A codified TikTok formula that's produced multiple seven-figure-view campaigns across DTC, hospitality, and B2B.

The context

Brand teams frequently request "viral TikTok content" without a clear definition of what that means in practice. The standard agency response is creative experimentation with limited repeatability. We approached TikTok as a system to study and reverse-engineer.

After a sufficient sample of both successful and underperforming campaigns, three consistent variables emerged: the first 1.5 seconds, the edit pacing, and the second-watch hook.

The framework (high level)

• Hook structure (first 1.5s): pattern-interrupt or named-stakes question. The most common failure mode is greeting the viewer; viewers don't want a greeting, they want a reason to stay.
• Edit pacing: cut every 0.8–1.4s in the first six seconds. Stop the doom-scroll reflex.
• Second-watch hook: a detail (visual or auditory) that rewards rewatching. This is the loop-rate driver and it's what TikTok's algorithm actually scores.
• Hashtag stack: one broad, two niche, one branded. Not more.
• Distribution timing: per-vertical day-parting; finance ≠ hospitality ≠ fitness.

Repeatability

The framework lives as an SOP document plus a Claude prompt that critiques storyboards against it before we shoot. Junior editors can use it on day one.

Outcome

• Multiple seven-figure-view campaigns across very different industries
• Predictable, not lucky — win rate is the metric we manage, not raw views
• Same framework now drives the Cowork-automated content pipeline (see HAZE Cowork Automation Pipeline)