Instagram insights export analysis with what to post more of
Reads an export of Instagram post and reel insights, compares formats, topics, hooks, lengths and posting times on fair rates rather than raw totals, and returns what to post more of, what to stop and a four-week test plan. Use it for brands, creators and social media managers who post daily but cannot tell what actually works.
The output this prompt produced, unedited.
The prompt
Constraints first
Compare posts on rates, such as saves and shares per 1,000 reach, not on raw likes.
Compare reels with reels and carousels with carousels before comparing formats with each other.
Do not call a pattern real when it rests on fewer than five posts; say "too few to tell".
Keep paid boosts apart from organic posts.
Inputs
Account and goal: [e.g. a home bakery, goal is WhatsApp orders]
Export: [paste rows: date, time, format, caption first line, topic, length for reels, reach, plays or views, likes, comments, saves, shares, profile visits, link clicks, boosted yes or no]
Period: [e.g. last 90 days]
Topics or series you post: [e.g. behind the scenes, product, customer review, recipe tip]
Output format
| Format | Posts | Median reach | Saves per 1,000 reach | Shares per 1,000 reach | Profile visits per 1,000 reach |
The same table by topic.
Top five and bottom five posts by the rate that best fits the goal, each with a likely reason from the hook, topic or format.
Best days and times, only if there is enough data.
Post more of, keep, and stop: three items each.
A four-week test plan with one change at a time and the number that would show it worked.
instagram insights analysisinstagram analyticsreels performancesocial media reportcontent performance analysisposting time analysis
How to run this
Copy the prompt and paste it into Gemini, then set the aspect ratio to 3:4. No reference photo is needed for this one.
The prompt is written in Markdown with each part under its own heading, which is what the copy action hands over. Change one section at a time, keeping the rest intact, so you can tell which change did the work.