<aside>

Testing Overview

Before finalizing the FitCheck prototypes, we planned two complementary forms of evaluation: an internal heuristic evaluation and an unmoderated usability study.

The heuristic review focuses on interface-level issues such as feedback, consistency, user control, error prevention, and clarity. The usability study focuses on how real participants navigate the prototype, complete representative tasks, understand cross-service connections, and respond to AI-assisted features.

Together, these methods help us identify both design-rule violations and real interaction problems before the final iteration.

</aside>

<aside>

Evaluation Strategy

<aside>

Heuristic Evaluation

We use a structured heuristic review to identify usability issues before interpreting participant behaviour.

Representative screens from the four FitCheck services are reviewed for system feedback, clarity of language, user control, consistency, error prevention, recognition over recall, and visual simplicity.

Issues are recorded together with their location, severity, and a proposed design response.

</aside>

<aside>

Usability Testing

We use Maze to conduct an unmoderated prototype study with participants from FitCheck’s target audience.

Participants complete a series of realistic tasks without being shown the intended interaction path. After each task, short questions capture perceived difficulty, confidence, and feature understanding.

The study ends with free exploration and overall feedback so that we can compare task performance with participants’ own impressions of the experience.

</aside>

</aside>

<aside>

01 Testing Approach

Exploring Usability Through Maze

To evaluate FitCheck before the final prototype iteration, we created an unmoderated usability study in Maze.

Rather than treating the study as a formal large-scale user research exercise, our goal was to become familiar with usability-testing workflows and understand how participants interact with realistic prototype tasks.

The study combines short background questions, four task-based prototype tests, free exploration, and final experience feedback. Most questions are optional so participants can move through the study at their own pace.

Participants were informed that background questions were optional before beginning the prototype tasks.

</aside>

<aside>

02 Study Structure

01 — Participant Context

Four optional questions capture age, outfit-planning habits, fashion-related activities, and previous experience with fashion applications.

02 — Task-Based Testing

Participants complete four representative FitCheck tasks without being shown the intended interaction path.

03 — Post-Task Reflection

Short questions capture perceived difficulty, confidence, feature understanding, and usefulness immediately after relevant tasks.

04 — Free Exploration

Participants are given time to explore FitCheck without a predefined goal and identify what naturally catches their attention.

05 — Final Feedback

Participants reflect on overall usability, cross-service clarity, AI assistance, and possible improvements.

</aside>

<aside>

03 Prototype Tasks

<aside>

Task 01 - Add an Item to Your Wardrobe

Add a clothing item to the Virtual Wardrobe, review the information, and save it.

</aside>

<aside>

Task 03 - Recreate a Shared Look

Find an outfit you like and recreate a similar look using items from your own wardrobe.

</aside>

<aside>

Task 02 - Create an Outfit

Create an outfit for a casual occasion using FitCheck and save the completed look.

</aside>

<aside>

Task 04 - Create and Share an Outfit Post

Create a new outfit post, review what will be shared, and publish it to the FitCheck community.

</aside>

</aside>

<aside>

Why We Added Free Exploration? - Beyond Goal-Based Tasks

Structured tasks are useful for evaluating specific flows, but they also direct participants toward predefined goals.

We therefore added a free-exploration stage after the four main tasks. Participants could browse any part of FitCheck that interested them and move on whenever they felt ready.

This gave us an opportunity to observe which areas naturally attracted attention once explicit task instructions were removed.

</aside>

<aside>

05 What We Evaluated

<aside>

Navigation & Discoverability

Can participants find important actions without being told where to click?

</aside>

<aside>

Cross-Service Understanding

Do transitions between FitCheck services feel understandable and connected?

</aside>

<aside>

System Feedback

Is it clear when actions such as saving, adding, or publishing have been completed?

</aside>

<aside>

AI & User Control

Does AI assistance feel helpful while leaving important decisions with the user?

</aside>

</aside>

<aside>

</aside>

<aside>

</aside>