Create intermediate Works with: Any AI

Performance Triage: Which Posts to Repeat

Table output Has example

Learn from results without over-reading them.

The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.

Help me analyse post performance. Data for my last [NUMBER_OF_POSTS] posts with format, topic, time and results: [POST_DATA]. Goals: [GOALS].

Group posts by format and topic and compare typical results, not only the best. Flag outliers and possible explanations (timing, shares, paid support), and note when the sample is too small to conclude. Recommend three things to repeat, three to test, and two to stop, each with the evidence. Design a simple test plan for the next four weeks that changes one thing at a time. Warn against reading too much into one viral post. Provide a table template for logging future posts so that analysis becomes easier.

Also add a worked example with ten invented posts and the conclusions that the data do and do not support, and a rule for waiting for a few weeks of data before changing strategy.

Include a short guide to calculating a simple engagement rate in two ways and explaining why I should use the same method every time, with the formula written out in words.

Fill in before use

Optional: type your details and the prompt above updates. Nothing is saved or sent.

Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.

Invented table (10 posts)
Format | Posts | Median reach | Median saves
Carousel tips | 4 | 1,250 | 38
Short video | 3 | 2,100 | 12
Photo with story | 3 | 800 | 9
Conclusion: videos reach more people; carousels earn more saves. Do not conclude too much from 10 posts.
Repeat: carousel tips. Test: video with tips. Stop: stock photos.
One viral video (9,000 reach) was paid support; treat as an outlier.
Test plan: for four weeks post one carousel and one short video each week with the same topic, compare reach, saves and clicks, and record the result in the log.

• social