Verified against Claude · 2026-08-01
Turn your LinkedIn post analytics into one specific next-post hypothesis
Feed in the numbers LinkedIn's own post-analytics panel gives you and get a plain-language read on what actually happened, plus exactly one testable change for the next post — not a re-explanation of the numbers you already have, and no invented or rounded figures.
The prompt
Ready to copy — highlighted parts are example details you can swap.
POST SUMMARY Text-only post, personal story about a missed deadline with a client, no link ANALYTICS AS SHOWN IN LINKEDIN'S POST-ANALYTICS PANEL Impressions: 14,200. Reactions: 310. Comments: 62. Reposts: 8. Discovery: 71% non-follower, 29% follower. COMPARABLE PREVIOUS POST, IF AVAILABLE Last month's carousel post: 9,800 impressions, 190 reactions, 14 comments, 44% non-follower WHAT YOU WANT THE NEXT POST TO DO BETTER More comments from people I could plausibly work with, not just raw reach TIME SINCE THIS POST WENT LIVE 5 days — the post has largely finished circulating DATA DISCIPLINE Treat every number in Impressions: 14,200. Reactions: 310. Comments: 62. Reposts: 8. Discovery: 71% non-follower, 29% follower. as correct exactly as given. Do not recalculate, round, estimate, or infer a number that wasn't provided, even if a rounder or more impressive-sounding figure would make the analysis read more cleanly. If a number needed for a clean read is missing, say what's missing instead of substituting an assumption for it. DELIVER FOUR THINGS 1. VERDICT — one sentence on whether this beat or underperformed Last month's carousel post: 9,800 impressions, 190 reactions, 14 comments, 44% non-follower. If no comparison was given, say so plainly instead of inventing an implicit baseline to compare against. 2. THE SPECIFIC SIGNAL — LinkedIn's Discovery breakdown separates impressions into a follower versus non-follower split, and separately reports reactions, comments, and reposts; these are different signals calling for different next moves. A high non-follower percentage suggests the algorithm pushed this post beyond the existing network, often via topic or hashtag; a high comment-to-reaction ratio suggests the content itself provoked genuine discussion rather than passive scrolling-by likes. Name which specific signal is actually present in Impressions: 14,200. Reactions: 310. Comments: 62. Reposts: 8. Discovery: 71% non-follower, 29% follower. — don't collapse everything into a vague "engagement was good." 3. ONE TESTABLE CHANGE — a single variable to change for the next post: hook style, format, posting time, or topic — never more than one at once, so a result from the next post can actually be attributed to something specific rather than to an unknown combination of simultaneous changes. 4. ONE THING THAT'S ALREADY WORKING — something to deliberately hold constant while testing the variable above, so a real improvement doesn't get accidentally undone by changing something that was already contributing to the result. TIME-CONTEXT CHECK Factor in 5 days — the post has largely finished circulating — a post analyzed six hours after publishing and one analyzed two weeks later are at very different points in their lifecycle, and a verdict on the six-hour version should be stated as provisional, not final, if the comparison post's numbers were captured at a later point in its own lifecycle. GAP HANDLING If the data given genuinely doesn't support a confident read on any of the four items above, say exactly what additional number or context you'd need instead of producing a plausible-sounding guess in its place. OUTPUT The four items in order, each clearly labeled, plus the time-context caveat if relevant, plus any explicit data gaps named rather than papered over.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
LinkedIn's own analytics panel breaks impressions into a follower-versus-non-follower Discovery split and reports reactions, comments, and reposts as separate figures specifically because they represent different underlying behaviors. A non-follower spike points at algorithmic distribution beyond the existing audience, usually topic- or hashtag-driven, while a high comment-to-reaction ratio points at the content itself provoking active thought rather than passive scrolling-by approval. Collapsing these into one "engagement was good" verdict throws away the one piece of diagnostic information the platform actually provides for free. The strict data-discipline rule — treat every given number as correct, never round or estimate a missing one — exists because a plausible-sounding fabricated number is worse than an honest gap. A model asked to interpret performance data will, left unconstrained, sometimes "helpfully" smooth over a missing figure with a reasonable-sounding estimate, and a business decision made on that estimate is a decision made on a number nobody actually measured, which compounds silently across a series of posts if it's never caught. Restricting the recommendation to exactly one changed variable is what turns the next post into an actual experiment rather than a fresh guess. A post that simultaneously changes the hook style, the format, and the posting time, then does better or worse than the last one, produces zero attributable information about which of the three changes mattered, whereas a single-variable change means the next comparison actually teaches something usable for the post after that. Requiring one thing to deliberately hold constant protects the experiment from its own success: a common failure after a strong post is to change everything for the next one out of enthusiasm, which accidentally discards whatever was already working alongside whatever gets newly tested. Naming the thing to protect is what keeps a real improvement from being undone by an unrelated, well-intentioned change made in the same pass.
What you get back
Verdict: this outperformed the comparison post on both reach and depth — impressions up 45%, but the more telling number is comments per impression, which is roughly 3x the carousel post's rate. Signal: 71% non-follower reach plus a high comment count together suggest this spread via genuine discussion, not just algorithmic hashtag push alone — the story format is doing real work here, not just topic luck. One variable to test next: try the same personal-story structure on a topic outside your usual niche, to see if the discussion-driving effect is about the format or about this specific story. Don't change: keep it text-only and link-free — you have no evidence a link or carousel format would have done better here, and changing it now would confound the next test.
Verified against
Claude Sonnet 5 · 2026-08-01
ChatGPT GPT-5.1 · 2026-08-02
Changelog
- 2026-08-02 — Initial publish for the rewritten linkedin category, scoped to interpretation-only after testing showed models will otherwise "helpfully" estimate missing metrics instead of asking for them.
Need this built into your business?
If a prompt isn't enough — Google Ads management, built and maintained for you — that's Scult's day job.
EXPLORE GOOGLE ADS MANAGEMENT
