensitive to tiny differences that are actually not that significant.
Example: Show your product's advantages next to a competitor so users notice the difference, even a small one.
The framing effect means that, facing the same problem, people choose whichever description sounds more favorable or agreeable at first.
Example: In most cases, a positive framing (e.g. the glass is half full) yields a higher conversion rate.
A bias caused by measurement error, sometimes also called observation bias or misclassification.
Example: When showing a product or service (on a website or in an app), be sure to include photos with detailed captions. The more information users get from the description, the more reassured they feel.
We perceive a product's value based on how it looks or how it is served. As people say: it's all in the packaging!
Example: Design matters to a product's success more than you think. Excess whitespace, the wrong border color, and misaligned text all affect conversion — make UI design a priority.
Also called choice overload. When there are too many options, our brain freezes and it becomes hard to choose.
Example: Too many choices = low conversion.
Compared with things they don't own, users tend to hold on to what they already have. They tend to overvalue their possessions while ignoring objective market value.
Example: The endowment effect is a form of loss aversion, and a free trial is its most common use. Once a user starts using a product and invests time in it, it becomes hard to let go when the trial ends — when they are about to leave, they feel they are about to lose many good things.
Also called present bias: people prefer immediate rewards over larger future ones.
Example: Offer a small discount (or free shipping) for an immediate purchase, rather than promising a bigger discount on a future purchase.
The IKEA effect: people tend to highly value products they took part in creating. The more labor (and emotion) a person invests in an item, the more they overvalue it.
Example: The key to using the IKEA effect for stickiness is to create low-effort, high-reward, high-contribution tasks where users can contribute their own value while still completing the task. In learning scenarios we can use it to keep users coming back and sticking with their studies.
The more we invest in something, the harder it is to give up. As a result, we often keep going with a failing course of action simply because we have already poured in too much time, money, or effort.
Example: When a gamer pays a $5 monthly membership fee, many force themselves to keep playing because they have already invested so much time and money in the virtual game.
When facing equal amounts of gain and loss, people find the loss harder to bear. The negative utility of a loss is about 2 to 2.5 times the positive utility of an equivalent gain. Loss aversion shows that risk preferences are not consistent: when gains are involved people are risk-averse, while when losses are involved they become risk-seeking.
Example: Limit the time on a special offer to create urgency: "This offer ends in X hours."
When we are under pressure, we fail to consider the probability that a risk will actually occur. As a result, smaller risks may be overestimated or ignored.
Example: In a conversion funnel, the smallest uncertainty can lead users to distrust the product and stop using it. Make sure all details are clear and shown upfront, especially anything involving money — total cost, discounts (if any), extra fees, and so on.
Also called the Peltzman effect. When we sense greater risk we become more cautious, and when we are more protected we become less careful.
So once we know many safety measures are in place, we are more likely to act in riskier ways.
Example: Make users feel as safe as possible before purchase or signup, giving them confidence and trust.
We love certainty, even when it is counterproductive. We prefer to completely eliminate one specific risk even when other options would reduce overall risk more. One example: deliberately reducing risk in your own area of responsibility at the cost of raising risk for the whole organization.
Example: Offer a money-back guarantee and a risk-free trial to lower the perceived risk and make users feel safe.
We raise the value of scarce items and lower the value of highly available ones.
FOMO (fear of missing out) makes us more susceptible to temptation and impulse, prompting rash decisions.
Example: Use phrases like "limited-time offer" and "limited quantity" to describe your product or service. Give the impression that many people are viewing this item and could grab the last few at any moment.
We cannot easily change people's beliefs: evidence and arguments that contradict a deeply held belief tend to backfire and only reinforce it.
Example: Influence others' behavior through emotional appeals — don't try to change their beliefs head-on.
We are more inclined to accept arguments that align with our prior knowledge, and to reject counterarguments against them.
Example: When describing your product's benefits, don't overstate them. If it sounds too good to be true, people won't believe it.
After making a decision, we tend to praise the option we chose and downgrade the alternatives.
Example: Right after a user makes a purchase and completes the transaction is a great moment to invite them to share the product (or leave a review); show an affirming message and congratulate them on reaching this step to motivate them.
When people already hold a view, their perception and attention to it are amplified, and they selectively recall or gather instances that support it. People accept ideas they already believe more readily, while setting opposing views aside.
Example: In user research, if your preconception is that users prefer design A over B, you may pay more attention to the upsides of A that users mention and collect more positive feedback on A. When a user says they like A, you think "just as I expected." This bias makes you miss a lot of other information.
When we are an expert in a field, we unconsciously assume others share our level of understanding — especially with jargon.
Example: Consider most users: their technical level may be lower than you imagine, and they may be far less familiar with things you take for granted.
When solving a problem, an individual often sees only the usual function of an object and misses the other functions it could serve. It is a fixed view of what something is for, formed over a long time.
Example: When a product challenges existing conventions of use, it can create usability hurdles. Be sure to run usability testing and try to address these potential issues in advance.
Hindsight bias traps people in an "I knew it all along" feeling, preventing them from drawing useful lessons from events and from judging facts and others fairly; subjectively, it is easy to selectively ignore many objective facts.
Example: Hindsight feels good but undermines the correctness and fairness of decisions at work, creating hidden risk. The cure is to write down your thoughts before knowing the outcome, then verify against the result and track the data.
We tend to overestimate the duration or intensity of our future emotional states.
Example: Depict the problems a user would face without your product or service, then introduce how your product solves them.
Compared with others, we often overestimate our own odds of success.
Example: Make sure the product hides no unpleasant features that could break the optimism bias (e.g. extra fees, possible delays).
We deliberately avoid negative information (or feedback inconsistent with our expectations), believing that if we bury our head in the sand, it will go away.
Example: Work side by side with customer support and proactively see what confuses users — there is a lot to learn from handling complaints.
We tend to underestimate the time needed to complete a task. The planning fallacy usually leads to time overruns, cost overruns, or insufficient returns. Breaking a large task into smaller parts helps address this.
Example: Multiply your initial estimate by two — no, actually by three. Not because you are lazy, but because your estimate is probably wrong.
People often overestimate their ability to control impulsive behavior.
Example: We all think clickbait is a cheap trick, yet we still fall for it, don't we?
When weighing all possible outcomes, we tend to focus on the ones that look plausible and familiar, ignoring the rest.
Example: Have you ever shipped a feature you expected to cause an unintended consequence, and then fixated only on that one consequence?
We assume that things which spring to mind quickly are more common and important than things that are harder to recall. As a result, recent, frequent, extreme, or memorable events carry more weight than most information.
Example: Make your design memorable by creating something distinctive (without hurting consistency and familiarity).
Also called the Forer effect. People give high accuracy ratings to personality descriptions they believe were tailored to them, even though those descriptions are vague and general enough to apply to almost anyone.
Example: In product copy, avoid words like "everyone" or "users"; words like "you" and "your" feel closer. When reading normative messages, people have a selective sense of self-insertion that unconsciously draws them to what fits their own experience, producing a feeling of "that's me too."
Once something first catches our attention, we soon notice related information in many places. For example, when someone wants to buy a new car, they suddenly see that car everywhere.
Example: When running a campaign, expose visitors to the same content across different marketing channels, so it feels like your product is everywhere, deepening the impression of this new product.
People develop a liking for something simply because they are familiar with it. The more frequently an external stimulus is presented (the more chances an individual has to encounter it), the more they tend to like it.
Example: Stick with UI concepts, behaviors, terms, symbols, and icons your users already know. Keep marketing materials, the website, and the product consistent to optimize the funnel, and align wording with industry terminology so your users feel comfortable.
Compared with neutral or positive things, things with a negative impact affect people psychologically more strongly.
Example: Demonstrate your product's value by solving a negative experience. If you want your story to have greater impact, try showing more emotionally negative content.
People more readily treat rhyming, rhythmic sentences as true and accurate. Once a statement has rhythm, it becomes more persuasive.
Example: Most ad slogans are catchy and sticky.；So make your product copy as rhythmic as possible — it persuades users more easily.
Our view of things is heavily shaped by our expectations, and when people concentrate, they filter things out of their environment.
Example: A product's conversion funnel doesn't start when a user logs in — it starts the first time they see your ad. Inconsistency between the marketing message and what the product delivers disappoints users and lowers conversion. Consistent messaging across all media (every stage of the funnel) sets the right expectations and improves conversion.
We assume things that are processed faster, more smoothly, and more fluently carry higher value. Sometimes an illogical argument can still win when it is communicated well (by someone with authority and experience).
Example: · Make things convenient for users (fast, simple, easy-to-understand navigation)；· Make content easy to scan (images, legible fonts)；· Create "mental shortcuts" that draw users in and make your product their first choice；· Provide meaningful default options, since users assume you have their best interests in mind and will often go with the default；· Add powerful cross-product search to simplify finding specific topics
A simpler way to put it: repeat a lie often enough and it becomes truth — the more publicly something is discussed, the more people believe it.
Example: Create a striking slogan and make sure it is repeated many times. If possible, ending it with a rhyme works even better.
We often change our own views simply because most people think the same way.
Example: In design we can create a sense of group choice to attract more users — showing purchase counts and a live feed of recent purchases signals popularity, nudging hesitant users to follow the crowd and buy.
We tend to feel for a specific identifiable person rather than an anonymous one.
Example: When telling your product's story, use an individual case rather than a general statement.
Reactance occurs when we feel that someone (or something) is trying to remove or limit our choices. When it happens, we have an urge to resist and do the opposite.
Example: Be careful when you "argue" with a user about their choice. Persuasion must be gentle and gradual — never confront a user head-on.
Actors tend to attribute their own behavior to situational factors, while observers tend to attribute the same behavior to the actor's stable personality traits.
Example: We explain our own behavior with external, uncontrollable situational causes — e.g. being late to work because of an accident on the road. But as observers explaining others' behavior, we use internal, stable personality causes — e.g. he was late because he's lazy and has no sense of time.
We fail to recognize our own lack of competence, and because we cannot distinguish high ability from low, we assume everyone is alike.
Example: · Low-ability people usually overestimate their own skill level;；· Low-ability people cannot correctly recognize the skill level of others who are genuinely competent;；· Low-ability people fail to recognize and face their own shortcomings, and how extreme those shortcomings are;；· If low-ability people are properly trained to greatly improve their skills, they eventually recognize and admit how incompetent they were before.
People tend to overestimate how much others agree with them, imagining that others are like us and share our views, beliefs, preferences, values, and habits — assuming everyone thinks the same way.
Example: Without ever being involved in politics (or a particular country), were you recently surprised by your country's election results?
When explaining others' behavior, we attribute it to inner traits (he must have that kind of personality to act that way) rather than the external situation (maybe he was forced by circumstances, or there are unwritten rules in that setting). The fundamental attribution error is a common source of conflict and problems, and this bias exists in users and designers alike.
Example: Designers need to clearly distinguish between "my view" and "my behavior," and to genuinely step out of their own position, role, and function to review the whole situation. First accept the full reality, including others' thoughts and views, without judging right or wrong, and understand that things change dynamically. Stop judging from a single angle — any one perspective inevitably carries a stance, and it is only one facet of many.
The halo effect means using one quality of a person (or thing) to make an overall judgment about them. In other words, our first impression of a person, product, company, or brand shapes how we interpret all of its characteristics.
Example: Make full use of this heuristic with authoritative photos and trust-building visuals.
Ingroup bias is often used to steer user decisions: people cognitively favor the group they belong to. The key to using it in experience design is to build a sense of group belonging and let the power of a small group influence user decisions.
Example: For example, Booking added a button to filter reviews by language. The intent may have been to help users understand reviews better, but in practice users trust reviews within their selected tag more, since a shared language often means a similar country or culture. Users find a small group this way, are influenced by it, and make decisions closer to that group's.
We assume that advice given by an authority figure is more accurate, and we are more easily influenced by it (even when the subject has nothing to do with that person's authority).
Example: Build authority by promoting notable endorsements of your product:；· Reach out to influential people and let them use your product or service for free；· Highlight well-known customers or brands；· Show testimonials from public figures；· Use authority figures (e.g. doctors, professors) in product demos and presentations
The foot-in-the-door technique is a persuasion method that gets a user to agree to a larger request by first making a simple, small request.
Example: · Build a relationship with users by offering a free trial.；· Don't frustrate users with an overly complex onboarding flow.；· Break large chunks of content into smaller, easy-to-handle pieces to keep users happy and engaged.
We tend to express more of what conforms to social expectations and hide what doesn't. For example, when asked about bad habits or number of sexual partners, people consciously or unconsciously conceal or downplay the truth.
Example: Social proof is a great way to build trust. Here are some examples you can use on a product page:；Instead of saying "we're great because…," let others vouch for you.；· Logos of well-known customers；· Logos of well-known partners；· Genuine reviews and testimonials from customers (from your target audience)；· Cumulative numbers (customer count, transaction count, any valuable figure)；· Media mentions and citations；· Awards and certifications from respected organizations；· Certificates；· Links to data-backed case studies
We easily forget information that can be readily found online.
Example: Do you still remember your best friend's or your child's phone number?
We remember information better when it is perceived as fun or humorous. This can help improve a product's conversion rate and contribute to overall business growth.
Example: Don't use humor when users may feel frustrated. For example, if your app isn't compatible with older devices, don't say so in a funny way — users may be upset and unable to appreciate the humor.
The more often something is repeated, the more we believe it.
Example: Repeat your message (the main benefit, the main difference) again and again across ads, your website, the App Store page, the landing page, newsletters, and more.
Users don't evaluate an experience by its average or by the sum of all moments. Instead, their brain heavily weights the peak (high or low) and the end of the experience. A pleasant peak usually corresponds to a memorable moment of delight in the user journey.
Example: Keep your competitive edge and raise the product's peak by making its best features even better. And don't ignore the "post-purchase" experience — make sure the experience ends on a note of refined taste.
Pictures and images are easier to remember than a thousand words.
Example: Always include images in your content — great visuals improve conversion.
Also called the recency and serial-position effect. Items at the beginning and end of a list are remembered more easily than those in the middle.
Example: If you plan to show a long list of information, make sure the first item is the most important one.
When a lot of information is presented together, the items that stand out and are specially emphasized are remembered more easily than the rest.
Example: Make sure CTA buttons stand out in a different style, size, color, and position to serve different purposes.
An ongoing task creates task-specific tension. Completing the task relieves this tension, but if the task is interrupted, the tension remains. This tension makes the related information easier to access and remember.
Example: Show an incomplete progress bar to encourage users to reach their goal.
Over-focusing on the people or things that survived, while ignoring those that were filtered out (possibly because they can't be observed), leads to mistaken conclusions.
Example: In needs analysis and research, ignoring survivorship bias means the data you collect may reflect only a small subset of users — turning false or niche needs into the main basis for design, and dooming the product from the start.
We tend to ignore general information and focus only on the specific case.
Example: Don't just share cold product facts. Instead, show reviews and use cases from other relevant users or companies. To be more persuasive, combine your quantitative data with a few personal stories — back your point reasonably with numbers, while concrete examples carry the message emotionally.
We wrongly believe that if something happens frequently in a given period, it becomes less likely to happen in the future.
Example: Stick to facts, rely less on intuition, and rely more on data.
Wrongly believing that someone who has experienced success is more likely to keep succeeding in other attempts. Typically, when someone has recently done something successfully, we assume they will keep succeeding at the same thing — ignoring that the success may have been random or coincidental.
Example: Build a strong brand by highlighting a series of success stories.
We often ignore sample size and rush to conclusions, even when the sample has not reached a statistically sufficient amount.
Example: Talking with customers matters, but don't build product assumptions on just a few interviews. Use plenty of data and make product decisions based on real data rather than assumptions.
Certain types of users are unintentionally excluded from the pool of research participants. It can also be understood as generalizing from a part — explaining the whole from local cases.
Example: You're designing a bike-tracking app and need to research cyclists. You decide to interview and observe cyclists in a major city, but you don't realize their riding habits may differ greatly from cyclists living in hilly suburbs. So your findings risk not applying to all of your target audience.
Researchers may sometimes expect a certain result and unconsciously manipulate the experimental process, or misinterpret the results, severely distorting the findings.
Example: In user research, a researcher's expressions and body language can reveal the result they expect. If users pick up on this, they may respond in ways that cater to the researcher's expectations.
When a reward comes unexpectedly, our joy at receiving it increases.
Example: Give users daily deals, free bonuses, badges, prizes, and more. The more "actions" you create each day, the more users feel connected to the product.