Repurposing has always been the advice everyone agrees with and nobody executes, because manual reformatting is a weekend per asset. AI tooling collapsed that cost — script generation, text-to-speech, captioning, clipping — to the point where "one article, ten assets" is an afternoon. The pipeline below is the one that actually survives contact with a real schedule.
Pick the source worth multiplying
Not every article deserves ten descendants — the ones with original data, clear steps or strong opinions do. Repurposing amplifies; it doesn't rescue. Choose the piece that already earned traffic, then work the pipeline top-down.
The pipeline, in order
1. Video script: feed the article, ask for a 3-minute spoken script with hooks — spoken rhythm, not read-aloud prose. 2. Voiceover or talking-head: TTS for speed, your face for trust — pick per piece, mix freely. 3. Long video (3-8 min): slides plus voiceover is the sustainable format; production polish is not the bottleneck, consistency is. 4. Three shorts: the hook, the best insight, the counterintuitive claim — vertical crops with captions (auto-captioning is the single highest-ROI tool in the chain). 5. Thread/carousel: the article's skeleton, one insight per slide. 6. Email version: the piece condensed to a two-minute read with one takeaway. 7. Infographic: the article's key numbers as a single image. 8. Audio version: the TTS track as a podcast episode — costs nothing after step 2. 9-10. Q&A and quote graphics from the comment section the video generates.
The honest constraints
Review every generated artifact — scripts misquote your own article with total confidence. Keep one canonical voice: the article is the source of truth, everything else cites it. And measure per format for a quarter: most sites discover two or three formats that earn disproportionate attention for their audience, and the professional move is dropping the rest. Ten assets is the menu, not the mandate.
