AI scene swaps with the original beside them
A vertical product demo pairs transformed footage with its source, then shows the upload workflow and a location wipe.
Field notes / why it works
At 0:00 the transformed result sits directly above the labeled original, so viewers can verify the effect immediately. New person and setting swaps around 0:04–0:15 keep the proof developing within the same readable layout. The interface at 0:30.8 and location wipe at 0:39.6 make the result feel achievable as well as surprising.
Plate 01 / The format in one breath
Show the transformed result above the source footage immediately, with both performances moving in sync. Escalate from one visual swap to a different setting, briefly name practical uses, then show the small input-to-output workflow and invite a comment.
Why it works
- At 0:00 the neon swapped object sits directly above a plain original. The viewer can judge the claim before learning the steps.
- The same two-panel comparison carries a person swap around 0:04–0:07 and a room-to-office change around 0:11–0:15, so the proof keeps changing while the layout stays legible.
- A workflow glimpse at 0:30.8 and the location wipe at 0:39.6 turn the spectacle into an actionable process.
Plate 02 / Format card
| Reference | |
|---|---|
| Platform · aspect · length | TikTok · 9:16 · 720 × 1280 · 30 fps · 44.1 s |
| Pace | Detector: 3 major shots, 2 cuts, 2.7 cuts/min, 14.71 s mean shot. The panes contain changing demonstrations that the cut count does not capture. |
| Script | 145 spoken words · 198 wpm overall, 228 wpm during speech · 9 measured sentences · 10.3 uses of “you” per 100 words |
| Voice | One animated, conversational, medium-register adult presenter; direct address and rising inflections |
| On screen | Comparison and transformed live-action proof for ~27 s; product workflow for ~8 s; final example plus presenter inset for ~9 s |
| Captions | No continuous word captions observed; persistent headline, “Original” source labels and a short feature callout |
| Sound | Continuous tonal bed measured beneath speech; platform credits an original sound. Effects are not identifiable from the supplied evidence. |
| Structure | Result first → escalating swaps → use cases → three-step workflow → comment CTA |
Tools you'll need: a video model (Veo, Kling, Grok Imagine) · a camera shoot · screen recordings of the product · Remotion for motion graphics · ElevenLabs or a recorded voice · Claude alone: partly
Frame by frame / measured
Stills taken at measured moments, each labelled with its time. Agents get the same sheets over MCP.
Pro / 13 plates + recipe
The first plates are open. The full dissection is Pro.
Get the timed shots, type and sound specs, and the runnable recipe.
Open the full blueprint- Plate 03 / The hook, frame by frame (first 3 seconds)Pro
- Plate 04 / Structure (beat sheet)Pro
- Plate 05 / Shot grammarPro
- Plate 06 / Visual style guidePro
- Plate 07 / Voice and deliveryPro
- Plate 08 / Sound designPro
- Plate 09 / PackagingPro
- Plate 10 / Script templatePro
- Plate 11 / Production planPro
- Plate 12 / PromptsPro
- Plate 13 / Edit recipePro
- Plate 14 / QA checklistPro
- Plate 15 / Make it yoursPro
- Recipe / runnable promptPro