Adaptation matrix (updated)

Each format has different constraints and affordances. Additions for 2026: carousels (LinkedIn/Instagram), short-form vertical video (TikTok/Reels/Shorts), and automated clipping tools.

The repurposing waterfall (updated)

Content flows downhill from high-effort canonical sources to constrained formats. New recommended paths:

Canonical source (blog, long video, whitepaper) ├── Blog (canonical SEO source) ├── Long-form video / recorded talk → automated clip extraction (OpusClip) → shorts ├── LinkedIn article (expanded narrative + slides) ├── Carousel (3–8 slides) for Instagram/LinkedIn (visual summary) ├── X thread (compress to 5–7 punchy posts) ├── Newsletter segment (summarize + subscriber-only hook) └── Documentation example (code + how-to)

Video-first source (YouTube/recording) ├── Run automated clipping (OpusClip / Descript highlights) ├── Produce short-form videos (vertical) and clips with captions ├── Extract transcript → blog post skeleton → expand with code/metrics └── Create carousel and thread from transcript highlights

Reverse waterfall still applies: a viral short or thread can be expanded into a long-form blog, video, and documentation.

Adaptation, not duplication

Repurposing means rewriting for the format, not reformatting the same text. Automation helps with formats but requires editing and verification.

Content freshness layers (unchanged but emphasized)

When repurposing, add a freshness layer to each format:

• New proof: Updated metrics, additional screenshots, follow-up results
• New angle: A different perspective on the same insight for a different audience
• New depth: Expand a surface-level point into a full walkthrough
• New context: Add background that was assumed in the original format
• New CTA: Each format should drive a different action (follow, subscribe, bookmark, star)

Workflow (revised to include tools)

Step 1 — Extract the content atom

From the source content, identify:

• The single core insight (one sentence)
• The proof that supports it (metric, screenshot, code, story)
• The audience who cares (and why)

Step 2 — Choose target formats

Use the adaptation matrix to select 2–5 formats based on:

• Where your audience is (platform presence)
• What format suits the atom (visual → video; step-by-step → thread; narrative → LinkedIn)
• Effort budget (start with highest-value, lowest-effort adaptations)

Step 3 — Map the freshness layer

For each target format, identify what new element you will add (new proof/angle/depth/context/CTA).

Step 4 — Draft in waterfall order (practical)

1. Create or update the canonical source (blog, long video, docs).
2. Produce a single-line post (X/Threads single post) to force the hook.
3. Expand to X thread or LinkedIn depending on audience.
4. Produce a carousel for visual platforms (Instagram/LinkedIn) using extracted highlights.
5. Produce short-form video clips and vertical edits (use OpusClip / Descript to accelerate clipping; then human-edit for hook and captions).
6. Draft newsletter segment with subscriber-only insight.
7. Update documentation with code examples if applicable.

Automation note: use AI tools to extract transcripts, identify highlight moments, and generate format-specific drafts — then human-edit strictly for accuracy and voice.

Step 5 — Schedule across time (unchanged guidance)

Stagger publishing across 5–14 days depending on platform and shelf life. Avoid simultaneous cross-posting unless each format has distinct value.

Step 6 — Cross-link

Every format should reference at least one other format. Ensure links are live and follow the waterfall order (publish canonical first when possible).

Tools (practical shortlist — monitored sources added)

• OpusClip / Opus.pro — AI clipping and automated clip-generation for long videos → short-form socials (use to find highlights, then human-edit). (source: opus.pro)
• LayerProof — Paste a URL and generate native-formatted posts for LinkedIn, X, Instagram, TikTok with traceable claims (Product Hunt listing). (source: Product Hunt)
• Descript — Transcription + overdub + editing; useful for video-first workflows and extracting talk transcripts.
• Repurpose.io — Automated publishing pipelines for audio/video to social formats and feeds.
• CastMagic / Notebooks — Transcript-driven note-taking and highlight extraction for newsletters and threads.
• Pictory / FlexClip / other AI video generators — quick blog-to-video or script-to-video experiments.

Operational rule: when you use an automated tool, add a single verification step: (1) fact/metric check, (2) voice/brand alignment, (3) legality/privacy (permission to clip or quote). Do not publish automated claims without these checks.

Examples (updated)

Example: repurposing a conference talk Source: 25-minute conference talk on "Evals for AI Agents" Content atom: "A basic eval suite takes 30 minutes and catches 80% of production failures." Repurposing plan (condensed):

1. Blog post (Day 1): transcript + code examples
2. OpusClip highlights (Day 2): 3 shareable short clips with captions
3. X thread (Day 3): 7 posts with the framework + link to blog
4. Carousel (Day 4): 6-slide visual summary for LinkedIn/Instagram
5. Newsletter (Day 6): "The one slide that got the most questions" + downloadable eval template
6. Documentation (Day 10): Eval setup guide for the project's docs

Example: repurposing a viral short or thread Source: Viral X/Threads post

1. Expand into a threaded narrative with proof and steps
2. Generate a blog skeleton from the thread (add code and benchmarks)
3. Use OpusClip/Descript on a recorded walkthrough to create a companion video
4. Create a carousel summarizing the blog and link back

Decision tree (unchanged logic)

1. Do I have a content atom worth repurposing? No → create original content first. Yes → continue.
2. Is the atom time-sensitive? Yes → prioritize fastest formats (X/Threads, short videos). No → use the full waterfall.
3. Which format is the source? Start from the source and flow outward in the waterfall.
4. What freshness layer can I add to each format? New proof, angle, depth, context, or CTA.
5. Do I have the effort budget for all planned formats? No → prioritize the 2–3 highest-value formats. Yes → execute the full plan.
6. Am I duplicating or adapting? If the text is the same across formats, rewrite. Adaptation means rewriting for each format's constraints.

Edge cases and gotchas (updated)

• Atom exhaustion — if you cannot state the atom in one sentence, it may not be worth repurposing.
• Format forcing — skip formats that do not fit the atom.
• Audience overlap — the freshness layer is mandatory.
• Stale repurposing — if the source content is 3+ months old, update the data before repurposing.
• SEO cannibalization — blog post and documentation should target different keywords.
• Cross-link dead ends — publish in waterfall order to avoid dead links.
• Effort creep — cap at 4–5 formats per atom unless ROI justifies more.
• Platform policy — some platforms suppress cross-platform links; adapt CTAs and use comments or link-in-bio where necessary.
• Automated tool hallucinations — verify any generated claims. Tools can speed formatting (OpusClip, LayerProof) but not replace editorial review.

Evaluation criteria (unchanged)