Skip to content
Open fromusers

Tasks and success criteria

Usability tests

Interactive tasks tell participants what to do on the prototype. Success criteria define how FromUsers evaluates whether a goal-based attempt succeeded.

TypeParticipant experienceTypical success signal
Goal-basedComplete a defined outcomeReach a screen or success URL, or meet configured criteria
Free explorationBrowse without a single required outcomeTime spent or completion of the exploration step
First clickClick where they would start for a promptFirst click location
Five-secondBrief exposure, then questionsAnswers after the timed view
PreferenceCompare options and chooseSelected option

Good instructions:

  • State the goal in one or two sentences.
  • Avoid giving away the exact click path unless you are testing comprehension of labels only.
  • Mention constraints (desktop only, use search, do not use the back button) when they matter.
  • Keep tone neutral. Do not coach the happy path.

Example: “You want to update your billing email. Find where to change it and save a new address.”

For goal-based and website tasks:

  1. Open the task in Build.
  2. Choose how success is detected for your prototype type.
  3. For URL tests, set the success URL participants should reach.
  4. Preview and confirm detection works.

Success criteria should match a real completion state, not an intermediate page that people hit by accident.

First click

  1. Show the prompt and the starting view.
  2. Capture where they click first.
  3. Optionally ask why they chose that control.

Five-second

  1. Configure the brief exposure.
  2. Follow with recall or impression questions.
  3. Avoid packing other interactive tasks on the same page.
  1. Provide the options you want compared (designs or labeled variants).
  2. Ask for a choice and a short reason.
  3. Randomize order when bias is a concern (use logic or presentation options available in the builder).

Use free exploration when you want qualitative browsing before structured tasks. Limit time in the instructions if you need comparable sessions, then follow with rating questions.

Recommended pattern:

  1. Screener page.
  2. One or two interactive tasks.
  3. Difficulty and open-text follow-ups.
  4. Ending.

Too many tasks in one study increases drop-off. Prefer multiple shorter studies when scenarios diverge.

Results Studio shows task-level outcomes, first-click data, preference distributions, and linked recordings when enabled. Failed tasks are as useful as successes when you review paths.

Common questions

What is a success URL?

A URL that indicates the participant reached the goal on a website or URL prototype. FromUsers can use it when judging task success, especially with the website testing extension.

Do all tasks need success criteria?

Goal-based tasks should. Free exploration often does not. Preference and five-second tasks usually rely on the participant response rather than a destination URL.

Can I add questions between tasks?

Yes. Use standard questions for difficulty, confidence, or open comments between interactive tasks.