Building Mental Models, Not Just Interfaces
Users don't experience your UI — they build a mental model of it. The quality of that model determines everything about how confidently and successfully they navigate your product.
When a user first encounters a new interface, they're not using it — they're building a theory about it. They're forming a mental model: a simplified internal representation of how the system works, what it can do, and how its parts relate to each other. Every subsequent interaction with the product is mediated by that model.
This is one of the most important concepts in interface design, and one of the least operationalized. Most UX work focuses on the surface: making things look clear, making flows logical, reducing steps. All good. But if those improvements don't cohere into a mental model that users can actually hold and use, they will require constant re-discovery of things they've 'already learned.'
A good mental model has three properties: it is accurate enough to predict system behavior; it is simple enough to hold in working memory; and it is stable — the system behaves consistently enough that the model doesn't need to be rebuilt with every new version.
Consistency is the most underrated design value precisely because of this. Inconsistency doesn't just confuse users in the moment. It actively degrades their mental model over time. They stop predicting and start reacting. They lose confidence. They slow down. They make mistakes.
I design for model-building explicitly. This means ensuring that similar actions have similar results throughout the product. It means labeling things in ways that match how users already think about them — not in ways that make internal architectural sense.
The measure of success isn't whether a user can complete a task with coaching. It's whether they can return to the product after two weeks and navigate without relearning. That durability is the signature of a product that built a good mental model rather than just a usable interface.