Tests & Results
Configure, run, and analyze UX tests.
Creating a Test
Every test starts with a URL and a task description. Here are the key configuration options:
- URL — The page where personas will start their session. For public sites, use the full URL. For localhost or VPN sites, use the Chrome Extension.
- Task description — A natural language instruction telling personas what to do (e.g. "Sign up for a free account").
- Success criteria — How the persona knows the task is complete (e.g. "You see a welcome dashboard").
- Panel size — The number of personas to run (1–20 depending on your plan).
- Execution mode — Cloud (CrowdRunner servers) or Your Browser (Chrome Extension).
Test Modes
Every new test starts with a mode chooser. Each mode configures the wizard and the persona behaviour differently:
Task Mode
The default mode. Personas follow your task description and try to complete it. Best for testing specific user flows like checkout, sign-up, or search.
Exploratory Mode
Personas browse your site freely without a specific goal. Useful for discovering general usability issues and understanding how different users naturally navigate your interface.
Chaos Mode
Personas deliberately try unusual interactions — edge cases, unexpected inputs, rapid navigation. Great for stress-testing your UI and finding error handling gaps.
UAT Mode
Upload a UAT document (or paste requirements directly) and personas execute each acceptance criterion as a separate test case, reporting pass/fail with evidence. Best for structured release sign-off.
Understanding Results
Session Timeline
Each persona's session is recorded as a step-by-step timeline with screenshots. You can click through each step to see exactly what the persona saw, what action they took, and what they were thinking.
Screenshots & Comparison
The comparison view shows screenshots from multiple personas side by side at each step. This makes it easy to spot where different user profiles diverge in their experience.
Friction Points
CrowdRunner automatically identifies moments where personas got confused, encountered errors, or struggled to complete an action. Each friction point includes the step number, a description of the issue, and a severity rating.
Reports
After all sessions complete, CrowdRunner generates an AI synthesis report that summarizes:
- Overall task completion rates across the panel
- Common friction points and their severity
- Demographic patterns — did certain user profiles struggle more than others?
- Actionable recommendations for improvement
Comparing Runs
Compare two completed test runs side by side. From the Tests page, switch to the Comparisons tab to create a new comparison and pick two runs (typically the same task across two URLs or before/after a change). The comparison report highlights differences in task completion, satisfaction, and friction points.
Sharing Results
Generate a shareable link for any test result to share with teammates or stakeholders. Shared links provide read-only access to the full results without requiring a CrowdRunner account.