You made a video. It's good — maybe 3-5 minutes, edited decently, real value in it. So you post it on YouTube and watch it get decent views. Then you chop it up into a 60-second clip for TikTok, thinking "same content, different shape, should still work."
And it falls flat. Meanwhile, someone tweets a three-sentence text insight about the same topic and gets 5x more engagement.
It's not random. It's not that video is "always better" or that text is "dying." It's that each platform's algorithm rewards fundamentally different things — and when you're repurposing across platforms, format choice isn't optional. It's the whole strategy.
Every platform measures engagement differently, and those measurements drive what gets distributed:
TikTok, Instagram Reels, and YouTube Shorts measure completion rate — did you watch the whole thing, or bounce in 3 seconds? For these platforms, video is structurally rewarded. TikTok's algorithm pushes video that holds attention. Instagram Reels completion rates hover around 72% for videos under 30 seconds, and that completion-time signal beats almost everything else in the ranking. YouTube Shorts, by comparison, sees far lower engagement (0.30% per some 2026 benchmarks), suggesting algorithm maturity matters: TikTok's system is honed, Shorts' is still scaling.
But here's the gap: that doesn't mean text loses everywhere. It means text loses on these platforms specifically, because they're built to measure watch time, not read time.
LinkedIn measures dwell time — how long you spend looking at something — not just whether you liked it. Document posts (PDF carousels) hit 6.6% engagement, nearly double what video achieves on the platform. Why? Because swiping through slides forces sustained attention. Carousels are the format that demands time; video is the format that can be skipped. When LinkedIn switched to its unified 360Brew AI ranking system in March 2026, this dwell-time signal became the core ranking input. Text-only posts on LinkedIn hit about 2% engagement — solid for the platform, but way behind carousels. Text works, but it's not the shape that plays to LinkedIn's specific reward structure.
X (Twitter) is the outlier. It's the only major platform where text outperforms video. Buffer's 2026 analysis of millions of posts across their state-of-social engagement report shows text at 3.56% median engagement vs. images at 3.39% — images are nearly tied, but video doesn't get the algorithmic boost it gets elsewhere. Why? X's core mechanic is speed and response — quick takes, fast replies, the culture of the platform is built around being current, not sitting still. Video slows that down. Text is the native format. (To be clear: media on X still helps — images do nearly as well as text — but video doesn't get the distribution bump it gets on TikTok or YouTube.)
Instagram splits its rewards. Reels reach 2.25x more people than single-image posts, so if your goal is raw reach, video wins. But carousels drive 109% more engagement per person reached. So you face a real trade-off: video gets the distribution to new people, carousels get the deeper engagement from people who actually see them. Choose based on what you actually want — reach or conversion.
Most people repurpose by format alone: "I have a blog post, make me a video. Make me a tweet. Make me a carousel." The repurposing tool turns the input into the right shape for each platform.
But shape isn't the whole story. The platform's reward structure is. So when you repurpose a video into a tweet by chopping the talking points into bullet points, you're losing the thing that made video work on its native platform (watch time) and landing it on a platform that doesn't reward watch time. You get a format-correct tweet, but not a tweet optimized for what X actually values.
Flip it: repurpose a killer tweet (a specific, fast idea that worked because of its brevity) into a LinkedIn carousel by stretching it into a 5-slide story with context and a payoff, and you're giving LinkedIn what it actually wants — sustained attention, depth, reason to keep swiping. Same idea, but now it fits the reward structure instead of fighting it.
The practical consequence: if you're trying to get the same idea to work across multiple platforms, the format and the depth/structure have to change together. "Repurposing" isn't copy-paste-different-platform, it's "rebuild for this platform's specific algorithm."
One clear pattern across all the 2026 data:
If you're taking one piece of content and sending it everywhere, the format change is step one. But step two — the one most repurposing tools miss — is adapting to the reward structure.
Same idea doesn't mean same structure. Platforms changed how they measure what matters, and your format has to follow, not fight it. That's what repurposing actually means.
Sources: Buffer's 2026 State of Social Media Engagement report (45M+ posts analyzed, covering format performance across all major platforms), LinkedIn's 360Brew AI ranking system documentation and March 2026 deployment (dwell-time signal confirmed via Dataslayer's July 2026 LinkedIn algorithm analysis), Instagram Reels completion rates from DigitalApplied's April 2026 Instagram statistics, TikTok engagement benchmarks from AutoFaceless's April 2026 video statistics report, and X's format performance data from PostEverywhere.ai's May 2026 X algorithm analysis.
Turn your own content into a week of platform-ready posts.
Try the Repurposer free →