Free, 16 biases, sorted by who has them
Cognitive biases in UX
Most lists of biases are alphabetical. This one is sorted by who has the bias, because that decides what you do about it. Your own biases you check with research. Your users’ biases you design with, and the only question is whether you do it in their interest.
New to the idea? Start with Types of Cognitive Bias for UX Designers and Marketers.
Part 1 Biases in your own judgement
These bend the research and the decisions before anything ships. You cannot switch them off, but you can check for them, and that checking is most of what user research is for.
-
Survivorship bias
You only hear from the people who stayed, so the answers sound happier than the truth.
Practise it: Lesson 4.4, asking who your data leaves out before you trust it.
-
Curse of knowledge
You forget what it is like not to know your own product, so the labels feel obvious to you and lose everyone else.
Practise it: Lesson 2.1, testing whether users understand what you take for granted.
-
Hindsight bias
Once the result is in, you remember having expected it, so the research seems to have taught you nothing.
Practise it: Lesson 4.5, writing down what would prove you wrong before the results come in.
-
Confirmation bias
You notice the interview quote that agrees with you and skim past the three that do not.
Practise it: Lesson 2.1, turning assumptions into questions you can test.
-
False consensus effect
You assume users think the way you do, so the persona ends up describing you.
Practise it: Lesson 4.2, building a persona from real behaviour instead of your own habits.
-
Selection bias
Your data only holds the people who stayed long enough to be counted, and it reads as everyone.
Practise it: Lesson 4.4, checking who your data leaves out before you trust it.
-
Bandwagon effect
A pattern gets chosen because every competitor uses it, not because your users need it.
Practise it: Lesson 1.2, moving design decisions from opinion to evidence.
Part 2 Biases in the people using your product
These shape how people read a screen and make a choice. Design works with them whether it means to or not, so the question is whether it does so in the user’s interest.
-
Peak-end rule
People judge an experience by its worst moment and its ending, not its average, so one bad step or a weak last screen can outweigh the rest.
-
Choice overload
Too many options can make a choice harder, but the evidence is mixed. Ease the choosing instead of only cutting the choices.
-
Loss aversion
A loss weighs more than an equal gain, so warnings about what you could lose pull hard. The effect varies, so use it to inform and not to pressure.
-
Anchoring bias
The first number a person sees sets the scale for every number after it, which is why pricing pages lead with the big plan.
-
Framing effect
The same fact, put as a gain or as a loss, gets a different answer.
-
Conformity bias
People follow what others around them appear to do, so "most teams choose this" moves more clicks than a feature list.
-
Aesthetic-usability effect
Attractive screens are judged easier to use, and forgiven for problems a plain one would be blamed for.
Practise it: Lesson 1.2, letting aesthetics support user goals instead of overriding them.
-
Default effect
People keep whatever is already selected, so the starting box decides the result for most of them.
-
Cognitive dissonance
Holding two conflicting beliefs is uncomfortable, and people change a belief or a choice to end the discomfort.
Practise the first part
Knowing a bias exists does not stop it. Checking for it does. The free UX Foundation Course has four chapters, and these checks happen all through it: evidence over opinion, testable assumptions, personas from real behaviour, and data that shows who it leaves out.
Start the courseFrequently asked questions
What is a cognitive bias in UX?
A cognitive bias is a predictable shortcut in how people judge and decide. In UX it matters twice: designers carry biases into their research and decisions, and users carry biases into how they read a screen and make a choice.
Which cognitive biases matter most for UX research?
Confirmation bias, the false consensus effect, selection bias, survivorship bias and the curse of knowledge. They make research agree with what you already believed, make the persona describe you, make the people who answered look like everyone, hide the people who left, and make your own labels look clear when they are not.
Is it ethical to design with cognitive biases?
It depends on whose interest it serves. Using anchoring to show a fair price in context helps people decide. Using it to make a bad deal look good is a dark pattern. The bias is the same; the intent is what differs.
Where can I practise avoiding these biases?
In the free UX Foundation Course. Lesson 2.1 finds the beliefs behind an idea, lesson 4.2 builds a persona from real behaviour, and lesson 4.4 checks who your data leaves out.
More chapters
A course chapter on these biases
Chapter 5 of the free UX Foundation Course will cover these biases. It will show you how to check for your own, and how to design fairly with your users'. It is planned but not written yet, and there is no date. Leave your email and you will get one message when it is out. Nothing else.