Your users already know how the web works. Design with that, not against it
Jakob's Law states that users bring expectations from every other product they've used to yours — and every pattern you break from convention makes them work harder. Understanding it is the difference between standing out for the right reasons and confusing users for no good reason.
The Law in Plain Terms
Jakob's Law, coined by Jakob Nielsen, states: users spend most of their time on other products, and they carry those expectations into yours. If your interface works differently from the products a user already uses daily, they have to unlearn before they can learn.
The implications go deeper than "follow conventions." Every novel interaction pattern you introduce carries an adoption tax. You're not just asking users to learn something new — you're asking them to override something they do automatically. That's a meaningful cost, even for motivated users who are genuinely trying to get things done.
The Real Cost of Being "Different"
Products that earn praise for being different are usually different in outcome — they're faster, better organised, or strip out steps that felt pointless. The underlying interaction conventions typically stay familiar.
When teams mistake novelty for differentiation, the cost surfaces as:
- Support tickets about basic navigation ("where did X go?")
- Onboarding drop-off caused by hesitation at unfamiliar patterns
- User interviews where people describe the product as "weird" or "hard to get used to" without being able to say exactly why
- NPS feedback where "too complicated" appears even when the product is objectively capable
The product hasn't failed. The interaction language has.
Where Convention Should Win
Not every familiar pattern deserves preservation. Jakob's Law is most useful as a forcing function: if you're about to break a convention, be deliberate about it. Ask what Mental Models users already hold, and what you're gaining by overriding them.
Convention usually wins at:
- Navigation placement (top bar or left sidebar for primary nav in web apps)
- Form field labelling and inline error placement
- Iconography for core actions — trash for delete, pencil for edit
- Search behaviour: results update, filters refine, enter submits
- Table sorting and list patterns
These aren't sacred. But the bar for changing them should be high, because you're designing for users who interact with dozens of other products every week.
When to Diverge — and How
Jakob's Law doesn't argue against innovation. It argues for understanding your users' Mental Models before you diverge from them.
There are legitimate reasons to break convention:
- The existing convention actively misleads users in your specific context
- Research shows users struggle with the standard pattern for your particular task type
- The conventional pattern conflicts with a core product promise — a data-heavy tool that needs a fundamentally different layout to surface what actually matters
The difference between good divergence and unnecessary friction is intentionality backed by evidence specific to your user base — not aesthetic preference or a desire to look distinctive.
How It Connects to Design Systems
A Design System that aligns with platform conventions — web, iOS, Android — is partly an expression of Jakob's Law at the component level. Familiar button states, predictable modal behaviour, standard input patterns: these reduce cognitive load not because they're elegant, but because users already know how they work.
When a design system drifts from platform norms — unusual scroll behaviour, non-standard dropdown interactions, unexpected focus states — it compounds the friction of every new user learning the product. The cost is diffuse and hard to attribute directly, but it shows up in onboarding data and support volumes over time.