Review which posts brought useful attention
Compare your own posts against a defined goal and choose a small experiment for next week.
What you’ll produce
A weekly content review with checked metrics, format comparisons and one or two proposed experiments.
Inputs & connections
- Platform analytics export with post IDs, dates and comparable observation windows
- The goal for each campaign, posting calendar and any tracked leads or purchases
Upload a platform export for the first review. Recurring collection requires supported analytics access and a saved schedule; the prompt alone does not activate either.
Your starting prompt
Replace the bracketed fields. This is our suggested prompt, not a quote from the source.
Review [content analytics] for [period] against [campaign goals]. Verify what each metric means and keep posts with different observation windows separate. Identify the strongest and weakest posts by the relevant metric, not by likes alone. Show denominators for rates and flag missing data or small samples. Distinguish reach, engagement, profile visits and attributable inquiries. Suggest at most two changes to test next week with a clear comparison and success measure. Do not claim a format caused sales without attribution evidence or publish new content. Return the calculations, source post IDs and a short recommendation.
How to run it
- 01
Choose one business goal and export the metrics relevant to it.
- 02
Check metric definitions, post age, paid promotion and missing rows.
- 03
Compare similar posts and inspect the underlying content before interpreting a pattern.
- 04
Select a small test with a defined metric and comparable review window.
- 05
Review the next period and keep conclusions proportionate to the sample.
Review before you act
Check rate calculations, paid versus organic distribution, observation windows and whether inquiries can actually be traced to a post.
Limitations & success checks
- A successful post may reflect timing or audience distribution rather than its format.
- Platform analytics may not connect an interaction to a sale outside the platform.
Measure: Correctly calculated metrics, experiments completed and relevant inquiries where attribution is available.
Sources & attribution
The sources below support the reported example or a tool capability. They do not validate our prompt or prove the outcome in your environment.
Source post and author checked through X public oEmbed on October 17, 2026. The author describes a recurring Instagram-insights review using what she calls Claude Coworker. The retrieved text is truncated and refers to a video; it does not provide the complete setup or establish measured growth. The prompt, calculations and evaluation steps here are original editorial additions.
Creator / publisher: @ashrodriguezugc
Editorial review: 2026-10-17. How we label evidence.