The thing that made me build this feature was a familiar pattern: a post does well on one platform, and then you look at your other platforms and they are dark. Instagram empty, TikTok empty, Twitter showing nothing because nobody clicks external links there. The content is trapped on one platform while the other four sit idle.
Every marketing guide says the same thing: repurpose it. Most marketers already do. But actually doing it is where things break down. Turning a LinkedIn article into a Twitter thread, an Instagram carousel, a TikTok script, and a Facebook post isn't just shortening the text. Each platform has its own limits, audience, format, and algorithm. A LinkedIn insight sounds pretentious on TikTok; a Twitter hot take sounds shallow on LinkedIn.
I was doing it by hand, for every post. It took longer than writing the original, and by the third adaptation I was tired and just wanted to publish, so the results were mediocre. So I built Repurpose Studio in Sydium. Here is how that went.
Where existing tools hit a wall
I tested everything before building my own. The tools split into two camps, and both stop short.
Format converters like Repurpose.io connect a source platform to destinations and handle format conversion. Upload a YouTube video and it becomes TikTok clips, Reels, and Shorts, watermarks stripped so it looks native. Genuinely useful, but it is video-in, video-out. If your content is text, it can't help.
AI rewriters like Blaze.ai and Typeface take text and generate platform-specific versions. Closer to what I wanted, but in testing the output had two consistent problems. First, they don't know platform rules: an AI might write a thread of 6 tweets at 450 characters each, fine for X Premium but useless for everyone capped at 280, or a 3,000-character Instagram caption when the limit is 2,200. Second, they don't keep your voice. The LinkedIn version sounds like LinkedIn-AI, the Twitter version like Twitter-AI. Neither sounds like you.
I wrote about building the brand voice system and building Autopilot separately. The repurposing engine sits on top of that voice profile and adjusts for platform context, not the reverse.
The three content sources
Repurpose Studio accepts content from three places, each needing different handling.
Calendar posts are the easy case: you published something in Sydium, it did well, now you want it elsewhere. I already have the full post data, nothing to reconstruct.
Top performers is where the value kicks in. The system surfaces your highest-engaging content across all connected platforms. Instead of guessing what is worth repurposing, the data tells you. A LinkedIn post that got 10x your average is worth adapting; an average one probably isn't.
Pasted content is the hard case: a blog article, an email, a podcast transcript. No metadata, no engagement signals, no platform context. The system has to read it cold and figure out what to do.
The platform rules engine
This is the part I am most frustrated took so long. It shouldn't be hard, since character limits and image dimensions are public. But the reality is messier, in three ways.
Character limits, hard and soft. The hard caps are easy to look up: 280 for standard Twitter, 2,200 for Instagram, 3,000 for LinkedIn, 500 for Threads, 300 for Bluesky. The real limits are the engagement-optimal lengths. Facebook posts of 40-80 characters get more engagement than longer ones. Twitter performs best at 70-100. Instagram's effective limit is the 125-character fold; most people never read past it. The engine tracks both and aims for the soft target while staying under the hard cap.
Media specs. Each platform wants different dimensions (1080x1350px for Instagram vertical, 1200x1200px for LinkedIn). The engine doesn't just flag a mismatch, it gives the fix: "This image is 1920x1080. For Instagram feed, crop to 1080x1350."
Hashtag culture. This is where platforms diverge most, and it is cultural, not technical. Instagram tolerates 5-15. LinkedIn looks spammy past 3-5. Twitter wants 1-2 or none. So a post with 15 Instagram hashtags gets trimmed to 3 for LinkedIn and stripped for Twitter.
How one post becomes five
When you choose "Repurpose," four things happen.
1. Content analysis. The system breaks the source into parts: thesis, supporting points, data, calls to action, hooks, stories. This matters because platforms want different parts: Twitter the hook, LinkedIn the argument, Instagram the emotion, TikTok the story.
2. Platform generation. For each platform, the AI writes a new version from your brand voice profile, the rules-engine constraints, the decomposed parts weighted by platform, and a format-specific structure. This isn't summarization, it is re-expression: rethinking how to say the same idea in a different format.
3. Content scoring. Each variation gets a score across voice match, platform compliance, content completeness (did the key insight survive?), and engagement prediction. A gut-check before you publish: 90+ is strong, below 70 needs editing, below 50 means the idea probably doesn't fit.
4. Caption editing. Each variation opens in an editor with platform-specific tools: hashtag suggestions, mention detection, and a live character count. The system does the first 80%; you handle the 20% that needs judgment.
The "schedule all at once" problem
One of the most requested features was scheduling all variations in one click. It sounds simple, but you can't post the same thing everywhere at the same moment. Audiences overlap, so someone who follows you on Twitter and LinkedIn sees the same insight twice within minutes, which feels lazy. Some algorithms may also penalize content that lands across platforms at once. So Sydium offers simultaneous posting for time-sensitive content and staggered as the default, spreading variations across hours or days.
Priority order matters too. A data-heavy insight does better on LinkedIn and Twitter; a visual story does better on Instagram and TikTok. The system suggests an order based on content type and your engagement history. The shift is from "blast it everywhere" to "lead with the platform where it lands best, then cascade."
Three things I got wrong
I treated repurposing as summarization. My first version compressed content for short platforms and padded it for long ones. Awful. Summarization strips the nuance that makes content interesting; expansion adds filler. The breakthrough was asking the AI not "shorten this for Twitter" but "what is the Twitter-native way to say this?" Now the same insight might be a question on Twitter, a personal story on LinkedIn, a carousel on Instagram, and a to-camera script on TikTok. Same idea, different expression.
I generated all platforms in one prompt. The first version sent one request for all five at once. Quality was terrible. The AI spread its attention thin and every version was off, too long here, too short there. I switched to one platform at a time, each with a dedicated prompt. This tripled the API cost, but each platform now gets full attention and a tuned prompt.
I tried to force-fit everything everywhere. A technical analysis with code snippets doesn't belong on TikTok. A selfie-style post doesn't translate to LinkedIn. Now scoring flags the mismatch and recommends skipping: "Your technical deep-dive scored 34/100 for TikTok. Consider skipping it." This felt backwards, since the whole point is posting on MORE platforms. But Siege Media's guide makes the point well: quality per platform beats presence on every platform.
The ROI math that made this a core feature
The average blog post takes over four hours to write. Repurposing it into 5 posts by hand adds another 2-3 hours; by machine it takes minutes plus 15-20 to review. The bigger number is reach: repurposed content can generate far more views than the original, and every long post contains 5-10 insights that stand alone as social posts, each able to reach a different audience. You already did the hard work; repurposing pulls more value out of it. For solo creators and small agencies, that is the difference between staying active on 5 platforms and going dark on 4.
The thing nobody tells you
Platform knowledge is the moat. Anyone can ask an LLM to rewrite content; the value is in knowing the rules and resisting the temptation to chase every platform. Repurpose.io connects to 20+, and the pull is to match that. Instead I built the 5 that matter most for Sydium's users. Better to repurpose excellently for 5 than badly for 15.
I wrote about the reality of building in public, and this feature is that grind: weeks on a rules engine nobody sees, days on character-count edge cases, watching the AI nail LinkedIn and bomb TikTok. But when I see someone turn one blog post into a week of content across five platforms in 20 minutes, each sounding like them, I know it was worth building.
If you spend hours adapting content by hand, try Sydium for free. Every piece you write has ideas trapped inside it, waiting for audiences you don't reach.
Questions creators ask
Can I repurpose video the same way as text?
Video works differently. Sydium generates platform-optimized captions, descriptions, and surrounding text, but the video itself still needs separate editing, since Reels are vertical, Shorts have different length limits, and TikTok has its own trends. The text layer around video is where AI repurposing helps. You can also pull key moments from longer videos as short-form starting points, but the editing stays a separate workflow.
How do I track which repurposed version performs best?
Sydium's analytics group performance across platforms for content from the same source, so you can see that your LinkedIn carousel beat the Twitter thread, or that the Instagram Reel fell flat. Over time the system learns which content types work where for your audience, so if data posts keep flopping on TikTok, it stops suggesting it. For a step-by-step version, see how to repurpose content across 5 platforms.
Related free tools
Free, no signup, runs in your browser.
- Social Media Character Counter - Check your caption length against all major platform limits in real-time. Optimize for engagement.
- Hashtag Generator - Generate relevant hashtags for your content using AI. Get a mix of popular and niche tags.
- Social Media Image Resizer - Resize images for every platform instantly. 14 presets across 7 platforms, no signup required.