A finished video is not one piece of content — it's a container holding several. The most valuable thing inside is usually a carousel: the same idea, restructured for swipers instead of watchers, earning saves instead of watch time. Repurposing one takes 15 minutes once you have the method (and near zero once you have the system). Here are both.
Why this specific conversion is worth it
Video→carousel is the highest-value repurposing move because the formats are complementary rather than redundant:
- Different discovery surfaces. Reels distribution and feed/Explore carousel distribution are separate lanes; posting both doubles your lottery tickets on one idea.
- Different consumption modes. Some people won't unmute or watch 60 seconds, but will swipe eight slides in silence — commuters, office scrollers, skimmers.
- Saves. Carousels are the most-saved format on Instagram, and saves punch far above likes in ranking weight. A how-to video becomes a how-to reference when it's slides.
- A second data point. When the carousel outperforms the video (it often will), you've learned something about your audience for free.
Method 1: The frame-grab (visual videos)
For videos where the visuals carry the content — transformations, processes, demos:
- Scrub for the 6–10 strongest frames: the before, the key steps, the after. Export stills at the highest quality your editor allows.
- Re-crop to 4:5 (1080×1350 — the carousel sizing rules apply in full). Video is 9:16 or 16:9, so decide per-frame what the crop keeps.
- Add a text layer per slide — one line each. The video had your voice explaining; slides need the caption baked in.
- Design slide 1 as a hook, not a screenshot: big type over your best frame.
Method 2: The transcript (talking-head and educational videos)
For videos where the words carry the content, the frames are irrelevant — extract the structure instead:
- Pull the transcript (your editor, YouTube's auto-captions, or any transcription tool).
- Reduce it to its skeleton: the core claim plus each supporting point, one line each. A 90-second talking-head video is almost always a 7-point list wearing a face.
- Pour the skeleton into a branded carousel template — hook slide, one point per slide, CTA slide.
- Ignore the video's visuals entirely. This method produces a designed text carousel; it just happens to share DNA with a video.
The transcript method is the automatable one, and it's how the conversion stops costing even the 15 minutes: in FeedForce, a Custom Agent node can take the transcript, reduce it to slide copy in your voice, and an Apply Template node fills your carousel template — leaving you a quick look at the drafted slides instead of a design session. (The general pattern is in our content automations guide.)
The full extraction, while you're at it
If you're opening the video anyway, take everything:
| From one video | Format | Where it goes |
|---|---|---|
| The carousel | 6–10 slides | Instagram, LinkedIn |
| The best 15 seconds | Clip | Reels/TikTok/Shorts (again) |
| The core claim | Text post | X, Threads, LinkedIn |
| The transcript | Article seed | Blog, newsletter |
| The best frame + quote | Quote card | Everywhere |
One production effort, five or six published assets. That ratio — not posting more often — is what "content at scale" actually means for small teams.
Make it a habit, not a project
The workflow that sticks: every time a video finishes editing, its transcript goes into the carousel pipeline before the video is even posted. Video goes out day one, carousel day four, quote card day seven. Three feed appearances, one idea, and your content calendar fills itself from work you were already doing.