# Cognitive Walkthrough

A learnability inspection that steps through a task from a new user’s perspective and questions whether each action and its effect will make sense.

- Type: method
- Domain: Research & behavior
- Level: Intermediate
- Last reviewed: 2026-09-21
- Original: [Soroush’s Product Wiki](<https://wiki.soroush.uk/topic/cognitive-walkthrough/>)

## What it means

A cognitive walkthrough is an expert inspection focused on learning by exploration. Evaluators define a user, goal, starting state, and correct action sequence, then examine each step: Will the user try to achieve the right effect? Notice the available action? Connect it to the goal? Understand the feedback after acting? The method builds a structured account of likely learnability breakdowns.

## Why it matters

Teams familiar with a product no longer see the inferences demanded from a newcomer. A walkthrough makes those assumptions explicit and links interface cues to a user’s probable knowledge and goals. It is especially good at finding hidden actions, unfamiliar labels, ambiguous feedback, and steps that rely on training or insider vocabulary.

## When to use it

Use it for onboarding, first-time setup, infrequent administrative tasks, safety-critical procedures, and workflows where users cannot rely on established habits. It can evaluate sketches or prototypes before recruitment, then inform tasks for usability testing.

## How it works

1. Define a specific user profile, their prior knowledge, a realistic goal, and the exact starting state.
2. Write the intended action sequence at a level that exposes each decision and interface response.
3. At every step, ask whether the user will form the right intention, see the action, connect it to the goal, and read the feedback.
4. Record a success or credible failure story with interface evidence rather than a yes-or-no score alone.
5. Prioritize breakdowns, revise cues or sequence, and test the assumptions with representative new users.

## A practical example

A team walks through connecting a smart thermostat for a renter with no home-network expertise. At the step labeled “Provision device,” evaluators conclude the user will neither recognize the term nor know that the pulsing icon is actionable. The flow is rewritten as “Connect thermostat,” with explicit progress and recovery guidance, then tested with first-time installers.

## Where it breaks

The quality of the result depends on an accurate model of the user’s knowledge and goals. Evaluators may imagine behavior that real users do not exhibit, and the method underrepresents expert use, social context, accessibility needs, and long-term efficiency. It can become laborious for branching workflows and should not replace empirical observation.

## Common mistakes

- Walking through as an experienced product team member rather than the defined new user.
- Using vague tasks that omit the user’s knowledge, goal, and starting conditions.
- Answering checklist questions without writing a plausible success or failure story.

## Visualization

[View the visualization on the website](<https://wiki.soroush.uk/topic/cognitive-walkthrough/#visualization-cognitive-walkthrough>)

## Sources

1. [Evaluate Interface Learnability with Cognitive Walkthroughs](<https://www.nngroup.com/articles/cognitive-walkthroughs/>) — Nielsen Norman Group · Professional guidance
2. [How to Conduct a Cognitive Walkthrough](<https://www.interaction-design.org/literature/article/how-to-conduct-a-cognitive-walkthrough>) — Interaction Design Foundation · Professional reference

## Related topics

- [Heuristic Evaluation](<https://wiki.soroush.uk/topic/heuristic-evaluation/>)
- [Usability Testing](<https://wiki.soroush.uk/topic/usability-testing/>)
- [First-Click Testing](<https://wiki.soroush.uk/topic/first-click-testing/>)
