Write one long-form source each week, then cut it into channel-native pieces: a LinkedIn post, a PDF carousel, an X thread, a Threads post, a Substack Note, a short video and a newsletter. Have AI draft each version from the source with that channel's rules, edit every draft by hand, and schedule the set in one tool.

This guide is the distribution layer of the B2B creator-operator playbook: the weekly system that turns one piece of thinking into a week of posts without writing seven things from scratch.

Key points

• Start from one long-form source with a single argument and at least one specific number, tool or example. Every channel version is cut from it.

• Each channel has hard limits: 3,000 characters on a LinkedIn post, 280 on a standard X post and 500 on a Threads text post.

• AI drafts the channel versions from the source, and a human edits each one before it is scheduled.

• A scheduler such as Typefully can hold LinkedIn, X, Threads and Substack Notes drafts in one calendar.

• Automation comes last, once the manual version has run for a few weeks and the prompts are stable.

What does a one-post-to-every-channel system look like?

It is a hub and spokes. One long-form source (a guide, a newsletter issue or a long LinkedIn article) sits in the middle, and each channel gets a version written for how people read there.

The seven outputs in the system:

1. LinkedIn text post. One idea from the source, with the hook inside the first two lines.

2. LinkedIn PDF carousel. The source's framework or steps, one point per slide.

3. X thread. The argument broken into short posts, each one readable alone.

4. Threads post. A shorter, more conversational take on one point.

5. Substack Note. A short observation that links back to the full post.

6. Short video. A 30 to 60 second script built on the strongest example, recorded or made with an AI clone.

7. Newsletter. The source itself, or a tighter edit of it, sent to subscribers.

Each version should point back to the source where the channel allows it. Over a month, the channel posts feed readers back to the source, and the source is the piece that can rank in search and be cited by ChatGPT and Perplexity.

The shift is from writing four separate things each week to capturing once and creating every format from one source.

What are the length and format rules for each channel?

Each channel caps length differently and rewards a different shape. Write to the shape first, then check the limit.

The suggested lengths are working ranges, not rules; the hard limits are the platforms' own. X's help center gives Premium longer posts a cap of 4,000 characters in one set of steps and 25,000 in another, so test before relying on either. Substack does not publish a Notes limit and instead says, "We dare you to try to figure out the character limit."

Here is what one capture can turn into in a week: a newsletter, a thread and daily posts, a LinkedIn post, infographic prompts and images.

How do you pick a source worth repurposing?

Pick a piece that makes one clear argument and contains specifics. A source with a number, a named tool and a step-by-step method gives every channel something concrete to cut from.

Good sources tend to be:

• Answer-first guides. The question is the hook, and the answer block fits a LinkedIn opener.

• Framework posts. A step-by-step method turns into a carousel with little restructuring.

• Data roundups. One table can feed a carousel, a thread and several standalone posts.

If you cannot state the source's argument in one sentence, the channel versions will read as generic, and AI drafting will make that worse.

How does an AI-assisted repurposing pipeline work?

Claude (or another model) drafts each channel version from the source using a fixed brief per channel, and a human edits every draft before scheduling. The AI handles length and format. You handle voice, accuracy and judgment.

The pipeline, step by step:

1. Store the source in one place. I keep sources in Notion, one page per piece, with the one-sentence argument, number and example at the top.

2. Keep a channel brief for each output. Each brief states the hard limit, the length target, the format and two or three of your own best past posts on that channel as examples.

3. Draft each version from the source, not from each other. Give the model the full source plus one channel brief, and ask for one version. Drafting the X thread from the LinkedIn post compounds errors.

4. Edit by hand. Read each draft aloud, cut anything you would not say, and check every number against the source.

5. Run a slop check. Strip banned phrases, recap endings and stacked one-line paragraphs. The guide on how to write with AI without sounding like AI lists the checks.

6. Schedule the set. Load the approved versions into a scheduler and space them across the week.

7. Log what ran. Note which hook and format got replies, and feed that into next week's briefs.

What goes in a channel brief?

A channel brief is a short, reusable prompt. A LinkedIn brief can read like this: "Write one LinkedIn post from the source below. Use one idea only. Put the hook in the first 140 characters. Stay under 1,800 characters. No hashtags in the body. End on a specific question. Match the voice of these three example posts."

The examples matter more than the instructions. A model copies the rhythm of real posts better than it follows adjectives like "punchy."

Where does the human edit matter most?

The hook, the facts and the ending. Models write hooks that sound like other hooks, round specific numbers into vague claims, and close with a summary line. Rewrite the first line yourself, check each figure against the source, and try cutting the last sentence.

How do you schedule the set in one tool?

Use a scheduler that holds every text channel in one calendar. Typefully is the one I use.

Typefully's site says it can "cross-post to X, LinkedIn, Bluesky, Threads, and Mastodon," and lists LinkedIn PDF carousels and natural language scheduling among its features. Its Substack Notes help article says Notes can be published right away or scheduled, and that Substack "can be enabled alongside other supported platforms."

Two limits to plan around:

• Video is not supported on Notes. Typefully's help article states that "videos aren't supported for Substack Notes," so short video goes out from another tool.

• Threads become one Note. The same article warns that "multi-post drafts are combined into a single Substack Note," so write the Note as its own version instead of reusing the X thread.

Typefully also does not post long-form Substack newsletters; per the same article, it supports "Substack Notes only." The newsletter still goes out from your newsletter platform.

For short video, a clipping tool covers the gap. OpusClip describes itself as a tool that "turns long videos into shorts, and publishes them to all social platforms in one click." If you would rather not record at all, the guide to AI video clones for creators covers making the short video from a script.

Can you automate the whole repurposing pipeline?

Yes, but only after the manual version has run for a few weeks and the prompts are stable.

A typical automated setup:

1. Trigger. A Notion database row moves to "Ready to repurpose."

2. Draft. An automation sends the source and each channel brief to Claude and writes the drafts back to the same Notion row, one property per channel.

3. Review. You edit and approve in Notion. Nothing posts without approval.

4. Schedule. Approved drafts are sent to the scheduler as drafts or scheduled posts.

The automation layer can be Relevance AI, n8n or Zapier. They differ in how much control and hosting you want, and Relevance AI vs n8n vs Zapier compares them for this kind of build.

Keep the review step manual: one wrong number on LinkedIn costs more than the minutes auto-posting saves. If you want the loop to improve its own prompts over time, loop engineering for AI workflows covers how to build that in safely.

I run this kind of pipeline for my own content, and the same pattern of intake, drafting, review and logging is what revenue teams need for lead routing, CRM cleanup and pipeline reporting. For Series A and B SaaS teams I build it as a 90-day AI RevOps Sprint. Contact me to get yours scoped.

What does a weekly repurposing cadence look like?

One source per week, written early, with channel versions spread across the following days. Keep writing and editing on separate days.

As an illustrative example with round numbers: a three-hour source plus ten minutes of editing per version comes to about four hours a week.

What should be on your repurposing checklist?

Run this before anything is scheduled:

• The source states one argument, one number and one example at the top.

• Each channel version was drafted from the source, not from another version.

• Each version fits that channel's hard limit.

• The LinkedIn hook fits inside about 140 characters, and the first X post stands alone.

• Every number matches the source, with its source named where it came from outside.

• The ending of each version is a real question or the last concrete point, not a recap.

• Sponsored or affiliate content carries the disclosure each platform requires.

What are the common mistakes when repurposing content?

The most common mistake is posting the same text everywhere.

1. Copy and paste across channels. A LinkedIn post cut down to 280 characters loses its argument. Rewrite from the source.

2. Drafting from a draft. Each AI pass drifts further from the source. Always give the model the original.

3. Skipping the human edit. Unedited AI drafts share the same hook shapes and filler.

4. Ignoring the mobile cutoff on LinkedIn. About 140 characters show on mobile before "see more," per AuthoredUp, so put the hook first.

5. Treating Threads as a copy of X. Threads caps text posts at 500 characters, so a long X thread needs its own shorter version.

6. Automating before the prompts are stable. Automation multiplies whatever the briefs produce.

7. Never reviewing results. Without a weekly log, the briefs never improve.

FAQ

How many platforms should I repurpose to?
Start with LinkedIn, one other text channel and your newsletter. Add channels once editing three versions a week feels routine.

Should I post the same content on LinkedIn and X?
Post the same idea, not the same text. LinkedIn allows 3,000 characters per post and X's standard posts allow 280, so each version needs its own shape.

What is the character limit on Threads?
Meta's Threads developer documentation says text posts are limited to 500 characters. Carousels take 2 to 20 items, and videos can run up to 5 minutes.

Can AI repurpose content for me?
Yes, AI can draft every channel version from your source. A human should still edit the hook, check every number and cut the ending before anything is scheduled.

Which tool schedules LinkedIn, X, Threads and Substack Notes together?
Typefully supports X, LinkedIn, Bluesky, Threads and Mastodon, and also schedules Substack Notes. It does not post long-form Substack newsletters or video to Notes.

How long should a short video repurposed from a post be?
YouTube Shorts can run up to three minutes. For a clip built on one example from a post, 30 to 60 seconds is usually enough.

Want help building the repurposing system behind this?

I work with a small number of B2B teams as a fractional GTM and AI engineer, setting up content systems like this one, from channel briefs and Claude drafting workflows to Notion review queues and scheduling. If you want a second set of hands, see how I work, browse the tools I build on or book a short intro call. Brands with sponsored content enquiries can use the same contact page.

Running revenue at a Series A or B SaaS company? Ask about the 90-day AI RevOps Sprint. I take two clients at a time.

About the author: Kevin Payne is an AI engineer and GTM operator who installs AI workflows for B2B teams. He has 10+ years in B2B SaaS demand generation and RevOps across Techstars, a16z and YC portfolio companies, and his writing has appeared in HubSpot, GetResponse, Outbrain and more.