BoardSnap is an iOS app that reads whiteboard photos and produces clean summaries and action items in about ten seconds. This usability test plan template structures a moderated usability test — tasks, success criteria, metrics, and observer protocol — on a whiteboard before the first session runs.
Use this template before running any moderated usability test — remote or in-person. The planning board is the spec the whole team works from: the moderator knows the tasks, the observers know what to capture, and everyone knows the success criteria before the first participant joins the call.
Budget 45 minutes to plan. Add 30 minutes to pilot the test with a colleague before running participants.
Write the specific question this test will answer. 'Can first-time users complete onboarding without help?' is a test objective. 'Understand the user experience' is not. The objective determines everything else — the tasks, the success criteria, and the metrics.
Write three to five tasks participants will attempt. Each task should be: written in the participant's language (not product language), scenario-framed ('You've just joined a new project. Find where your team's previous boards are stored.'), and completable without the moderator's help. Write the tasks in the order participants will attempt them — logical progression reduces cognitive load.
For each task: what counts as success? Completion without assistance? Completion in under 90 seconds? Correct navigation path? Write the success condition explicitly. Without this, observers will disagree about what they saw.
What will you measure? Task completion rate, time on task, number of errors, and System Usability Scale (SUS) score are the most common. Write the metric and the method for capturing it. If someone needs to take notes during the session, assign a note-taker.
Who is observing? What are they capturing? How do they communicate questions to the moderator without interrupting? Write the observer roles and the communication protocol. Observers who don't have a defined capture role generate unhelpful ad-hoc commentary.
Read the design hypothesis or the product decision this test will inform. Write the single question the test will answer. Everything else flows from this question — don't skip it.
Draft each task in language the participant would use, not product terminology. If your product calls them 'projects' but your participant says 'folders,' write 'folders' in the task. Participants who hear a term they don't recognize will fail the task for the wrong reason.
For each task: write a single observable success condition. 'Participant navigated to the correct screen without clicking on the wrong section first' is observable. 'Participant understood the navigation' is not.
More than three metrics per session creates data overload that prevents action. Choose the metrics that most directly answer the test objective. Task completion rate is almost always one of them.
For each observer: write their capture responsibility on the board. Observer A tracks verbal commentary. Observer B tracks navigation paths. Observer C watches body language and hesitation. Defined roles produce richer, more consistent data than open observation.
Run the full test on a colleague before the first participant. The pilot reveals tasks that are ambiguous, success criteria that are unmeasurable, and session flows that are too long. Write the changes on the board and snap the updated version.
BoardSnap reads the objective, tasks, success criteria, metrics, and observer protocol. The output is a structured test plan ready to share with all observers before the first session.
Usability test plans written in a Google Doc often don't get read before the session. A whiteboard test plan gets discussed before it's finalized — the observer roles get challenged, the success criteria get sharpened, and the tasks get rewritten until they're real. The result is a test that produces usable data instead of a session that produces opinions.
BoardSnap captures the agreed-upon plan before anyone runs a session. The structured output goes into the test folder and serves as the spec for every observer.
Five per user segment for qualitative usability testing (based on Nielsen's research showing five users reveal ~85% of usability issues). If your product serves multiple distinct user types, test five per type. Quantitative usability tests require larger samples — consult a statistical power calculator.
Moderated testing has a facilitator present during the session who can probe participants' thinking in real time. Unmoderated testing uses a tool (Maze, UserTesting) to run tasks asynchronously without a facilitator. Moderated produces richer qualitative data; unmoderated produces larger samples faster. This template is designed for moderated testing.
Task-based (directed) rather than question-based (open-ended). Give participants something to do, not something to answer. 'Try to add a new board to your project' is a task. 'How would you add a new board?' is a question — it produces verbal descriptions instead of observable behavior.
The free tier includes one project and 30 boards. Pro is $9.99/month or $69.99/year for unlimited boards and AI chat on every board you snap.
No exporting, no transcription. Snap the board, get the action plan.