Use Cases

Check an AI Experiment Summary Against a Test Plan

Use a fictional signup test to catch a cohort mix-up, an unsupported conclusion, and a rollout recommendation that ignores a stop rule.

Audience and review notes

For: People checking an AI-written experiment recap before deciding what to ship.

Check facts against original sources and remove sensitive information. Review AI output before relying on it.

Check The Plan Before The Verdict

A recap can report a real improvement and still recommend the wrong next step. Compare three claims about a fictional workshop signup test with the plan and the results. The plan specifies who counts, which outcome matters, and when to pause.

Get The Pack

Give the AI only the test input. Use the separate review key to check its first answer, then compare it with a second run using the worksheet.

Download test input (.txt)

Download review key (.txt)

Download comparison worksheet (.txt)

Check Three Claims

CaseLook forWhat to verify
TEST-01Staff trial traffic counted as eligibleCompare the groups named in the plan.
TEST-02More confirmations reported as more attendanceName the measured outcome and what is not known.
TEST-03Rollout recommended despite a stop ruleApply the guardrail before proposing a rollout.

For real work, keep observed results separate from the decision. Confirm the eligible group, primary outcome, and stop rules with the test owner before sharing a recommendation.

Return to the experiment summary practice pack or choose your next check.

From reading to a decision

Keep the workflow connected

Choose a task-first path, save candidates locally, compare the shortlist, and recheck official details before use.