LinkedIn

Verified against ChatGPT · 2026-08-12

Turn a client result into a LinkedIn case-study post that leads with the number, not the client's name

Structures a client or project win into a LinkedIn post around the before/after numbers and the specific decision that caused the change, built to work as credibility-building proof rather than a thinly-veiled ad.

ChatGPT (GPT-5.1)6 fillable variables

The prompt

Ready to copy — highlighted parts are example details you can swap.

PHASE 1 — CONFIRM THE STORY HAS A REAL BEFORE/AFTER
Before drafting anything, check whether what I've given you actually has a measurable before-state and after-state. If it doesn't, tell me what's missing and what to ask the client or pull from records before this can become a credible case-study post — don't draft around a vague improvement.

CLIENT OR PROJECT (anonymized if needed)
A 40-person logistics company that came to us with a broken lead-routing process.

THE STARTING NUMBER OR SITUATION
Average lead response time was 26 hours, and only 12% of inbound leads got a same-day reply.

THE RESULTING NUMBER OR SITUATION
Average lead response time dropped to 45 minutes, with 91% same-day reply rate.

THE SPECIFIC DECISION OR CHANGE THAT CAUSED IT
We rerouted inbound leads by territory automatically instead of a single shared inbox one person checked manually.

TIMEFRAME
6 weeks from implementation to the after-numbers being stable.

PERMISSION TO NAME THE CLIENT
Cannot name them by contract, must anonymize as 'a logistics company.'

PHASE 2 — DRAFT THE POST (only after phase 1 passes)
Lead with the after-number as the hook, stated plainly, before any context about who the client is — the number earns attention faster than a company name most readers don't recognize does. Follow with the before-state so the gap is immediate and concrete, then name the single specific intervention that caused the change — not a vague "we optimized their strategy," but the actual decision, tactic, or change made. State the timeframe explicitly since an unstated timeframe makes any number impossible to evaluate — a 40% increase over 3 years reads very differently than over 3 weeks. If client_permission indicates you cannot name the client, anonymize consistently throughout ("a mid-size logistics company" used the same way every time it's referenced) rather than accidentally including an identifying detail elsewhere in the post. Close with what this means for the specific type of reader who'd have the same starting problem, phrased as a takeaway they could apply, not a sales pitch to hire you — the credibility is supposed to do the selling implicitly.

WHAT NOT TO DO
Do not inflate or round the numbers I gave you to sound more impressive. Do not add a hard sales CTA ("DM me to get these results too") unless I explicitly ask for one — a case study that reads as proof outperforms one that reads as an ad, and adding a pitch at the end undercuts the credibility the numbers just built.

OUTPUT FORMAT
1. A note on whether phase 1 passed, and what's missing if it didn't.
2. The drafted post, formatted with LinkedIn-appropriate line breaks.
3. One alternate hook line using the same numbers, in case the first doesn't land.

Customize

Optional — swap in your own details for the highlighted parts above.

Why this works

The gating phase 1 check exists because a model asked to write a case-study post will draft something plausible-sounding even from a vague 'we helped them improve' input, since fluent prose generation doesn't require the underlying claim to be verifiable — building in an explicit checkpoint that refuses to proceed without a real before/after number forces the actual evidentiary gap to surface before word-smithing hides it, which matters specifically because a case study's entire persuasive value collapses if a reader can tell the numbers are vague or unverifiable. Requiring the timeframe to be stated explicitly, and flagging that an unstated timeframe makes a percentage meaningless, targets a specific reasoning gap general-purpose instruction following doesn't automatically fill: a percentage change is mathematically incomplete information without a denominator of time, but a model asked simply to 'write a case study post' has no built-in incentive to demand that context unless it's named as a requirement, since the sentence 'we improved response time by 40%' is fluent and complete-sounding on its own even though it's unverifiable as stated. The instruction against a hard sales CTA reflects an actual behavioral pattern rather than a stylistic preference: models tend to close persuasive or promotional content with an explicit call to action by default, because that's the standard shape of marketing copy in their training distribution, but a results-driven case study post specifically works through implicit credibility rather than direct pitching, so the close has to be deliberately overridden rather than left to the model's default marketing-copy instinct.

What you get back

Phase 1: passed — concrete before/after with timeframe. Post: "91% of inbound leads now get a same-day reply. Six weeks ago it was 12%. The fix wasn't a new CRM or a bigger sales team — it was killing the shared inbox one person was checking manually, and routing leads by territory automatically instead. If your team's average response time is still measured in hours, the bottleneck probably isn't effort. It's the routing."

Verified against

ChatGPT GPT-5.1 · 2026-08-12

Changelog

  • 2026-08-12 — Initial publish, verified against ChatGPT GPT-5.1.

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
Write a first line that survives LinkedIn's "see more" cutoffGenerate six hook-line options for a LinkedIn post, each built on a different mechanism and tested against the actual character count where LinkedIn truncates a post, so every option reads as a complete thought before the cutoff rather than a fragment that only resolves after the tap.ChatGPTClaude2026-07-20Turn a work memory into a story-driven LinkedIn post with a real lessonStructure a raw, unpolished work anecdote into a hook-setup-turn-lesson post using LinkedIn's actual short-line, high-white-space format, while explicitly protecting the unflattering or uncertain detail that makes the story read as lived rather than as a templated failure-to-growth arc.ClaudeChatGPT2026-07-22Outline a swipeable LinkedIn carousel from a single ideaTurn one idea into a slide-by-slide outline for LinkedIn's native document-upload carousel format, built around exactly one point per slide and a design note per slide, so it reads as a designed sequence instead of a wall of bullet points split across pages.ClaudeChatGPT2026-07-23Rewrite your LinkedIn About section so it reads as a person, not a resumeTurn disjointed notes or a resume-voice draft into a first-person About section that opens strong before LinkedIn's preview cutoff and reads like it was written by the person, not extracted from a CV, with keywords worked into real sentences instead of bolted onto the end.ClaudeChatGPT2026-07-24
All LinkedIn prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY