Current:
New: This reel provides a micro-level, *visual* retention tactic focused specifically on text caption styling (font size, style, shadow, two-word cuts) for the *first 10 seconds* of a short-form video. The existing 'MrBeast' plan likely focuses on broader narrative structures, pacing, and visual dynamics, but not this granular detail of text caption strategy for engineered nonchalance/authenticity.
Implement aggressive two-word caption cutting with Basic shadow style to boost short-form retention across DDB and TFWW content.
Update DDB editing SOP to use two-word caption cuts in opening hooks, Basic style with custom small sizing (not the default large TikTok captions)
Apply aggressive caption pacing and high-contrast Basic style to any 'website wizard' educational Reels to increase completion rates
Use this caption style for AIAS feature showcase videos—nonchalant aesthetic aligns with 'AI that sounds human' positioning
We should adopt the two-word caption pacing immediately for DDB hooks—it's a perfect fit for the 'unpolished expert' brand positioning. The 'engineered authenticity' insight is particularly valuable for high-ticket service marketing.
The 'two words at a time' rule is elite. Been testing similar pacing for boring B2B content and the retention jump is wild. Definitely A/B testing this vs the default TikTok sizing.
What it is: A specific caption editing workflow in TikTok/CapCut: auto-captions → custom style → small font size → Basic style (black outline) → aggressive two-word cutting for pacing. Targets the 'messy/nonchalant' aesthetic that performs well on short-form platforms.
How it helps us: Directly applicable to DDB content strategy and any TFWW/Lead Needle short-form video content. Provides concrete technical specs for caption formatting that we can implement immediately in CapCut for Reels/TikToks. The 'two words at a time' pacing technique is especially useful for explainer content.
Limitations: The claim that this single technique drives millions of followers is overstated—editing is one variable among many (content quality, timing, algorithm luck). Also, this is mobile-specific editing; our current ReelBot pipeline processes content server-side, so we'd need to adapt these specs to ffmpeg/Python if automating captions.
Who should see this: Dylan (DDB content creation) and any team member producing short-form vertical video for TFWW or AIAS marketing
| Step | Prompt | Completion | Cost |
|---|---|---|---|
| analysis | 14,859 | 2,387 | $0.0119 |
| similarity | 1,403 | 432 | $0.0005 |
| plan | 11,297 | 6,319 | $0.0190 |
| Total | $0.0314 | ||