Metrics change diagnosis

Help me understand why this metric changed.

Data:
{{data}}

What else I know (releases, campaigns, tracking changes, seasons):
{{context}}

1. Describe exactly what changed: which metric, how much, from when, and which segments moved and which did not.
2. Check measurement first: could tracking changes, bot traffic, definition changes, time zones or incomplete recent data explain it?
3. List up to four explanations, most likely first, each with the evidence for and against it from the data and context.
4. For each, the quickest check to confirm or rule it out, naming the cut of data or report to look at.
5. What I should do now, and what I should wait to decide until the checks are done.

State as fact only what the data shows; label everything else as a hypothesis. Do not treat a correlation in timing as proof of cause, and do not assume figures, exchange rates, dates or user behaviour that the data and context do not give. Calculations from the data are fine if you show them.

How to use it

Replace {{data}} with metric figures, dates, and segment breakdowns. Replace {{context}} with known releases, campaigns, tracking changes, and seasonal factors, then submit the prompt.

Example

data

Week | Visitors | Signups | Conversion Sep 29 | 18,400 | 552 | 3.0% Oct 6 | 19,100 | 573 | 3.0% Oct 13 | 26,900 | 565 | 2.1% Oct 20 | 27,300 | 573 | 2.1% By device, week of Oct 20: desktop 9,800 visitors / 401 signups; mobile 17,500 / 172. Week of Oct 6 by device: desktop 9,600 / 392; mobile 9,500 / 181.

context

We launched a TikTok campaign on Oct 12. We also redesigned the signup form on Oct 14. Analytics is Plausible.