Cognitive biases shape how users experience our products long before we've had a chance to explain them.

They operate below the level of conscious decision-making, which means designing without accounting for them is not neutral - it's designing with your eyes closed to half of what determines the outcome.

This piece covers seven more biases worth keeping close; each one touches a different corner of the design process.

Noble edge effect

The noble edge effect describes something most designers already suspect but rarely make explicit: users prefer products that demonstrate social responsibility.

When a brand signals genuine care for something beyond its own commercial success - environmental commitments, community investment, ethical sourcing - users extend it a degree of trust and goodwill that a purely functional product rarely earns.

The irony the original research keeps surfacing is that the businesses most likely to chase short-term revenue at the expense of this signal tend to underperform against the ones that don't; the noble edge effect is one of the few places in commercial design where doing the right thing and doing the smart thing point in exactly the same direction.

For us as designers, the implication is straightforward: where we have influence over how a product communicates its values, we should be making that communication visible, specific, and credible. Vague altruism reads as marketing; concrete commitment reads as character.

Default bias

People tend not to change what has already been set for them, and the default bias captures this tendency precisely.

Once a behaviour or setting is established as the starting point, the effort required to change it - even small effort - is often enough to prevent any change at all.

The design implication cuts both ways. Set a harmful default and many users will never correct it, not because they want that outcome but because inertia carries them there; set a helpful one and the same inertia becomes an ally.

This is why the default option in a product is never neutral: it is, in effect, the choice we have made for the users who won't choose for themselves - and that is a significant portion of any audience. Making the default genuinely serve the user, rather than serve the product's metrics at the user's expense, is one of the quieter ethical responsibilities in our work.

Self-serving bias

We overestimate our own contribution to success and underestimate our contribution to failure; external successes we attribute to luck, external failures we attribute to factors outside our control.

The self-serving bias is a deeply embedded feature of human cognition, and it has direct implications for how users interpret the feedback a product gives them.

When a user fails at a task, their instinct is to look outward for the cause, not inward - which means a design that responds to errors with anything resembling blame will produce defensiveness rather than correction.

The practical consequence is asymmetric: error messages need to work much harder to remain non-judgmental and externally-attributed ("something went wrong" rather than "you entered the wrong format"), whilst success states can afford to be warmer and more generous without triggering scepticism. Users are predisposed to agree that they've done well; designing success states is, in that sense, the easier half of the problem.

Halo effect

The halo effect - sometimes called the halo error - is the tendency for an initial positive impression to shape how we perceive everything that follows.

Once we've formed a favourable view of something in one domain, we extend that favour to domains we have not yet evaluated; once we've formed a negative one, the inverse applies with equal force.

In product design, this makes the first moment of contact disproportionately important. An onboarding flow that is clear and well-considered creates a perceptual frame through which users will interpret every feature they encounter next; a clunky or confusing first screen creates a different frame, and that frame is difficult to reset even when subsequent interactions are better.

The same principle operates across all first touchpoints: the website before the download, the app icon before the open, the loading screen before the product. We are always in the business of first impressions, whether or not we treat it that way.

Chunking

Chunking - or grouping - describes the cognitive tendency to remember information more reliably when it is organised into clusters rather than presented as a continuous stream.

The classic example is a phone number: eleven digits in sequence are hard to hold in working memory; broken into three groups, the same digits become manageable.

For interface design, this has a direct structural implication. Navigation benefits from chunking; long forms benefit from chunking; content-heavy pages benefit from chunking - not because grouping is an aesthetic choice, but because it corresponds to how memory actually works.

The groupings we create don't just reduce cognitive load in the moment of use; they also contribute to the sense that a product has internal logic, that things belong where they are. A well-chunked interface feels considered without advertising that fact.

IKEA effect

The IKEA effect describes the increase in value people attribute to things they have had a hand in creating.

The name comes from the furniture company's assembly model: people consistently rate products they've assembled themselves more highly than identical pre-assembled ones, even when the quality of the result is objectively the same.

The application in digital product design extends further than it might first appear. Onboarding flows that ask users to make meaningful choices - not just click through defaults but actually configure something - create a sense of investment that purely passive onboarding doesn't.

Customisation options, user-generated content, and visible responsiveness to community feedback all tap into the same principle; the user who has shaped an experience in some way, however small, feels more ownership of it. This isn't a trick; it's recognition that participation is itself a form of value, and designing for it creates a relationship that passive consumption rarely does.

Authority bias

The authority bias describes our tendency to attribute greater credibility to statements from people or institutions we recognise as authoritative, sometimes to the point of accepting those statements without the scrutiny we'd apply elsewhere.

The Milgram experiments are the extreme case - a series of studies in which participants administered apparent electric shocks to strangers at the instruction of a figure in a lab coat - but the everyday version of this bias is far more benign and far more common.

In product design, authority signals are everywhere: expert endorsements, professional certifications, institutional affiliations, credentials displayed near claims.

Used well, these signals reduce the cognitive work a user has to do to decide whether something is trustworthy; they borrow credibility from an existing authority to extend it to a new context. The ethical boundary lies in the gap between genuine authority and performed authority - between expert endorsement that is accurate and representative, and social proof that is cherry-picked or misleading.

The bias is powerful precisely because users are not always interrogating it; that's exactly why it demands more care from the people deploying it, not less.

What these seven biases share is that none of them require the user to be irrational or inattentive for them to operate; they function within perfectly ordinary human cognition.

The user we design for is not a flawed version of some rational ideal - they are a complete person whose behaviour is shaped by exactly these kinds of tendencies, as ours is too. Designing with that in mind isn't a methodological trick; it's accuracy about the nature of the problem.