Flink Linkedin Carousel

---
name: flink-linkedin-carousel
description: Builds a LinkedIn Carousel Outline of 6 to 10 text slides from your own post or client framework, one idea per slide, slide one the claim, each slide readable without you there, ready to export as a PDF document post. Use for "run flink-linkedin-carousel", "turn this post into a carousel", "make slides from my framework", "a PDF post people can save", "carousel outline for LinkedIn", "document post from this idea", part of the Claude Playbook for Authentic LinkedIn Presence Pack by Polar Bear.
---

# LinkedIn Carousel

## When To Use
A post or a framework you use with clients deserves a version people can save and pass on. You want slides that make sense when someone forwards them to a colleague who never saw your post.

## When Not To Use
If you are still deciding which pieces of a talk or article to reuse, run Content Repurposing Plan first. If the idea fits in a few short paragraphs, keep it as a post; a carousel of one thin idea feels padded.

## Inputs
- The post, framework or notes the carousel comes from, in your words
- Any number or client example in it, with its source from your Claim and Permission Check
- Your Personal Voice Guide, if you have one
If you have none of this, I start from the idea in three sentences and mark the output as a first draft.

## Approach
Situation, complication, question, then the answer, from Barbara Minto's pyramid principle, with the point first. Each slide carries one idea and a full-sentence headline, so the deck reads without a speaker. In Claude Slides (beta) you can lay it out and export to PDF or PowerPoint; in any chat you get the outline to build yourself. The failure mode is the slide with a two-word title and a stock photo: forwarded on its own, it says nothing.

## Workflow
1. I ask three questions: what is the one claim the carousel makes; who will read it when it is forwarded; is there a number or client example, and is it cleared for this channel.
2. Slide one: the claim, point first, as a full sentence a reader could agree or disagree with.
3. Slides two and three: the situation the reader is in, then the complication that makes it hard. Both in your clients' words where you have them.
4. Answer slides: one idea each, a full-sentence headline and a few lines of body. If a slide needs two headlines, it becomes two slides; if the count passes ten, I show you which ideas to merge or cut, and you choose.
5. Forward test: I read each slide alone and flag any that only makes sense after the one before.
6. Last slide: a plain summary of the claim and its parts. No "comment below", no "follow for more". Text only: no stock images, no invented charts or figures.
7. You build or adjust the slides, export a PDF, and upload it yourself as a document post. I never post.

## Output Format
```markdown
# LinkedIn Carousel Outline
## The claim
[One sentence, point first]
## Slides
| Slide | Headline (full sentence) | Body (a few lines) | Source |
|---|---|---|---|
| 1 | [the claim] | [one line of why it matters] | [your post or framework] |
| 2 | [situation] | [body] | [source] |
| 3 | [complication] | [body] | [source] |
| 4 | [answer, idea one] | [body] | [source] |
| [n] | [plain summary] | [body] | |
## Forward test
| Slide | Reads alone | Fix |
|---|---|---|
| [n] | [yes / no] | [what to change] |
## Decision
[You decide the final slide count, build and export the PDF, and post it as a document post, by [date].]
```

## Done When
- There are 6 to 10 slides, each with one idea and a full-sentence headline.
- Slide one states the claim; the last slide summarises it plainly.
- Every slide passes the forward test.
- Every number or client example traces to your source and permission.

## Quality Bar
- One idea per slide; nothing added that is not in your post or framework.
- Text only; no stock images, decorative charts or invented data.
- Client examples anonymised unless the client agreed to LinkedIn.
- No call to comment, no engagement bait.
- Built from your post or framework; Claude never invents a claim, number or client, and you upload it yourself.

## Next
Run flink-warm-conversations-list (Warm Conversations List) to choose who you would like to talk to when people respond.

## About the makers

This pack is made by Polar Bear, a consultancy built by ex-McKinsey founders with a dream to make AI work for People, not instead of them. We help our clients build people systems and AI-first ways of working, and we run our own company on Claude. If your team has outgrown the self-serve version, message Pauline (linkedin.com/in/paulinebertry).