Check whether the outcome improved
Set task acceptance checks
Require a neutral goal, realistic trigger, starting state, completion condition, recovery observation, and product question. Add a check that the task does not repeat a control label or disclose the intended sequence.
Pilot the risky moments
Ask a participant or colleague to attempt the task without coaching. Record whether the scenario is understandable, whether the data state supports it, and where facilitation was tempted to help. Revise only the wording or setup that blocks the intended learning.
Approve the evidence boundary
State which behavior the task can reveal and which outcomes require another method. Record pilot deviations, access needs, and review owner. Keep an inconclusive result available when the task or prototype was not usable enough to test the question.
What to carry forward
The verification record should show task criteria, pilot behavior, wording changes, access needs, and evidence limits. Approve the task for the context it represents. Retain a follow-up when prototype fidelity or scenario realism could still determine the result.
Technical background: Nielsen Norman Group research guidance.
Keep the decision with the work.
Use a Work Item in Aglet to record the problem, the evidence you have, and the next decision. Add an owner and priority, then keep updates in the discussion so the next person can follow the reasoning.
Create an account See the product workflow