Add /ultraresearch-local for structured research combining local codebase analysis with external knowledge via parallel agent swarms. Produces research briefs with triangulation, confidence ratings, and source quality assessment. New command: /ultraresearch-local with modes --quick, --local, --external, --fg. New agents: research-orchestrator (opus), docs-researcher, community-researcher, security-researcher, contrarian-researcher, gemini-bridge (all sonnet). New template: research-brief-template.md. Integration: --research flag in /ultraplan-local accepts pre-built research briefs (up to 3), enriches the interview and exploration phases. Planning orchestrator cross-references brief findings during synthesis. Design principle: Context Engineering — right information to right agent at right time. Research briefs are structured artifacts in the pipeline: ultraresearch → brief → ultraplan --research → plan → ultraexecute. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
606 lines
19 KiB
Markdown
606 lines
19 KiB
Markdown
# Video Scripting & Production Strategy Guide
|
|
|
|
Comprehensive video scripting reference for LinkedIn thought leadership. This guide focuses on **scripting, pacing, and production workflow** — for general video format specs, algorithm data, technical requirements, editing tools, and thumbnail strategy, see `linkedin-formats.md` (Video Content Deep Dive section).
|
|
|
|
---
|
|
|
|
## Length-Specific Script Templates
|
|
|
|
### 30-Second Script (75 words)
|
|
|
|
**Use case:** Single punchy insight, reaction to news, quick tip.
|
|
|
|
```
|
|
[0:00-0:03] HOOK (8 words)
|
|
Bold claim or pattern interrupt. No warm-up.
|
|
Energy: HIGH — lean into camera.
|
|
|
|
[0:03-0:08] CONTEXT (15 words)
|
|
One sentence: why this matters right now.
|
|
|
|
[0:08-0:25] INSIGHT (40 words)
|
|
The one thing they need to know.
|
|
Deliver with conviction. No hedging.
|
|
|
|
[0:25-0:30] CTA (12 words)
|
|
Direct question or follow prompt.
|
|
```
|
|
|
|
**Pacing:** 2.5 words/second = 75 words total.
|
|
**Energy curve:** Start at 8/10, sustain at 7/10, end at 8/10.
|
|
**Visual cues:** 1 text overlay (hook keyword), 1 CTA overlay.
|
|
|
|
---
|
|
|
|
### 60-Second Script (150 words)
|
|
|
|
**Use case:** Framework intro, single lesson, "here's what I learned."
|
|
|
|
```
|
|
[0:00-0:03] HOOK (8 words)
|
|
Pattern interrupt or bold claim.
|
|
Energy: HIGH.
|
|
|
|
[0:03-0:10] CONTEXT (18 words)
|
|
Brief setup: what happened or why this matters.
|
|
Transition: "And here's the thing..."
|
|
|
|
[0:10-0:45] MAIN CONTENT (88 words)
|
|
2 key points, ~44 words each.
|
|
|
|
Point 1 [0:10-0:27]:
|
|
State it. Explain why. Quick example.
|
|
Transition: "But that's only half of it..."
|
|
|
|
Point 2 [0:27-0:45]:
|
|
State it. Explain why. Connect to point 1.
|
|
Pause for emphasis before CTA.
|
|
|
|
[0:45-0:55] TAKEAWAY (24 words)
|
|
"Here's what this means for you:"
|
|
One clear actionable sentence.
|
|
|
|
[0:55-1:00] CTA (12 words)
|
|
Engagement question or follow prompt.
|
|
```
|
|
|
|
**Pacing:** 2.5 words/second = 150 words total.
|
|
**Energy curve:** 8/10 → 6/10 → 7/10 → 8/10 (wave pattern).
|
|
**Visual cues:** Hook keyword, point numbers, takeaway highlight, CTA.
|
|
|
|
---
|
|
|
|
### 90-Second Script (225 words)
|
|
|
|
**Use case:** Complete framework, story with lesson, detailed insight. **This is the optimal LinkedIn video length.**
|
|
|
|
```
|
|
[0:00-0:03] HOOK (8 words)
|
|
Pattern interrupt, bold claim, or stat shock.
|
|
Energy: HIGH — this determines 70% of retention.
|
|
|
|
[0:03-0:12] CONTEXT (23 words)
|
|
Why this matters. Brief personal connection.
|
|
"I discovered this when..." / "After analyzing X..."
|
|
|
|
[0:12-0:65] MAIN CONTENT (133 words)
|
|
3 key points, ~44 words each.
|
|
|
|
Point 1 [0:12-0:30]:
|
|
State the principle. Explain briefly.
|
|
Real example or data point.
|
|
Transition: "Second..."
|
|
|
|
Point 2 [0:30-0:48]:
|
|
State the principle. Why it matters.
|
|
Counter-intuitive angle if possible.
|
|
Transition: "And the most important one..."
|
|
|
|
Point 3 [0:48-1:05]:
|
|
The strongest point saved for last.
|
|
Deliver with increased energy.
|
|
Pause after for emphasis.
|
|
|
|
[1:05-1:20] TAKEAWAY (38 words)
|
|
Synthesize all 3 points into one insight.
|
|
"So here's what I want you to remember..."
|
|
Add personal reflection: "This changed how I think about..."
|
|
|
|
[1:20-1:30] CTA (23 words)
|
|
"Which of these resonates most with you?"
|
|
Or: "Follow for more [topic] breakdowns."
|
|
```
|
|
|
|
**Pacing:** 2.5 words/second = 225 words total.
|
|
**Energy curve:** 8/10 → 6/10 → 7/10 → 6/10 → 8/10 → 7/10 (double wave).
|
|
**Visual cues:** Hook keyword, 3 point numbers, takeaway highlight, CTA.
|
|
|
|
---
|
|
|
|
### 2-Minute Script (300 words)
|
|
|
|
**Use case:** Detailed story, multi-step process, deep contrarian take. **Use sparingly — retention drops after 90s.**
|
|
|
|
```
|
|
[0:00-0:03] HOOK (8 words)
|
|
Must be your strongest possible hook.
|
|
At 2 minutes, you need extra retention power.
|
|
|
|
[0:03-0:15] STORY SETUP (30 words)
|
|
Set the scene with specific details.
|
|
Create tension or curiosity.
|
|
"6 months ago, I was facing a problem..."
|
|
|
|
[0:15-1:30] MAIN CONTENT (188 words)
|
|
3-4 key points, ~47 words each.
|
|
|
|
Point 1 [0:15-0:35]:
|
|
The foundation. Set the baseline.
|
|
Use a concrete example.
|
|
Transition naturally.
|
|
|
|
Point 2 [0:35-0:55]:
|
|
Build on point 1. Add complexity.
|
|
"But here's where most people get it wrong..."
|
|
|
|
Point 3 [0:55-1:15]:
|
|
The revelation. The unexpected insight.
|
|
This is your strongest moment.
|
|
|
|
Point 4 (optional) [1:15-1:30]:
|
|
Only if essential. Otherwise extend point 3.
|
|
"And one more thing..."
|
|
|
|
[1:30-1:50] TRANSFORMATION (50 words)
|
|
The "so what" — what changed because of these insights.
|
|
Before/after contrast.
|
|
Specific outcome or result.
|
|
"Since implementing this, we've seen..."
|
|
|
|
[1:50-2:00] CTA (24 words)
|
|
Strong engagement hook.
|
|
"I'm curious — have you experienced this too?"
|
|
"Save this if you want to try it."
|
|
```
|
|
|
|
**Pacing:** 2.5 words/second = 300 words total.
|
|
**Energy curve:** 8 → 7 → 6 → 7 → 8 → 7 → 6 → 8 (sustained wave).
|
|
**Visual cues:** Hook keyword, story context, point numbers, transformation stat, CTA.
|
|
**Warning:** Only use 2-minute format when the content genuinely requires the extra time. Most ideas fit better in 60-90 seconds.
|
|
|
|
---
|
|
|
|
## Format-Specific Production Guidance
|
|
|
|
### Talking Head
|
|
|
|
**Best for:** Personal stories, opinions, lessons learned, authenticity-building.
|
|
|
|
**Script style:**
|
|
- Conversational — write for the ear, not the eye
|
|
- Contractions: "I've found" not "I have found"
|
|
- Short sentences: max 15 words when spoken
|
|
- Direct address: "you" not "people"
|
|
- Personal pronouns: "I", "we", "my team"
|
|
|
|
**Visual cues to include in script:**
|
|
```
|
|
[CAM: direct] — Look at camera (default)
|
|
[CAM: slight left] — Break eye contact for storytelling
|
|
[CAM: lean in] — Emphasize key point
|
|
[GESTURE: count] — Use fingers for numbered points
|
|
[GESTURE: open hands] — Openness, invitation
|
|
[PAUSE: 1s] — Let a point land
|
|
[ENERGY: up] — Increase enthusiasm
|
|
[ENERGY: down] — Slow for emphasis
|
|
```
|
|
|
|
**Production tips:**
|
|
- Film in natural light facing a window
|
|
- Eye-level camera on tripod
|
|
- Clean, non-distracting background
|
|
- Lavalier mic for consistent audio
|
|
- Film in one take if possible (authenticity > perfection)
|
|
|
|
---
|
|
|
|
### Screen Recording
|
|
|
|
**Best for:** Tool walkthroughs, demos, data analysis, process tutorials.
|
|
|
|
**Script style:**
|
|
- Instructional — clear, step-by-step narration
|
|
- Announce what you're about to do before doing it
|
|
- Narrate mouse movements: "I'll click on..." / "Notice how..."
|
|
- Pause between steps for viewer processing
|
|
|
|
**Visual cues to include in script:**
|
|
```
|
|
[SCREEN: show app] — Full screen capture of application
|
|
[SCREEN: zoom to X] — Zoom into specific UI element
|
|
[SCREEN: highlight X] — Circle or arrow pointing to element
|
|
[CAM: picture-in-picture] — Small webcam overlay (corner)
|
|
[CAM: full] — Switch to full webcam (for intro/outro)
|
|
[CURSOR: circle] — Draw attention to cursor location
|
|
[TEXT: step N] — On-screen step number
|
|
```
|
|
|
|
**Production tips:**
|
|
- Clean desktop, close notifications
|
|
- 1080p minimum, 1440p preferred
|
|
- Pre-load all tabs/apps before recording
|
|
- Use mouse highlights (pointer circle)
|
|
- Record voiceover separately if needed (cleaner audio)
|
|
|
|
---
|
|
|
|
### Slideshow / Visual Sequence
|
|
|
|
**Best for:** Frameworks, data visualization, step-by-step processes, lists.
|
|
|
|
**Script style:**
|
|
- Concise narration per slide — voiceover guiding the visual
|
|
- Each slide has a clear purpose statement
|
|
- Transitions are verbal: "Now let's look at..." / "Next..."
|
|
- Build suspense between slides
|
|
|
|
**Visual cues to include in script:**
|
|
```
|
|
[SLIDE: title] — Title slide with hook text
|
|
[SLIDE: point N] — Content slide with numbered point
|
|
[SLIDE: data] — Chart, graph, or statistic
|
|
[SLIDE: comparison] — Before/after or A vs B
|
|
[SLIDE: summary] — Key takeaway recap
|
|
[SLIDE: CTA] — Call-to-action slide
|
|
[TRANSITION: fade] — Smooth transition between slides
|
|
[TRANSITION: cut] — Hard cut for emphasis
|
|
[BUILD: animate] — Reveal elements progressively
|
|
```
|
|
|
|
**Production tips:**
|
|
- 4:5 aspect ratio slides (1080x1350)
|
|
- Large text: 28pt+ body, 36pt+ headings
|
|
- Max 3 bullet points per slide
|
|
- Use brand-consistent colors
|
|
- Record narration after designing slides
|
|
- 3-5 seconds per slide minimum
|
|
|
|
**Programmatic video with Remotion:**
|
|
Slideshow/visual sequence videos can be generated directly from Claude Code using [Remotion](https://www.remotion.dev/) — a React framework for programmatic video creation. This is especially powerful for:
|
|
- Data-driven slides with dynamic content (stats, charts)
|
|
- Branded templates with consistent styling across videos
|
|
- Batch-generating multiple video variations from scripts
|
|
- Animated text overlays, transitions, and build sequences
|
|
|
|
When a script uses `[SLIDE:]`, `[BUILD: animate]`, or `[TRANSITION:]` cues, these map directly to Remotion components. The video-scripter agent's output can serve as a blueprint for a Remotion composition.
|
|
|
|
---
|
|
|
|
## 2026 Video Algorithm Nuances
|
|
|
|
### Completion Rate: The #1 Video Metric
|
|
|
|
LinkedIn's algorithm weights **completion rate** above all other video metrics. A 60-second video watched to the end outperforms a 2-minute video watched 50%.
|
|
|
|
**Completion rate targets:**
|
|
| Length | Target Rate | Signal |
|
|
|--------|------------|--------|
|
|
| 30s | 70%+ | Strong — short enough for most viewers |
|
|
| 60s | 55%+ | Good — 2026 sweet spot for depth vs completion |
|
|
| 90s | 45%+ | Risky — retention drops, only for complex frameworks |
|
|
| 2min | 35%+ | Dangerous — most viewers won't hit 30% completion gate |
|
|
|
|
**Critical (2026):** LinkedIn requires **30% minimum completion rate** or the video gets **zero distribution**. This makes shorter videos significantly safer. 60 seconds is the new recommended default.
|
|
|
|
**How to optimize:**
|
|
- Front-load the most interesting content (not chronological order)
|
|
- Use "open loops" — tease what's coming ("and the third one surprised me...")
|
|
- Vary pacing to prevent monotony
|
|
- Place visual changes every 5-7 seconds
|
|
- Strong hook → the 3-second test determines 70% of retention
|
|
|
|
### Vertical Video Preference
|
|
|
|
LinkedIn now strongly prefers vertical (4:5 at 1080x1350) video on mobile feeds. Vertical videos get approximately 20-30% more feed real estate than landscape.
|
|
|
|
### Native Upload Signals
|
|
|
|
- **Native upload** (direct to LinkedIn) vs link from YouTube/Vimeo: native gets 2-3x the distribution
|
|
- LinkedIn cannot index or optimize external video players
|
|
- Always upload the .mp4 file directly
|
|
|
|
### Caption Detection
|
|
|
|
LinkedIn's 2026 system can detect whether captions are present:
|
|
- Videos with burned-in or SRT captions get a distribution boost
|
|
- Auto-generated captions are now better but still imperfect
|
|
- Professional captions signal quality and effort
|
|
- Recommendation: Always add custom captions (not just auto-generated)
|
|
|
|
### Video + Text Caption Synergy
|
|
|
|
The text accompanying the video matters for algorithm classification:
|
|
- Write 200-400 character captions (not just a title)
|
|
- Include keywords matching your topical authority
|
|
- The caption should add context, not repeat the script
|
|
- Include a question or CTA in the caption for comment engagement
|
|
|
|
---
|
|
|
|
## Pacing Reference
|
|
|
|
### Words Per Second Targets
|
|
|
|
| Speaking Style | WPS | When to Use |
|
|
|---------------|-----|-------------|
|
|
| Slow/Deliberate | 2.0 | Key insights, emotional moments |
|
|
| Normal/Conversational | 2.5 | Default for most content |
|
|
| Energized/Excited | 3.0 | Enthusiasm moments, lists |
|
|
| Rapid (briefly) | 3.5 | Quick asides, building energy |
|
|
|
|
**Default:** Script at 2.5 wps. Most people speak slightly faster than they think.
|
|
|
|
### Length-to-Word-Count Mapping
|
|
|
|
| Duration | Words at 2.5 wps |
|
|
|----------|------------------|
|
|
| 30 seconds | 75 words |
|
|
| 45 seconds | 113 words |
|
|
| 60 seconds | 150 words |
|
|
| 90 seconds | 225 words |
|
|
| 2 minutes | 300 words |
|
|
|
|
### Pause Placement
|
|
|
|
Strategic pauses increase perceived authority and improve comprehension:
|
|
|
|
- **After the hook** — 0.5s pause lets it land
|
|
- **Before each key point** — 0.5s signals "pay attention"
|
|
- **After a surprising statement** — 1.0s pause for impact
|
|
- **Before the CTA** — 0.5s shift in energy
|
|
- **After numbers/data** — 0.5s for processing
|
|
|
|
### Energy Curve Patterns
|
|
|
|
**Single Wave (30-60s):**
|
|
```
|
|
Energy: 8 → 6 → 7 → 8
|
|
hook body take CTA
|
|
```
|
|
|
|
**Double Wave (90s):**
|
|
```
|
|
Energy: 8 → 6 → 7 → 6 → 8 → 7
|
|
hook p1 p2 p3 take CTA
|
|
```
|
|
|
|
**Sustained Wave (2min):**
|
|
```
|
|
Energy: 8 → 7 → 6 → 7 → 8 → 7 → 6 → 8
|
|
hook story p1 p2 p3 transform CTA
|
|
```
|
|
|
|
Never maintain flat energy — monotone kills retention faster than anything.
|
|
|
|
---
|
|
|
|
## Video-Specific Hook Techniques
|
|
|
|
### Lean-In Start
|
|
|
|
Open mid-thought, as if the camera caught you in the middle of an idea:
|
|
|
|
- "...and that's exactly why this doesn't work."
|
|
- "...so I tested it. And the results were not what I expected."
|
|
- "...which brings me to the one thing nobody talks about."
|
|
|
|
**Why it works:** Pattern interrupt. Viewers feel they walked into something already happening — curiosity forces them to keep watching.
|
|
|
|
### Mid-Sentence Open
|
|
|
|
Start speaking before the video visually begins (cut the first 0.5s of silence):
|
|
|
|
- Camera already rolling, you're already talking
|
|
- No "settle in" moment — instant engagement
|
|
- Signals confidence and energy
|
|
|
|
### Visual Hook
|
|
|
|
Use a prop, on-screen text, or unexpected visual in the first frame:
|
|
|
|
- Hold up a whiteboard with a surprising stat
|
|
- On-screen text: "This changed everything" (before you speak)
|
|
- Unexpected background or setting
|
|
- Something physically happening (writing, drawing, showing)
|
|
|
|
### Stat Shock
|
|
|
|
Open with a number that demands attention:
|
|
|
|
- "87% of AI projects fail before reaching production."
|
|
- "I analyzed 200 LinkedIn posts. Only 3 formats actually work."
|
|
- "The average LinkedIn user scrolls past 300 posts per day."
|
|
|
|
### Direct Challenge
|
|
|
|
Challenge a common belief or practice immediately:
|
|
|
|
- "Stop posting motivational quotes on LinkedIn."
|
|
- "Your 'thought leadership' content isn't leading anyone's thoughts."
|
|
- "The advice to 'post daily' is destroying your credibility."
|
|
|
|
---
|
|
|
|
## Video Content Calendar Integration
|
|
|
|
### Frequency Guidelines
|
|
|
|
Optimal mix when incorporating video:
|
|
- **1 video per 4-5 text posts** — video as accent, not primary format
|
|
- **Maximum 2 videos per week** — quality over quantity
|
|
- **Best days:** Tuesday-Thursday (higher professional engagement)
|
|
- **Best times:** 8-9 AM or 12-1 PM in your audience's timezone
|
|
- **Avoid:** Mondays (low engagement) and Fridays (early drop-off)
|
|
|
|
### Content Type Rotation
|
|
|
|
When planning video alongside text posts, alternate types:
|
|
|
|
```
|
|
Week example:
|
|
Mon: Text post (educational)
|
|
Tue: VIDEO — talking head (personal insight)
|
|
Wed: Text post (contrarian take)
|
|
Thu: Text post (framework)
|
|
Fri: (optional) Text post or rest day
|
|
```
|
|
|
|
Never post two videos back-to-back — each needs time to gain traction.
|
|
|
|
### Seasonal Video Opportunities
|
|
|
|
- **January:** Goal-setting, prediction videos
|
|
- **Q1:** Strategy/planning content
|
|
- **Conference season:** Behind-the-scenes, key takeaway summaries
|
|
- **Q3:** Mid-year reviews, pivot stories
|
|
- **Q4:** Year-in-review, lessons learned
|
|
- **Breaking news:** Quick reaction videos (30s format)
|
|
|
|
---
|
|
|
|
## Caption/Subtitle Best Practices
|
|
|
|
### Why Captions Are Non-Negotiable
|
|
|
|
- **85%+ of LinkedIn users watch video without sound**
|
|
- Captions increase watch time by 12-15%
|
|
- Algorithm detects caption presence and rewards it
|
|
- Accessibility compliance (critical for professional content)
|
|
|
|
### Caption Scripting Rules
|
|
|
|
Write captions alongside the script, not after:
|
|
|
|
1. **Sync points:** Mark where each caption line starts/ends
|
|
2. **Line length:** Maximum 2 lines, 42 characters per line
|
|
3. **Reading speed:** 150-180 words per minute (slower than speech)
|
|
4. **Key words:** Bold or highlight critical terms
|
|
5. **Numbers:** Write as digits ("3 steps" not "three steps")
|
|
6. **Punctuation:** Use periods and commas for pacing. Never ALL CAPS for entire sentences.
|
|
|
|
### Caption Formatting
|
|
|
|
```
|
|
Good caption flow:
|
|
"I tested this with 50 companies." [2s]
|
|
"Only 12 succeeded." [1.5s]
|
|
"Here's what separated them." [1.5s]
|
|
|
|
Bad caption flow:
|
|
"I tested this with 50 companies and only 12 succeeded here's what separated them" [5s]
|
|
```
|
|
|
|
### Caption Style Options
|
|
|
|
| Style | When to Use | Example |
|
|
|-------|-------------|---------|
|
|
| **Minimal** | Talking head, clean aesthetic | White text, no background |
|
|
| **Highlighted** | Key words need emphasis | Word-by-word highlight animation |
|
|
| **Boxed** | Busy backgrounds | Dark semi-transparent box behind text |
|
|
| **Branded** | Series content | Your brand colors, consistent font |
|
|
|
|
---
|
|
|
|
## First-Comment Strategy for Video Posts
|
|
|
|
### Why First Comment Matters More for Video
|
|
|
|
Video posts get fewer organic comments than text posts. A strong first comment:
|
|
- Adds context the video couldn't cover
|
|
- Provides a text-searchable summary (SEO)
|
|
- Gives algorithm additional engagement signal
|
|
- Creates a conversation anchor
|
|
|
|
### First Comment Templates for Video
|
|
|
|
**Template 1: Extended Insight**
|
|
```
|
|
For context on this video:
|
|
|
|
[1-2 sentences expanding on a point you didn't have time to cover]
|
|
|
|
The TL;DW (too long, didn't watch):
|
|
- [Point 1]
|
|
- [Point 2]
|
|
- [Point 3]
|
|
|
|
What would you add to this list?
|
|
```
|
|
|
|
**Template 2: Behind the Scenes**
|
|
```
|
|
Behind the scenes on this one:
|
|
|
|
[What prompted you to make this video]
|
|
[Something you cut from the script]
|
|
|
|
Drop a comment if you've experienced this too.
|
|
```
|
|
|
|
**Template 3: Resource Link**
|
|
```
|
|
I wrote a detailed breakdown of this on my profile.
|
|
|
|
Key resources mentioned:
|
|
- [Resource 1 — described, not linked]
|
|
- [Resource 2 — described, not linked]
|
|
|
|
Full article in my Featured section.
|
|
```
|
|
|
|
### Timing
|
|
|
|
Post the first comment **immediately** after the video goes live — within 30 seconds. Have it pre-written and ready to paste.
|
|
|
|
---
|
|
|
|
## Script Output Format
|
|
|
|
Every video script should follow this standardized output structure:
|
|
|
|
```
|
|
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
VIDEO SCRIPT: [Title]
|
|
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
|
|
Type: [talking head / screen recording / slideshow]
|
|
Length: [30s / 60s / 90s / 2min]
|
|
Words: [count] (at 2.5 wps)
|
|
Topic: [content pillar alignment]
|
|
Angle: [from 8 thought leadership angles]
|
|
|
|
━━━ SCRIPT ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
|
|
[Full script with timing markers, visual cues,
|
|
energy cues, and transition markers]
|
|
|
|
━━━ CAPTIONS ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
|
|
[Line-by-line caption text with timing]
|
|
|
|
━━━ POST CAPTION ━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
|
|
[200-400 char text to accompany the video post]
|
|
|
|
━━━ THUMBNAIL ━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
|
|
Expression: [description of ideal facial expression]
|
|
Text overlay: [3-5 words for thumbnail text]
|
|
Style: [minimal / branded / text-heavy]
|
|
|
|
━━━ FIRST COMMENT ━━━━━━━━━━━━━━━━━━━━━━━━
|
|
|
|
[Pre-written first comment — post within 30s of video going live]
|
|
|
|
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
|
|
```
|