Convert LinkedIn articles into repurposed video content with captions and distribution

LinkedIn article writers often find themselves trapped in a content multiplication trap: you spend three hours writing a 1,500-word article, then face the prospect of recording a video, generating captions, scripting social snippets, and scheduling cross-platform posts. Most content creators simply don't do it. The article sits in isolation, reaching only the subset of your network who scroll to written posts that week. Meanwhile, the same core ideas could work beautifully as a 60-second video on TikTok, a carousel on Instagram, a thread on X, and a motivational clip for YouTube Shorts. The solution isn't to work harder. It's to set up a workflow that extracts your article once and distributes it everywhere, with minimal human intervention after that first publish button. This guide walks you through building exactly that: a system that pulls your LinkedIn article content, transforms it into video with captions, and schedules the results across social platforms automatically. The real payoff arrives when you publish on Sunday evening and wake up Monday to find your video already live on TikTok, Instagram, LinkedIn, and X, with matching captions and graphics queued for the week.

The Automated Workflow

The workflow runs on Make (Integromat) as the orchestration layer, though Zapier or n8n will work equally well if you prefer. The process has four phases: extraction, transformation, production, and distribution.

Phase 1: Trigger and Extract

You'll start with a simple but essential manual step: save your LinkedIn article as a PDF immediately after publishing. This PDF becomes your single source of truth. Rather than scraping LinkedIn directly (which breaks their terms), you'll feed the PDF into Chat With PDF by Copilot.us through its API. In Make, set up a webhook that triggers when you upload a file to a designated folder in Google Drive or Dropbox. This folder becomes your content hopper.

CopyPOST /webhook/process_article
{ "file_path": "article_pdf_url", "title": "Your LinkedIn Article Title", "publish_date": "2026-03-15"
}

Once triggered, Make calls the Chat With PDF API to extract the article's key points, main argument, and a short summary.

CopyPOST https://api.copilot.us/chat-with-pdf/extract
{ "pdf_url": "gs://drive.google.com/file/d/...", "prompt": "Extract the 5 core takeaways from this article in bullet points. Keep each under 20 words.", "model": "gpt-4o"
}

Store the response in a JSON variable within Make. You now have structured content ready for the next phase.

Phase 2: Script Generation and Video Creation

Pass the extracted takeaways to Claude Opus 4.6 (via the Anthropic API) to generate a 90-second video script. Claude excels at condensing longer arguments into tight, spoken narratives with clear hooks.

CopyPOST https://api.anthropic.com/v1/messages
{ "model": "claude-opus-4.6", "max_tokens": 800, "system": "You are a video scriptwriter. Write a 90-second spoken script (roughly 225 words) based on these article takeaways. Include a hook in the first 10 seconds, three main points, and a call-to-action at the end.", "messages": [ { "role": "user", "content": "[Extracted takeaways from Phase 1]" } ]
}

Feed this script to Hour One, which will render it as a professional video with a virtual presenter and automatic B-roll suggestions. Hour One's API accepts script text and returns a video URL within minutes.

CopyPOST https://api.hourone.com/videos/create
{ "script": "[Claude-generated script]", "presenter_style": "professional", "video_length": "90_seconds", "background": "corporate_office", "aspect_ratio": "9:16"
}

Make waits for the video to render (polling the job status endpoint), then stores the video URL.

Phase 3: Subtitles and Clipping

Next, send the video URL to Clipwing, which will automatically generate captions, detect scene breaks, and produce short clips optimised for TikTok, Instagram Reels, and YouTube Shorts.

CopyPOST https://api.clipwing.io/jobs
{ "video_url": "https://hourone-output.com/video/abc123", "auto_caption": true, "caption_style": "white_bold", "clip_targets": ["tiktok", "instagram_reels", "youtube_shorts"], "output_format": "mp4"
}

Clipwing returns an array of 3-5 short clips, each with hardcoded captions and platform-specific dimensions. This step eliminates manual editing entirely.

Phase 4: Social Distribution via SocialBu

Now you have a 90-second video, several 15-30-second clips, and captions. Use SocialBu's API to schedule these across your accounts. SocialBu integrates with LinkedIn, TikTok, Instagram, X, and Facebook, so a single API call can distribute to multiple platforms.

CopyPOST https://api.socialbu.com/v2/posts/schedule
{ "platforms": ["linkedin", "tiktok", "instagram", "x", "facebook"], "content_type": "video", "primary_video_url": "https://clipwing.io/output/long.mp4", "short_clips": [ { "url": "https://clipwing.io/output/clip_1_tiktok.mp4", "platform": "tiktok" }, { "url": "https://clipwing.io/output/clip_2_insta.mp4", "platform": "instagram" } ], "captions": { "linkedin": "Just published a close look on [topic]. Watch the summary here.", "tiktok": "Here's why [key insight]. Full article in bio.", "instagram": "Nobody talks about this. [Call-to-action].", "x": "[Tweet version of key point] Full video on TikTok.", "facebook": "A new perspective on [topic]. See the full discussion." }, "schedule_dates": { "linkedin": "2026-03-16T09:00:00Z", "tiktok": "2026-03-16T18:00:00Z", "instagram": "2026-03-17T10:00:00Z", "x": "2026-03-17T12:00:00Z", "facebook": "2026-03-17T19:00:00Z" }
}

The Make workflow ends here. From PDF upload to fully scheduled, multi-platform video content takes roughly 45 minutes of processing time, with zero manual handoff in between.

The Manual Alternative

If you prefer more creative control, you can pause the workflow at Phase 2. Generate the Claude script and Hour One video, then review and edit both before proceeding. This is sensible if you're trying new messaging or testing audience response. Simply approve the video in Make, which resumes the workflow at Phase 3. Alternatively, skip SocialBu and manually craft captions for each platform using the AI-generated scripts as a foundation. You'll retain editorial voice while still eliminating the most repetitive work: video production and clip generation.

Pro Tips

Rate Limiting and Batching

Hour One and Clipwing have concurrent video processing limits.

If you publish multiple articles per week, queue them in Make with staggered start times rather than triggering all four workflows simultaneously. Use Make's built-in delay modules to space requests 5 minutes apart.

Cost Optimisation

Claude Opus 4.6 costs more per token than Claude Sonnet 4.6, but for script generation, Sonnet is entirely adequate. Switch the API call to Sonnet and save roughly 60% on LLM costs. Reserve Opus for complex multi-step reasoning tasks where nuance matters.

Captions and Accessibility

Clipwing's automatic captions are good but not perfect, especially for industry jargon. Add a manual review step before final distribution if your content uses technical terms. A human reading captions takes five minutes; caption errors can cost you engagement.

Video Length Mismatches

LinkedIn's algorithm favours 60-120 second videos, whilst TikTok performs better with 15-30 second hooks. Hour One's 90-second output is deliberately sized to work across both. If your article is very technical or dense, consider requesting a 120-second version from Claude instead.

Tracking Performance Across Platforms

SocialBu provides basic analytics, but platform-specific insights matter. Set up Make to log video URLs and scheduled times to a Google Sheet alongside the original article title. After two weeks, manually compare views, shares, and saves across platforms to learn which video lengths and captions perform best for your audience.

Cost Breakdown