Conversational Interfaces Break Jakob's Law

Think of every website’s three-line hamburger menu. It wasn't always intuitive. You learned it.

We all did.

But what if the most natural interface was the one humans mastered thousands of years before computers existed?

Conversation might be the ultimate interface. It requires no learning curve beyond what we already possess.

This idea was sparked by a tweet thread from Denis Jeliazkov (Jeliazkov, 2024):

Jakob's Law vs Natural Language

The Clash Between Learned Patterns and Innate Human Behavior

Jakob's Law comes from Jakob Nielsen (Nielsen Norman Group), who observed that users spend most of their time on other websites. This creates an expectation that your site should work the same way as all the sites they already know.

The principle states: Users expect websites and digital products to be designed consistently with others they've used before.

This law has guided digital design for decades. It explains why drastic redesigns often fail. Users rely on accumulated experience to navigate digital spaces.

According to Nielsen Norman Group research (Nielsen, 2018), users prefer your site to work the same way as other sites they know, making consistency a critical factor in user experience design.

But conversation follows different rules.

A child learns to speak and have conversations naturally. No instruction manual needed.

Compare this to the hamburger menu. Remember the first time you saw those three lines? You had to learn what they meant. Then you saw them again and again until they became "intuitive."

But they were never truly intuitive. They were learned.

Conversational interfaces tap into abilities we've refined since childhood. They don't require special training or experience with similar systems.

The Memory Factor

When AI Remembers What You Care About

Standard UX design provides gravitational points that support Jakob's Law. Buttons stay in predictable places. Navigation follows patterns. Icons maintain consistent meaning.

Conversations don't work this way. They flow. They adapt. They branch in unexpected directions.

This creates a challenge for traditional UX thinking. How do you apply familiarity principles to something inherently fluid?

The answer might lie in memory.

When you interact with traditional interfaces, you build memories of how they work. You remember where the search box lives, what icons mean, and how to use them. You develop muscle memory for common tasks.

Jakob's Law leverages these memories. It says: "Design your interface to match what users already remember from elsewhere."

But conversational interfaces flip this dynamic. Instead of you remembering the interface, the AI remembers you.

An AI assistant might notice you frequently ask about the Boston Celtics, Jacksonville real estate trends, or machine learning developments. It builds familiarity by remembering your interests and needs.

This creates a different kind of consistency. Not visual consistency, but contextual consistency.

Recent research into memory-enhanced conversational AI (Oni, 2025) demonstrates how advanced memory systems enable chatbots to remember past conversations, creating more personalized and context-aware interactions.

Context Over Components

How Conversational UX Creates New Design Patterns

Traditional interfaces achieve familiarity through visual and interactive components. Buttons look like buttons, menus behave like menus, and links appear clickable.

Conversational interfaces achieve familiarity through context and memory. They remember:

• Your frequent topics

• Your vocabulary preferences

• Your interaction style

• Your previous questions

• Your reaction to the answers

This shifts design thinking from visual patterns to memory patterns. The system doesn't just remember where you left off. It remembers who you are.

According to UX Magazine (Gravestein, 2024), conversational AI represents the first new UI paradigm in 60 years. It reverses the "locus of control" so that computers learn to talk to us rather than us learning to speak to computers.

Google's search interface has changed little in 25 years. Why? Because Jakob's Law suggests that changing a familiar interface costs users mental effort.

But conversation allows constant evolution without breaking familiarity. The interface remains stable: you ask questions, you get answers.

What changes is the underlying intelligence. The responses improve. The understanding deepens. The memory expands.

This creates an interesting paradox. The most advanced interface might change the least visually while changing the most intellectually.

Redesigning for Conversation

Breaking Free From Jakob's Constraints

For designers and product teams working with conversational interfaces, this paradigm shift demands new thinking:

• Focus on memory architecture over visual architecture. How will your system remember user preferences and history?

• Consider conversational flows rather than user flows. Map the many ways a conversation might branch.

• Build intelligence that recognizes context shifts. Humans naturally signal when they're changing topics.

• Develop personality consistency. Voice and tone become your interface elements.

• Create retrieval systems that surface relevant previous conversations. Context often spans multiple interactions.

Questions to consider:

1. How might your conversational interface signal its capabilities without visual cues?

2. What memory structures would make your users feel truly understood?

3. Where does Jakob's Law still apply within conversational experiences?

4. How can you maintain consistency when each conversation is unique?

5. What metaphors from human conversation should apply to AI conversation?

The shift from Jakob's Law to conversational interfaces represents more than a technological evolution. It marks a return to our most natural form of information exchange.

We didn't need to learn how to have conversations. We needed to learn how to use hamburger menus.

As AI advances, our interfaces may finally catch up to our oldest and most refined skill: talking to each other.

The most intuitive interface might not be the one that matches other websites and apps.

It might be the one that matches how we communicated for thousands of years before digital interfaces existed.

That realization breaks Jakob's Law wide open.