Overview
Method category: Evaluative product experiment
Usability Testing involves observing how easily users complete tasks using a product or prototype to assess the product's understandability.
Within GLIDR, conduct a Usability Test as an Experiment connected to Assumptions about product UX and UI. During the Run phase, create one Evidence - Interview for each user tested (typically 5-7 people). Identify patterns among testers, then in the Analyze phase, determine necessary product improvements.
In Brief
Usability testing observes people completing tasks with interactive products at any fidelity level, from paper mockups to fully functioning systems. Observers record task completion and note user confusion or frustration.
Helps Answer
- How do people use the product or service?
- Do people understand how to use the product or service (as intended)?
- What do people experience at different touchpoints while using the product or service?
Tags
- Qualitative
- Value proposition
- Customers
- Description
Time Commitment and Resources
Five-user usability tests can be completed in half a working day with minimal resources. Tests typically require 5-7 users unless tasks are complex and collaborative. While extensive equipment (labs, cameras, eye tracking, one-way mirrors) exists, it's not necessary.
How To
1. Prepare: Critical preparation minimizes biases. Prepare:
- Introduction script
- List of use cases and tasks
- Task context (e.g., "You are thinking of buying a car.")
- Task description (e.g., "Compare prices of various cars.")
- Reasonable time limit per task
- Recording equipment (notebook or camera)
Consider environment carefully. Usability labs offer detailed observation but lack realism, while real environments provide authentic behavior but complicate observation.
2. Frame for Feedback
Explain the experiment's purpose and reassure users that difficult tasks aren't their fault—problems represent desired feedback. Use prewritten scripts for consistency.
3. Explain the Task
Present a single task and its context to the user.
4. Observe the User
Observe while asking users to verbalize impressions, intentions, and expectations. Avoid providing guidance; only ask about thought processes, feelings, or experiences. Record using audio, video, screen capture, or eye tracking.
5. Repeat if Necessary
Assign additional tasks and repeat steps 3-4 until completion.
6. Exit Interview
Thank the user and ask open-ended follow-up questions clarifying their experience.
Interpreting Results
Multiple observers help eliminate subjective bias. Synthesize observation notes, noting emotional responses like frustration. Identify functional issues reported by most/all participants.
Given small sample sizes, prioritize consistent usability problems found across cases and retest with proposed solutions.
If all users complete tasks successfully, the product is usable but may not be desirable or deliver the value proposition.
Potential Biases
- Hawthorne effect (observer effect): Users behave differently when aware of observation.
- Social desirability bias: Users attempt tasks favorably to impress experimenters.
- Confirmation bias: Experimenters design use cases confirming preconceptions.
- Selection bias: Incorrect audience selection severely biases results (e.g., testing with existing users won't reveal new user issues).