arrow_back Research

Usability Testing

Discover what Usability Testing is and how this UX research technique helps you find and fix problems in your design by observing real users.

toc Table of Contents expand_more
Info Icon Definición Rápida
A usability test is a fundamental UX research method for evaluating how easy a product is to use. It consists of observing representative users as they attempt to complete typical tasks on a prototype or real product to identify where they encounter problems and frustrations.

What Is Usability Testing?

Imagine you design the instructions for assembling an IKEA piece of furniture. You think they are perfect. Then, you ask someone to assemble the furniture following them and you watch in silence. You see that they get stuck on step 3, that they try to put a screw in the wrong place, and that they end up frustrated. You are not evaluating whether the person is “smart” or not; you are evaluating the clarity of your instructions.

That is a usability test. We do not test the user – we test the design. We give a user a task (e.g., “find and buy a pair of red sneakers in your size”) and observe how they interact with the interface to accomplish it.

The key components are:

  • A representative participant: Someone who fits your [[Personas]] profile.
  • A prototype or product: From a low-fidelity Wireframe to the final product.
  • Realistic tasks: Scenarios that a real user would attempt to accomplish (e.g., “change your password,” “add a friend”).
  • Observation: A moderator who guides the session (or software that records it) and takes notes on usability issues.

Why Is It Important?

  • It identifies real problems: It reveals where the interface is confusing, inefficient, or simply does not work as users expect.
  • It reduces risk and cost: It is much cheaper and easier to fix a design problem in a prototype than in a product that has already been coded and launched.
  • It provides evidence for decision-making: Instead of arguing about which design is “better,” usability testing data shows which one works better for users.
  • It increases satisfaction and conversion: A product that is easier to use generates happier users who are more likely to achieve their goals (and the business’s goals).

Key Methods

Tests can be:

  • Moderated vs. Unmoderated: A moderator guides the user in real time (moderated) or the user takes the test alone, following software instructions (unmoderated).
  • In-person vs. Remote: The test takes place in the same physical location or over the internet.

The “Think Aloud” Protocol: This is the gold standard. You ask the participant to verbalize their thoughts while performing the tasks. “What are you looking at?”, “What do you expect to happen if you click there?”, “What confuses you?”. This gives you direct access to their mental model.

Mentor Tips

  • “We are testing the design, not you”: Repeat this phrase at the beginning and during the test. It is crucial that the user feels comfortable and does not think they are being examined. There are no “right” or “wrong” answers.
  • Resist the urge to help: If a user gets stuck, do not tell them what to do. Ask them “What would you try to do now?”. The goal is to see where the design fails, not to help the user succeed.
  • 5 users are enough (to start): The Nielsen Norman Group demonstrated that with just 5 users you can discover about 85% of the most important usability problems. You do not need a large budget to get started.
  • Focus on behaviors, not opinions: What users do is much more important than what they say they like. If 5 out of 5 users say the button is “nice” but none of them can find it, you have a usability problem.

Resources and Tools