There is a myth that interaction design is about creativity. In practice, it's almost the opposite: most good interaction decisions consist of choosing the established pattern rather than inventing a new one.
The user does not want to be surprised when trying to log in or fill out a form. He wants it to work as he expects it to. Originality in the wrong place becomes friction, and friction becomes abandonment.
This text shows, with concrete examples, how to choose interaction patterns that reduce friction. It's not about aesthetics. It's about making the user's path so obvious that they don't even realize there was design there.
What actually is interaction design
Interaction design is the design of how the user and the system talk to each other: what happens when they click, what the system responds, how it knows something went right or wrong. It is the choreography between action and reaction.
Good interaction design answers three questions at all times: where am I, what can I do here, and what happens if I do it. When any of these questions are left unanswered, the user freezes.
Choosing patterns is the heart of the work. And it is done by evidence and context, not by personal taste.
How to choose: criteria with examples
Familiarity beats originality
Example: think of the shopping cart icon in e-commerce. Nobody innovates on this. The user has learned this symbol from thousands of websites, and changing it would only create confusion. The cost of relearning is high; the gain from "being different" is zero.
The criteria: before creating a new pattern, ask if there is already a convention that the user has mastered. If it exists, use it. Reserve creativity for where it creates value, not where it creates doubt.
Give immediate feedback for every action
Example: when you make a transfer via Pix, the app immediately shows a clear confirmation, with the amount and recipient. Imagine if, when you tapped "confirm", the screen sat still for five seconds without any signal. The anxiety would be immediate, and many people would play again, causing errors.
The criteria: every relevant action needs a visible and instantaneous response, a charge, a vibration, a message. System silence is interpreted as failure.
Reduce the number of decisions
Example: government forms were famous for asking for everything at once, on endless screens. Newer good public digital services have broken this down into short steps, one question at a time, with progress bar. The completion rate goes up when the user sees the end.
The criteria: each screen must ask for the minimum number of decisions. When there is too much choice together, the user freezes, the classic paradox of choice.
Prevent error before treating it
Example: a good date field won't let you enter February 31st. A good form validates the CPF as you type it, and not just after you click submit and lose everything. Prevention is better than correction.
The criteria: design the interaction so that the error is difficult to make. And when it happens, the message needs to tell you what to do, not just "error 400".
Respect the context of use
Example: a banking app and a gaming app require opposite interactions. At the bank, every important action deserves confirmation, because mistakes cost money and generate anxiety. In the game, constant confirmations kill fluidity and irritate. The same pattern, asking for confirmation, is right in one context and wrong in another.
The criteria: the choice of pattern depends on what is at stake and the emotional state of the user at that moment. There is no universally good interaction; there is interaction appropriate to the context. Copying a pattern from an app without understanding why it works there is a recipe for disaster.
Think tap, not click
Example: buttons designed for the mouse, small and close together, turn into a nightmare on a cell phone, where the finger is imprecise and covers a large part of the screen. Apps poorly adapted to the desktop leave the user touching the wrong button all the time, hitting the button next to the one they wanted.
The criteria: in Brazil, where the majority accesses via cell phone, designing for touch is the rule, not the exception. Generous touch areas, adequate spacing and important elements within thumb reach. Finger ergonomics are as important as screen logic.
The mistake that destroys good interactions
The most expensive mistake is the designer designing for himself. He knows the system by heart, navigates it easily and concludes that everything is clear. But he is not the user.
In Brazil, this is especially serious in services that serve diverse populations, the elderly, people with low digital familiarity, people using entry-level cell phones with a small screen and unstable internet. A standard that works for the designer may be insurmountable for the real citizen.
Another mistake is inconsistency. When the same type of action behaves differently on different screens, the user needs to relearn at each step. A button that is in one place on one screen and in another place on the next breaks the automatism that makes use flow.
And there is the mistake of following blind trends. A trendy pattern emerges, a hidden menu, a new gesture, and everyone copies it, even without evidence that it works for their audience. A beautiful pattern in a portfolio is not always an effective pattern in production.
Maintain consistency across screens
Example: in well-built apps, the main action button is always in the same place, with the same color and the same behavior, screen after screen. The user learns once and applies it throughout the product. In inconsistent apps, each screen appears to have been created by a different person, and the user needs to relearn where things are at each step.
The criteria: standardize repeated elements, buttons, navigation, messages, icons, so that the user can build automatisms. Consistency is what allows usage to flow without conscious effort. Each unjustified exception breaks this automatism and demands attention that the user would prefer to spend on the task. A design system, even a simple one, is the instrument that guarantees this consistency at scale, it transforms scattered decisions into a single standard, decided once and applied everywhere.
Choosing well means making the interface invisible
The best interaction design is the one that no one notices. The user enters, does what they need and leaves, without friction and without thinking about the interface. When design becomes the protagonist, it's usually because something is getting in the way.
Choosing patterns is, in essence, an act of humility. It means recognizing that the user has habits, expectations and limits, and designing to respect them, not to impress market peers.
The examples above have a common thread: they all come from the reality of the user, not the vanity of the creator. This is the criterion that separates interaction that works from interaction that just looks modern.
If you're reviewing your product's interactions and suspect you're designing for yourself rather than the user, it's worth testing with real people before deciding. There are other articles here about UX and usability that delve deeper into this topic, and I am available to discuss specific cases.