A content repurposing strategy only works if the derivative list is written at brief stage, before the source piece is outlined. This covers the twelve-asset slate I plan off one substantial idea, which formats feed which channels, what I cut, and why the derivatives shouldn’t become indexable pages of their own.
I plan around twelve assets off every substantial idea and decide all twelve before I outline the source piece. At a B2B eMobility company in 2024, that’s how one person covered eight buyer segments across 39 countries in seven months: 12+ articles, 15+ web pages, two whitepapers and a case study, plus 70+ social posts pulled from the same research rather than researched separately. The content repurposing strategy lives in the brief, otherwise the derivatives get written off the finished article, and then they read like a copied introduction, which is what most of them are.
Atomisation, the term the industry uses for this, means breaking one source asset into smaller channel-native pieces. My version has one constraint the standard version doesn’t: I repurpose the research, never the text.
What this assumes, and what I won’t do with it
This runs on a lean function, usually one person, in B2B with a considered purchase, and it assumes you have something substantial to multiply: original data, a position of your own, named sources, evidence somebody had to go and gather. Applied to a thin post, the multiplier faithfully produces twelve thin assets. It also assumes you own the calendar, because where an editor hands you a list of titles, the derivative planning comes too late to change the source piece and half the value goes with it.
The refusal: I could point a model at any published article of mine and have it generate the full twelve derivatives in about twenty minutes. I don’t, because derivatives generated from finished prose inherit the prose, arriving as compressed paragraphs with the transitions removed, and the LinkedIn version of an article opening is the most recognisable format failure on that platform.
Why a content repurposing strategy beats publishing more
The pressure to produce more shows up in the survey data. In the Content Marketing Institute’s 2026 B2B research, a survey of 1,015 B2B marketers fielded with MarketingProfs in mid-2025, resource constraints came out as a top-three challenge for 39%, and 28% named producing enough quality content for the organisation’s needs. Human resources sat last on the 2026 investment list at 9%, behind AI tools at 45%, so the same small team feeds more channels with more tooling and no more people.
AI hasn’t closed that gap. The same research found 87% of marketers using AI for content creation reported better productivity while only 39% saw content performance improve, which tells you the extra output isn’t working. Ann Handley summed it up:
AI is like giving every marketer a turbo-charged typewriter.
Typing faster doesn’t produce a second idea, and multiplication is cheap for a reason unrelated to tooling: the expensive part of a piece is the research, the interviews and the argument, all of which happen once, so everything downstream is an editing job against material that’s already been checked.
1. Write the derivative list before the outline
The brief has a section listing every asset that will come out of the piece, filled in before I write a word of it. Knowing I owe a slide set changes which comparison I build, and if there’s a sales one-pager on the list I answer the main objection head on instead of in passing. Here’s the slate off a research-led article, the highest-yield source format I run:
| Asset | Channel | What it reuses | Goes out |
|---|---|---|---|
| The source article | Blog | Everything | Publish day |
| 4 to 6 LinkedIn posts | Personal social | One claim each, stated on its own | Publish, then weekly |
| 1 newsletter issue | The argument, rewritten shorter | +3 days | |
| 1 comparison table, standalone | Social carousel | The table without the prose | +1 week |
| 1 sales one-pager | Enablement | The objection section | +1 week |
| 1 FAQ block | Product or help pages | The three Q&A pairs | +2 weeks |
| 2 to 3 quote cards | Social | Named sources already cited | Rolling |
| 1 slide set | Webinar, speaking | The structure, expanded | +1 month |
| 1 short video script | Video | The single sharpest claim | +1 month |
| Refresh with new data | Blog, same URL | The whole piece, updated | +6 to 12 months |
Twelve is what a substantial idea supports, not a target, and some pieces only yield five.
2. Split the argument into entry points, not extracts
An extract is a paragraph lifted out; an entry point is a different way into the same argument, which is what makes a derivative worth existing in its own channel. Off one article I’ll usually find the counter-argument standing alone, one statistic with its source and nothing around it, the failure case, the diagnostic test a reader can run in thirty seconds, and what I got wrong before I got it right. Each of those needs its own opening sentence, written natively.
3. Keep one canonical page and let everything else point at it
Here I disagree with most published repurposing advice, including guides I otherwise rate, like Cognism’s and Blend’s, which claim repurposing improves search performance because each new format targets new keywords and earns backlinks. For blog-to-blog derivatives, that’s how you end up competing with yourself for one query, splitting authority across two thin pages instead of stacking it on a good one.
The AEO and LLMO case points the same way and it’s stronger. AEO (Answer Engine Optimisation) is the work of being the answer a search engine serves directly; LLMO (Large Language Model Optimisation) is the work of being a source an AI model cites. The Princeton GEO study (Aggarwal et al., ACM KDD 2024, across 10,000 queries) found citation lift came from attributed statistics, quotes from named sources, and citing external sources, worth up to 115% for lower-ranked content, while extra words did nothing. Twelve derivatives are twelve copies of the same substance, so they add no citability the source page doesn’t already have.
With 58.5% of US Google searches ending without a click and 73% of B2B buyers using AI tools in purchase research, that page carries its answer in full anyway. The derivatives do a different job: they get the argument in front of people who’ll never search for it, which builds the salience that makes someone type your name later. How the citable page gets built is in the AEO/LLMO Content Optimisation Framework and, at sentence level, in the tactical guide to AI-citable content.
4. Sequence the releases, then kill what needs new research
A piece published once and shared once on the day reaches whoever happened to be online that morning. Everything in the table has a lag against it, and the lag is the first casualty when a deadline moves, which is why it sits in the brief with a date instead of in someone’s memory. The full checklist for getting a piece seen without budget is in content distribution for B2B SaaS with no budget.
The kill rule is simpler: any derivative that needs research the source piece didn’t already do gets cut, or it gets promoted to a source piece of its own with its own slate. A “customer quotes” carousel that requires four new customer emails is a project wearing a derivative’s clothes, and it’s how a twelve-asset plan becomes a six-week one.
What twelve assets looked like on two product launches
The clearest version I’ve run was two product launches at that eMobility company, where launch content accounted for 35% of total website traffic during the campaign windows. The research happened once and everything downstream came off it: segment-specific web pages, a whitepaper, social sequences across three platforms, and the sales material. Across five months the social work alone ran to 70+ posts, impressions up 24% and reach up 217% against the pre-launch baseline. Mechanics are in the product launch piece and the 217% reach turnaround.
Before I worked this way, at a climate tech company, I produced 60+ pieces over 18 months as mostly separate efforts. Similar volume, far worse hours, and none of it reinforced anything else, because nothing had been planned to.
Where this currently breaks
Video is the weak link and has been for two years. I plan the script, I write the script, and it sits there, because the production step needs a skill I’m still slow at, so the slate ships eleven assets. I’m testing whether cutting straight from the slide set gets me past it.
The refresh line at the bottom of the table is the other one. Updating an existing URL returns more than a new post on the same topic and it’s the least satisfying work in the slate, so it slips first when the calendar tightens. Tracking it as a review date, not a good intention, helps and doesn’t fix it. What’s worth refreshing at all is a call I make with the Content Prioritisation Framework.
What the multiplication frees up is the research budget. When one idea has to carry twelve outputs, three days of interviews and data stops looking extravagant and starts looking like the cheapest decision available, which is the argument I’d make for original research to anyone still funding it by the piece.
Frequently asked questions
A content repurposing strategy is a documented plan for turning one substantial source asset into multiple channel-native derivatives, decided before the source asset is written and not after it’s published. It names each derivative, the channel it serves, which part of the source research it draws on, and when it goes out. The version I run yields around twelve assets from one research-led article and treats the research as the reusable material, never the finished sentences.
Resharing puts the same asset in front of an audience again with a new caption. Content repurposing rebuilds the argument for a different channel with that channel’s constraints in mind, so a LinkedIn post derived from an article has its own opening line and makes one claim instead of summarising nine. The test is whether the derivative makes sense to someone who never sees the source piece.
Not for derivatives of the same argument. Publishing several blog pages off one idea splits topical authority across thin pages and puts them in competition for the same queries, where one deep, well-sourced, cross-referenced page earns more in both search and AI citation. Keep one canonical page, put the derivatives in channels that don’t compete with it (social, email, enablement, slides), and link everything back.
As many distinct entry points as the argument genuinely has, which for a research-led B2B piece with original data and named sources usually works out at ten to twelve, while a thinner opinion piece honestly yields four or five. Forcing a fixed number produces derivatives with nothing in them, and a content repurposing strategy that manufactures assets to hit a count has recreated the volume problem it was meant to solve.
