One substantial blog post already contains a month of social content. The mistake is treating repurposing as cutting: slicing the post into quote graphics and link captions that all say the same thing in shorter clothes. The stronger method is decomposition. Pull out every claim, number, example, objection, process step, and quote, and treat each one as the seed of a standalone post with its own job.
A 1,200 to 1,800 word post that argues something real usually contains 15 to 25 seeds. Some die in editing, so plan on 12 to 20 published posts across four weeks. The same method works on a launch announcement or a meaty changelog entry, because those are arguments too: something changed, for a reason, with evidence.
The six seed types
Read the post once with a highlighter and sort what you find into six buckets. Each bucket produces a different shape of social post, which is why decomposition beats excerpting. Excerpts all inherit the blog post's voice and pacing. Seeds get rewritten for the platform they land on.
| Seed type | What to look for | The post it becomes | Typical yield |
|---|---|---|---|
| Claims | Any sentence you would defend in an argument | A standalone opinion post built to start replies | 3 to 5 |
| Numbers | Results, costs, timelines, counts, before-and-after pairs | A short post about one figure and what it changed | 2 to 4 |
| Examples | Customer stories, specific scenarios, screenshots of real work | A mini case study with the lesson stated plainly | 2 to 3 |
| Objections | Counterarguments you answered, tradeoffs you accepted | A steelman-then-answer post; strong on LinkedIn and Reddit | 2 to 3 |
| Process steps | Any numbered how-to buried in the prose | A checklist post or a thread where each step advances | 3 to 5 |
| Quotes | Customer emails, support tickets, your own sharpest sentence | A quoted line with one paragraph of context | 1 to 3 |
Worked example: decomposing a free-plan postmortem
Imagine a founder publishes a 1,600-word post titled 'Why we removed our free plan.' It argues the free tier attracted users the product was never built for, shows the support math behind the decision, walks through the migration, and answers the loudest objection. The post is invented for this exercise; the structure is what matters.
A single read produces roughly 17 seeds:
- Claims: free users were not future customers here; a free plan is a pricing decision, not a marketing decision; support load is a product signal, not a cost line.
- Numbers: the share of support tickets that came from free accounts, the free-to-paid conversion rate over six months, the three weeks the migration took.
- Examples: the customer who upgraded the day the email went out; the free user whose export request revealed a missing feature.
- Objections: 'you will lose your top of funnel' and 'every competitor has a free plan'.
- Process steps: announce with a date, open a grandfather window, ship export tooling, rewrite the pricing page.
- Quotes: one line from a customer reply, plus the post's own sharpest sentence.
Seed from the fictional post
Most of our support volume came from accounts that had never paid us and, by their own description, never would. We were staffing a help desk for a different product.
X post from that seed
We audited our support queue before killing the free plan. Most of the volume came from accounts that would never convert. A free tier is not marketing spend if it buys you the wrong users' problems.
The excerpt trap
New blog post! Why we removed our free plan. Link below, would love your thoughts!
The four-week distribution calendar
Order matters. Lead with the strongest claims and numbers while the argument is fresh in your own head, because those posts generate the replies that tell you which seeds deserve a second angle. Save process steps for later weeks; a checklist is useful to people who never saw the announcement. The link itself goes out twice in the month, not twelve times.
| Week | Posts | Seeds to use | Rule for the week |
|---|---|---|---|
| Week 1 | 4 to 5 | The core claim as an announcement, the sharpest number, one objection answered | The only link-forward week; the link rides in one or two posts, never all |
| Week 2 | 3 to 5 | Both examples and a second claim, rewritten per platform | Reply-heavy week; the examples invite people to compare their own |
| Week 3 | 3 to 5 | Process steps as a checklist or thread, plus one quote post | Must be useful to a reader who never saw week one |
| Week 4 | 2 to 4 | The remaining objection, a follow-up on the numbers, one link repost to slower channels | Publish follow-up numbers only if they are real by then |
When repurposing turns into spam
Repurposing reuses ideas, not captions. The moment two communities receive the identical text, you have crossed from distribution into noise, and community platforms notice first. Reddit's spam guidance (reviewed August 26, 2026) tells users to contribute authentically and warns against posting the same content across communities; a decomposed seed only belongs on Reddit if it is rewritten as a contribution to one subreddit's actual conversation.
- One seed, one post per platform. Do not run the same seed twice on the same channel within the month.
- The link earns two outings. Every other post must stand alone with the useful part inside it.
- If a post only makes sense with the blog post open in another tab, it is an excerpt, not a post.
- Never push identical captions into multiple communities or groups; rewrite for one, or skip it.
- Stop when the seeds get thin. Twelve strong posts beat twenty that trail off into paraphrase.
Where this fits in the system
Decomposition is the supply side of a founder content system. The blog post gives you a month of raw material; a batching session turns the seeds into scheduled drafts in one sitting, and the platform rewrite is what keeps twenty posts from sounding like one press release. If your source is a product update rather than a blog post, the same seed thinking applies at smaller scale.