Decide if a metric improvement actually came from the feature you shipped
A metric moving up right after a launch is a correlation you want to be a story about your feature — check before telling it that way.
- Use it for
- Product people who shipped something and want to know honestly if it's actually responsible for a metric change that followed.
A metric improving shortly after a feature ships is genuinely exciting to report, which creates real pressure to credit the feature without checking whether something else — seasonality, a marketing push, an unrelated change — happened at the same time and is a more likely actual cause. This prompt checks the timing and alternative explanations honestly before the story gets told either way.
When not to use this
If you ran a genuine controlled experiment (A/B test) for this feature, that evidence already settles attribution directly — this is for the more common case of a launch without a controlled comparison.
Did this work?
You have an honest read on whether the feature is plausibly responsible for the metric change, or whether something else happening at the same time is a more likely explanation.
Tested on claude-opus-5. Evidence status is draft; it moves to battle-tested only on recorded runs, never by hand.