Reaction / duet

A duet or stitch-style clip: the creator reacting to something else on screen, with their response as the content.

When to use it

For responding to a trend, a competitor, a review or a piece of news quickly. Its value is speed — a reaction is only worth running while the thing being reacted to is still live.

The thing to get right

Only three beats, deliberately. Reactions are short because the source material is doing half the work, and stretching one to a full ad length drains it.

The format at a glance

• Output — a video ad.
• Structure — 3 beats, each one a separately planned shot.
• Family — Creator & UGC video.
• Recipe id — reaction, which is what you name when you ask for it by hand or from an agent.

Why is it only three beats?

Because the source material is doing half the work. Stretching a reaction to full ad length drains it, so the cap is the format rather than a limitation of it. If you need longer, the thing being reacted to is not strong enough.

Related formats

Formats built around a person on camera — the talking-head, the unboxing, the try-on, the demo. These are the highest-volume paid-social formats on TikTok, Reels and Shorts, and the ones where realism matters more than polish.

• ASMR / satisfying — an oddly-satisfying / ASMR sensory clip
• Get ready with me — a "get ready with me" routine clip while talking to camera
• Q&A / answering — a creator answering a common question to camera
• Street interview — a man-on-the-street / vox-pop style Q&A clip
• Myth-busting — a "myth vs fact" debunking clip
• Storytime — a relatable first-person story told to camera ("the time I…")

The full catalog is on the creative recipes index.

Models that render it

This format is rendered by a video model. The plan picks one unless you name it, and Seedance 2.5 is the default pick. You can ask for another by name, such as Wan 2.7, Veo 3.1 Lite or Kling 2.6. All three render native sound, so a voice comes out of the same take as the picture. The video models page compares every option.

Hermoso plans and renders a reaction / duet from your brand rather than from a prompt you write, composites every on-screen word after the render so the copy comes out exact, and exposes the same format to an AI agent over MCP. How Hermoso renders a format covers all three.