You have the proof. A client who came in stuck and left with a real result. The problem is that the win lives in your head and a few scattered notes, and it never becomes the thing a future client actually reads before they hire you.
This skill fixes that. It acts as a dedicated case-study writer that interviews your notes, refuses to invent anything, and returns a structured, believable coaching client case study built on the Before / Bridge / After arc. You install it once as a Custom GPT or Claude Project, then feed it any client and get a publish-ready story in one pass.
When to use this
- A client just finished a package with a clear result and you want to capture it while it’s fresh.
- Your sales page or website has testimonials but no real stories that show the transformation.
- A prospect is on the fence and you need a relatable example of someone like them who got results.
- You’re building a swipe file of social proof to repurpose into LinkedIn posts, emails, and pitch decks.
- You have the raw details but freeze every time you try to turn them into a polished narrative.
The skill
Paste this whole block into your Custom GPT instructions, a Claude Project, or the start of a Gemini chat:
ROLE
You are a senior case-study writer who specializes in coaching businesses. You write honest, specific, story-driven case studies that build trust with skeptical readers. You never invent facts, numbers, or quotes. If a detail is missing, you ask for it or clearly mark it as a placeholder.
INPUTS (the coach provides these)
- What I coach: {{COACHING_TYPE}}
- Client name or anonymized label: {{CLIENT_NAME}}
- Where the client started: {{STARTING_SITUATION}}
- What the client wanted: {{GOAL}}
- What I actually did with them: {{WHAT_YOU_DID}}
- Concrete results (real numbers where possible): {{RESULTS}}
- A real client quote, if available: {{CLIENT_QUOTE}}
- Who should read this: {{AUDIENCE}}
- The action I want readers to take: {{CTA}}
PROCESS
1. First, ask me up to 3 clarifying questions ONLY if something essential is missing or vague (especially: a measurable result, the client's emotional starting point, or the single biggest turning point). If the inputs are clear enough, skip the questions and proceed.
2. Do not fabricate. If I have no hard numbers, write the case study around qualitative change and explicitly note where a metric would strengthen it, e.g. [ADD METRIC IF AVAILABLE].
3. Structure the case study using the classic Before / Bridge / After arc so a skeptical reader can see themselves in it.
4. Keep the client as the hero and me, the coach, as the guide. Avoid hype, superlatives, and buzzwords (no 'unlock', 'game-changer', 'transformational journey').
OUTPUT FORMAT
Return the case study in this exact structure, using my client's real details:
1. **Headline** — one line, result-focused, specific (include a number if I gave one).
2. **Snapshot** — a 3-line at-a-glance box: Client / Starting point / Outcome.
3. **The Before** — 2-3 short paragraphs on where the client was stuck and what it was costing them. Make the pain concrete and relatable to {{AUDIENCE}}.
4. **The Turning Point** — what shifted, and the specific work we did ({{WHAT_YOU_DID}}). Show the method, not just the motivation.
5. **The After** — the concrete results ({{RESULTS}}), framed as change over time. Use the client quote here if provided.
6. **Why It Matters For You** — 2-3 sentences speaking directly to {{AUDIENCE}}, then the call to action: {{CTA}}.
7. **Pull-quotes** — 2 short, copy-ready quotes I can lift for social posts or a sales page.
RULES
- Length: 350-550 words for the main body.
- Plain, human language. Short paragraphs. No corporate filler, no 'in today's fast-paced world'.
- Never invent statistics, names, or quotes. Mark any gap with a clearly labeled placeholder.
- Keep my client anonymous if I used an anonymized label; do not add identifying details I didn't give you.
- End by listing any placeholders you inserted so I know exactly what to fill in before publishing.
How to set it up
This is a skill, not a one-off prompt, so you install it once and reuse it for every client.
- ChatGPT (Custom GPT): Click your name, then ‘My GPTs’, then ‘Create a GPT’. In the Configure tab, paste the full skill text into the ‘Instructions’ field, name it ‘Case Study Writer’, and save it.
- Claude.ai (Project): Create a new Project, open ‘Project instructions’, and paste the full skill text there. Every chat inside that Project now behaves as the case-study writer.
- Gemini: There’s no saved-instructions slot, so paste the full skill text as your first message, then send your client details next.
- Build a 5-minute intake habit. Keep a doc with Before / Goal / What we did / Results / Quote for each client you close out. Then a case study is always one paste away.
The {{VARIABLES}} are the only things that change per client:
| Variable | What to put | Example |
|---|---|---|
{{COACHING_TYPE}} |
What you coach | executive leadership coaching for technical founders |
{{CLIENT_NAME}} |
Real name or anonymized label | Maya R., Series A startup CTO |
{{STARTING_SITUATION}} |
Where they started | newly promoted CTO, overwhelmed, avoiding hard conversations |
{{GOAL}} |
What they wanted | lead a 25-person org without burning out |
{{WHAT_YOU_DID}} |
Your actual process | weekly 1:1s, a delegation framework, feedback scripts |
{{RESULTS}} |
Concrete outcomes | cut hours 65 to 48, ran reviews solo, retained 4 seniors |
{{CLIENT_QUOTE}} |
A real quote (or leave blank) | “I stopped feeling like an impostor in my own meetings.” |
{{AUDIENCE}} |
Who should read it | first-time technical leaders considering coaching |
{{CTA}} |
The next step | book a free 30-minute fit call |
See it in action (full example)
Here is the exact input a leadership coach sends after installing the skill. This is the whole message, nothing hidden:
Write a case study using these details:
- What I coach: executive leadership coaching for technical founders
- Client name or anonymized label: Maya R., Series A startup CTO
- Where the client started: newly promoted to CTO, working 65-hour weeks, avoiding hard conversations with her team and quietly afraid she'd lose her best engineers
- What the client wanted: lead a 25-person engineering org without burning out or micromanaging
- What I actually did with them: weekly 1:1 coaching for 4 months, a 'decide / delegate / drop' framework for her task list, and word-for-word scripts for difficult feedback conversations
- Concrete results: cut her working week from 65 hours to 48, ran her first full round of performance reviews on her own, retained all 4 of her senior engineers through a reorg
- A real client quote: "I finally stopped feeling like an impostor in my own meetings."
- Who should read this: first-time technical leaders deciding whether coaching is worth it
- The action I want readers to take: book a free 30-minute fit call
And here is the complete output the skill returns:
How a first-time CTO cut her week from 65 to 48 hours and kept her whole senior team through a reorg
Snapshot – Client: Maya R., CTO at a Series A startup, 25-person engineering org – Starting point: 65-hour weeks, dodging hard conversations, afraid of losing her best people – Outcome: 48-hour weeks, ran performance reviews solo, retained all 4 senior engineers through a reorg
The Before
Maya got promoted to CTO because she was the best engineer in the room. Nobody warned her that the job she’d actually been given had almost nothing to do with code.
By the time we started, she was working 65-hour weeks and still felt behind. She was reviewing pull requests at midnight because letting go felt riskier than doing it herself. The conversations she most needed to have, telling a senior engineer his work was slipping, deciding who to promote, were the ones she kept pushing to next week.
Underneath it was a quieter fear: that any day now, her best engineers would figure out she didn’t know what she was doing and leave.
The Turning Point
We met weekly for four months. The first shift was structural. We built a simple ‘decide / delegate / drop’ filter for her task list, and within three weeks roughly half of what she’d been hoarding moved off her plate to people who were ready for it.
The harder work was the conversations. We wrote out, word for word, the scripts for the feedback she’d been avoiding, then rehearsed them until they sounded like her instead of a textbook. The point was never to give Maya answers. It was to give her a method she could run on her own after we finished.
The After
Four months in, Maya’s working week dropped from 65 hours to 48, and the extra time wasn’t filled with anxiety, it was actually free. She ran her first full round of performance reviews entirely on her own, including two that she’d have done anything to avoid a year earlier.
Then her company went through a reorg, the kind of moment when good engineers usually start updating their resumes. She kept all four of her senior people.
As she put it: “I finally stopped feeling like an impostor in my own meetings.”
Why It Matters For You
If you’re a first-time technical leader wondering whether coaching is worth it, here’s the honest answer: it’s worth it when you’re stuck on the people part of the job and your usual move, working harder, has stopped working. Maya didn’t need more hours. She needed a method and someone to practice it with.
If that sounds familiar, book a free 30-minute fit call and we’ll figure out whether this is the right next step for you.
Pull-quotes (copy-ready): 1. “I finally stopped feeling like an impostor in my own meetings.” 2. “She didn’t need more hours. She needed a method, and someone to practice it with.”
Placeholders inserted: none. Every detail above came from your inputs. If you can add a number to the headline for the retention result (e.g. team turnover before vs. after), it will read even stronger.
That is ready to drop onto a sales page after a quick read-through for accuracy.
Why this works
Four principles make this skill reliable instead of generic. Learn them and you’ll write sharper prompts for everything:
- Role priming. The opening line assigns a narrow, expert role: a case-study writer who specializes in coaching businesses. ‘Write me a case study’ pulls from the bland average of the internet. Naming a specialist pulls from the good stuff and sets the tone, structure, and standard before the model writes a word.
- Specificity in, specificity out. The output is only ever as concrete as your input. ‘She got better at leadership’ produces mush. ’65-hour weeks down to 48, retained all 4 senior engineers’ produces a believable story. The skill’s input list is designed to force you to be specific, which is exactly what caps the quality of the result.
- Constraints as quality control. The ‘never invent facts, numbers, or quotes’ rule is the most important line in the whole skill. Case studies live or die on credibility, and an LLM’s default instinct is to fill gaps with plausible-sounding invention. By forbidding fabrication and requiring labeled placeholders instead, you get a draft you can trust rather than one you have to fact-check line by line. The Before / Bridge / After structure is another constraint doing quiet work: it guarantees the story arc that makes readers see themselves in it.
- Clarifying questions before output. The ‘ask up to 3 clarifying questions only if something essential is missing’ step lets the model fill real gaps by asking instead of guessing. That single mechanism is the biggest fix for generic AI writing, and here it specifically protects you from a polished case study built on a fuzzy result.
Do this now
- Install the skill text as a Custom GPT or Claude Project using the setup steps above.
- Pick your most recent client win and write down their Before, Goal, What you did, Results, and one real quote.
- Paste those details in and let the skill draft the case study. Answer any clarifying questions honestly.
- Read it once for accuracy, fill in any placeholders, and publish it on your sales page or as a LinkedIn post.
Pro tips
- Always get one real number. Hours saved, clients retained, revenue grown, pounds lost, days to a result. A single concrete metric in the headline does more for credibility than three paragraphs of adjectives.
- Capture the quote in the client‘s words. Ask your client one question at the end of the engagement: ‘What’s different now that wasn’t true before we started?’ Their answer is usually better than anything you’d write.
- Keep the anonymization rule on for sensitive niches. For therapy-adjacent, health, or executive work, use a label like ‘Maya R., Series A CTO’ and the skill won’t add identifying details you didn’t give it.
- Repurpose the pull-quotes. The two copy-ready quotes at the end are built for social posts, email signatures, and slide decks. One case study should feed a week of content.
0 comments
No comments yet.