The human brain is designed to make quick and effective decisions rather than stick to facts at all times. Instead of acting rationally, we prefer to act fast. This may lead to better outcomes indeed, but it might also lead you astray. Cognitive biases can be both a blessing and a curse.

In this article:

  • How both designers and users are prone to cognitive biases
  • What are the most common cognitive biases in design

Why cognitive biases matter for UX designers

It’s important to note that both the designers and the users are prone to cognitive biases. None of us are fully immune to heuristics and prejudices — and this is exactly why we should be aware of it. The same rules that affect how the users make decisions apply to UX designers.

For example, we’re all prone to framing:


Even when we know about the bias, we’re still prone to it!

From the designer’s perspective, the context, as well as our previous experiences, all affect design decisions. All these external factors make us focus on specific aspects of the issue — or ignore the other ones.

Kathryn Whitenton of the Nielsen Norman Group used a brilliant example to describe how this works for UX designers. Imagine that you’ve conducted a usability test with 20 users. The outcomes can be described in two different ways:

  • 20% of users couldn’t find the function on the website
  • 80% of users found the function on the website.

See the difference? The researchers from Nielsen Norman Group decided to test both versions in an online quiz. Here’s what happened: 39% of UX designers who saw the success rated voted for a redesign. In the case of respondents who saw the failure rate, 51% of them thought that the feature needs to be redesigned.

This shows how framing research results and statistics in a different manner may lead to significantly different design decisions. On the other hand, the framing bias can affect the users too. It’s mostly visible in the perception of the pricing, what is expensive, and what can be considered a bargain.

Below, how the most common cognitive biases apply to both the users and the UX designers.

Common cognitive biases in UX design

Anchoring bias

The anchoring bias(also known as the anchoring principle) is about relying on a single aspect and ignoring the other ones involved. Because people use it to make decisions, it’s also a judgment heuristic, exactly like framing.

How it works for the user: Anchoring bias can be hugely helpful in understanding user interfaces. It’s the bright side of this phenomenon — the user can stick to one clue, and therefore, learn to use the application faster.

Suggested values are also a good example here. Most non-profit websites, such as GoFundMe, don’t charge a set fee for their services. Instead, they provide a suggestion:


How it works for the UX researcher: Because of the primacy effect, the first piece of information we receive usually becomes the anchor. It’s good to keep that in mind for user testing. The respondents may prefer the first version they were introduced to simply because it was the first one, not because it was actually better. One of the ways to avoid this is through A/B testing — you can show different version to your users in a different order:


Wording bias

What is the best qualitative research method and why is it session recording?

This is a classic (a little exaggerated, yes) example of the wording bias, which is also known as response bias. It occurs when the question itself influences the answer.

How it works for the user: Imagine taking a survey, where the first question is: How difficult was it for you to use this feature? The question itself implies that using this particular feature is difficult, at least to some extent.

How it works for the UX researcher: The wording bias largely affects the validity of surveys. Here’s how different types of biased questions can affect research results:

  • The leading question, e.g. Do you prefer the previous or the improved version of the app? The word improved suggests the new version is better than the old one.
  • The double-barreled question, e.g. Are you enjoying the new inbox and text editor? What if the user is a fan of the inbox yet they’re skeptical about the editor? In this case, the respondent has to express the same opinion about both features.
  • The absolute question, which may not always work. Sometimes the yes/no answer misses the point. For instance, when you ask the user if they enjoy using the new feature, you might want to leave space for their comments and ideas. This way, you can gain much more in-depth insights.

Sunk cost fallacy

Again, this is an example of how past decisions influence current choices. Sunk cost fallacy means that if we already invested a lot of resources in something, we’re going to keep investing more. It happens because we don’t want our efforts to go to waste. This is what makes us finish bad movies — we’re not likely to drop it if we’re already halfway through, right?

The sunk cost fallacy goes in line with another phenomenon called loss aversion, which states that our brains consider all losses as more severe than gains. What’s more, some studies suggest that losses are perceived as psychologically twice as important as gains! For instance, losing $100 hurts more than gaining 100$ satisfies.

Same rules apply to user experience design.

How it works for the user: The sunk cost bias is often a part of the user flow, especially when it comes to signing up for a service, or making a purchase. This mechanism can be triggered by different design tricks, such as:

  • Progress bar — This solution encourages users to complete their actions, even if they require a certain commitment. Have a look at this sign-up form example from DataFeedWatch:


  • At this stage, the potential customer is most likely to drop out, as they need to provide payment details. The progress bar, however, indicates that it’s the final step of the process. It doesn’t make much sense to abandon the form now that you’ve gone so far, right?
  • Sunk rewards — Used in loyalty programs, sunk rewards encourage the user to stick to your product. It’s a bit like collecting stamps in a cafe so you can get the 10th coffee for free. When you’ve got just one or two purchases left to get something extra, you probably won’t resist the temptation.
  • Top-up suggestions — Have a look at this example from Revolut:


To use the service, you need to add money to your account first. But here’s the trick: you can’t add less than a certain amount (in this case, 20 PLN). This often means that you’re obliged to have more money in your account than you first needed. Because of that, you’re likely to use Revolut more often. What’s more, suggested top-up amounts right above the keyboard fulfil the same function. See the pattern?

How it works for the UX designer: You’ve probably already guessed it: we tend to stick to bad design because of the sunk cost fallacy. If you spent long hours working on a feature, you would want to stick to it even if it turns out to be a bad idea. One of the ways to avoid this is through agile development. Short sprints and iterations make it easier to apply changes along the way and avoid wasting resources.

Social desirability bias

This one’s about our tendency to seem likeable and be accepted. Social desirability bias (also known as the friendliness bias) encourages the user to answer in a way they think is expected by the researcher.

How it works for the user: What’s interesting, the social desirability bias typically appears unconsciously. Most of the time, the respondents don’t realize that they respond in a kinder, more favorable manner. For instance, if you ask “How do you like the new dashboard?”, the user is likely to select a higher rating, just to make you feel better about it.

How it works for the UX researcher: Now that you know how social desirability bias works, you might be wondering how to prevent it from happening. This bias can be avoided through indirect questions. Instead of asking how the user feels about a feature, you can ask “How do you think an average user would interact with this feature?”. This way, they won’t feel the urge to appear as nice and friendly.

Conclusion and cognitive biases reading list

The most important finding? We’re all prone to cognitive biases and the best we can do is to stay aware. Educating yourself about psychological principles that affect our minds is sure to pay off in the long run. It will help you understand user behavior, and it’s going to improve the quality of your work as a UX professional as well.

If you’d like to dig deeper:

Thinking, Fast and Slow by Daniel Kahneman
Biases and Heuristics by Henry Priest
Everyday Bias by Howard J. Ross
The Art of Thinking Clearly by Rolf Dobelli

Cognitive Bias in UX: the user side and the designer side

 There’s a great infographic with 100+ cognitive biases, all categorized and branched. It’s a great starting point to start looking for less-known, interesting cognitive biases. For example, start with a loss aversion bias and go deeper into adjacent ones, like IKEA effect, zero-risk bias, escalation of commitment or pseudocertainty effect.
Full-size image.
List of all the cognitive biases.