/director

GEO Packaging — Transcript + VideoObject for AI Citation

Purpose: Package every externally distributed demo so AI Overviews, ChatGPT, AI Mode, and YouTube search can cite specific timestamped segments. Used by the geo recipe.

Why GEO Matters (2026)

• AI assistants quote 20–660 second segments from videos — not whole files. Without a transcript and schema, your demo is invisible to citation.
• Combined YouTube + branded web mentions correlate strongest with brand visibility in ChatGPT / AI Overviews / AI Mode.
• Multi-channel distribution of a video can increase AI citation by up to 325%.
• VideoObject structured data correlates with up to +41% CTR on search.

If your demo is not transcribed and schema-tagged, it loses to competitors who do this for free.

Deliverables

Every external demo ships with all four:

1. WebVTT transcript (.vtt) — same cue structure as captions, accessible to crawlers.
2. Plaintext transcript (.txt) — for blog embeds, social copy reuse, screen-reader fallback.
3. VideoObject JSON-LD — schema.org markup with hasPart clips, transcript, thumbnailUrl.
4. Chapter cue map — timestamp → topic, embeddable in YouTube description and on-page.

WebVTT Transcript

Same shape as closed captions but optimized for search:

WEBVTT

NOTE
  Demo: Acme Migration Tool
  Duration: 0:60
  Locale: en-US

00:00:00.000 --> 00:00:03.000
Schema drift broke production. Again.

00:00:03.000 --> 00:00:10.500
Most teams manually diff three environments
and pray nothing else changed.

00:00:10.500 --> 00:00:40.000
With Acme, one click previews and applies
the migration across staging, canary, and prod.

Differences from captions:

• Longer cues are OK (this is for reading, not display).
• Cue text repeats narration verbatim — no [music] cues required.
• Speaker labels optional.

Plaintext Transcript

Single-paragraph or paragraph-per-chapter prose. Used for:

• Blog post embeds
• Social copy reuse
• Screen-reader / accessibility fallback
• Direct AI ingestion (some models prefer plain text over VTT)

[0:00] Schema drift broke production. Again. Most teams manually
diff three environments and pray nothing else changed.

[0:10] With Acme, one click previews and applies the migration
across staging, canary, and prod.

[0:40] Result: zero manual SQL, zero overnight rollbacks. Acme
catches drift before it ships.

[0:55] Start free at acme.dev/migrate.

VideoObject JSON-LD

Place this in the page that embeds the video, or alongside the video file as demo.jsonld.

{
  "@context": "https://schema.org",
  "@type": "VideoObject",
  "name": "Acme Migration Tool — 60s demo",
  "description": "One-click schema migration across staging, canary, and production with drift detection.",
  "thumbnailUrl": [
    "https://cdn.acme.dev/demos/migrate_thumb_1280x720.jpg",
    "https://cdn.acme.dev/demos/migrate_thumb_1080x1080.jpg",
    "https://cdn.acme.dev/demos/migrate_thumb_1080x1920.jpg"
  ],
  "uploadDate": "2026-05-15T09:00:00+09:00",
  "duration": "PT1M",
  "contentUrl": "https://cdn.acme.dev/demos/migrate_16x9_20260515.mp4",
  "embedUrl": "https://www.youtube.com/embed/EXAMPLE_ID",
  "transcript": "Schema drift broke production. Again. Most teams manually diff three environments...",
  "inLanguage": "en-US",
  "publisher": {
    "@type": "Organization",
    "name": "Acme",
    "logo": {
      "@type": "ImageObject",
      "url": "https://cdn.acme.dev/logo.png"
    }
  },
  "hasPart": [
    {
      "@type": "Clip",
      "name": "The schema-drift problem",
      "startOffset": 0,
      "endOffset": 10,
      "url": "https://cdn.acme.dev/demos/migrate_16x9_20260515.mp4#t=0,10"
    },
    {
      "@type": "Clip",
      "name": "Preview migration across environments",
      "startOffset": 10,
      "endOffset": 40,
      "url": "https://cdn.acme.dev/demos/migrate_16x9_20260515.mp4#t=10,40"
    },
    {
      "@type": "Clip",
      "name": "Apply with zero rollback risk",
      "startOffset": 40,
      "endOffset": 55,
      "url": "https://cdn.acme.dev/demos/migrate_16x9_20260515.mp4#t=40,55"
    }
  ]
}

Required Fields

Chapter Cue Map (YouTube Description Format)

Paste directly into YouTube description for native chapter navigation:

0:00 The schema-drift problem
0:10 Preview migration across environments
0:40 Apply with zero rollback risk
0:55 Try it free at acme.dev/migrate

Rules (YouTube enforces):

• First chapter must start at 0:00.
• Chapters must be at least 10 seconds long.
• At least 3 chapters per video.

This unlocks YouTube "Key Moments" snippets that AI Overviews and Google Search use.

Multi-Channel Distribution Plan

To maximize AI citation (up to +325%), distribute the same demo across multiple surfaces. AI models fuse citations across sources.

AI models that crawl YouTube + your domain see the same video twice, which boosts citation confidence.

Workflow

INPUT          →  finalized demo video + closed-caption .vtt
               →  brand metadata (name, description, publisher)

EXTRACT        →  parse .vtt cues → plaintext transcript
               →  identify chapter boundaries (≥10s each, ≥3 chapters)

SCHEMA         →  generate VideoObject JSON-LD
               →  populate hasPart[] with timestamped Clip objects
               →  attach thumbnailUrl per aspect

PUBLISH        →  upload video to YouTube + CDN
               →  paste chapter cues into YouTube description
               →  embed JSON-LD on hosting page
               →  add transcript section on the page (visible or collapsed)

VERIFY         →  Google Rich Results Test: search.google.com/test/rich-results
               →  YouTube chapters appear in player
               →  spot-check transcript matches narration

DISTRIBUTE     →  cross-post per Multi-Channel Distribution Plan

HANDOFF        →  Quill: doc-side embed + transcript
               →  Growth: A/B thumbnail + UTM tagging
               →  Builder: CDN upload + schema embed in templates

Anti-Patterns

References

• schema.org VideoObject — schema.org/VideoObject
• Google Rich Results Test — search.google.com/test/rich-results
• YouTube chapter rules — support.google.com/youtube
• "VideoObject Schema" — AEO Expert (2026)
• "Video Citing AI Overviews" — Reel n Reel (2026)
• "YouTube Optimization for AI" — Am I Cited (2026)