AI Blocks › Track your numbers
Numbers drop explainer
Paste the numbers before and after a drop. Get the likely causes ranked by evidence, and what to check.
Numbers drop explainer is a free AI prompt for the job "Track your numbers". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · Made by AI Blocks · Thinking: high
What it does
Sales fell twenty percent and everyone has a theory by lunchtime. The marketing lead blames the platform, the founder blames the price change, and the number that would settle it sits in a dashboard nobody opened. Eric Ries has argued in public that the honest version is to show cause and effect with evidence, and to say what you do not know. This prompt sizes the drop against the usual swing, breaks the metric into its parts to find which one moved, and puts every change in the window on the same timeline, because a change that came after the drop cannot have caused it. It ranks the causes with evidence for and against, always including normal variation and broken measurement, flags the vanity numbers that do not answer the question, names the one check that would settle the top cause, and writes six to ten lines for your audience…
You give it
- The metric by day or week, before and after
- The numbers that feed it
- Everything that changed in the window, with dates
- Who the explanation is for, and the usual swing (advised)
You get
- The drop sized, with where it sits
- Changes on the timeline and causes ranked by evidence
- Vanity numbers flagged, and the one check to run
- A six to ten line explanation for your audience
Use it when
- A weekly number fell and you have to explain it on Monday.
- Two people on the team are sure of two different causes.
- You suspect the drop is inside normal variation and want to say so with evidence.
How to use it
- Copy the prompt, then paste it into Claude, ChatGPT or Gemini with the numbers under it.
- List every change in the window with its date, including the ones you think are irrelevant.
- Read CAUSES RANKED. If a LIKELY cause started after the drop, something in your dates is wrong.
- Run THE ONE CHECK before you send the explanation.