A Lost Lamb's Imagined Journey
A story-led Gemini demo turns a lost plush toy into a sequence of generated travel images, wild videos, and a personal message.
Field notes / why it works
The lost-toy problem is spoken by 0:01 and shown with a reference photo by about 0:02, giving the interface an immediate human purpose. Prompt/result intercuts accelerate into 0.9–1.4-second travel images, then a rapid action burst at 0:33–0:35 that aligns with the strongest replay peak. The 7.59-second final generated scene slows down for a personal message before the product card.
Plate 01 / The format in one breath
An adult tries to soften a child's disappointment about a missing stuffed toy. A dark chat interface becomes a stage for turning one photo into a playful fictional trip: still images, then an absurd action video, then a gentle personal message. The product demonstration is carried by the story rather than a feature list.
Why it works
- The problem arrives in the first spoken line at 0:01, while the toy photo appears over the interface at about 0:02. The viewer understands the emotional stake before any feature explanation.
- The examples expand from search/help (0:08–0:12) to travel images (0:21–0:29) to moving adventures (0:30–0:37), giving each generated result a larger visual payoff.
- The rapid parachute-to-bull sequence at 0:32–0:35 sits at the strongest measured replay peak, 0:32.9–0:33.5. The narrator then pulls the idea back at 0:35–0:38, which keeps the joke tied to the child's original problem.
Plate 02 / Format card
| Reference | |
|---|---|
| Platform · aspect · length | YouTube · 16:9 · 60.0 s · measured source 640×360 at 23.98 fps |
| Pace | 26 shots, 25 cuts/min, 31 total picture changes/min; mean shot 2.31 s, median 1.83 s. The opening holds 4.59 s; the middle accelerates to several 0.4–1.4 s shots. |
| Script | 136 transcribed words; 140 wpm across the whole film, about 233 wpm within voiced segments; 34 short utterances averaging 4 words. Some auto-transcribed words appear unreliable. |
| Voice | Two conversational adult roles (inferred from the dialogue): one concerned, one improvising; medium pitch, animated rise and fall. No announcer voice. |
| On screen | Roughly half dark product UI, one third generated still/video examples, remainder transition/end card (estimated from shot durations and sheets). |
| Captions | No dialogue captions. UI prompt text, result labels, brief end-card copy only. |
| Sound | Continuous tonal/music-like bed under dialogue; no consistent hit on cuts. −16.9 LUFS integrated, −1.0 dBFS true peak. |
| Structure | Lost-object setup → product clue → still-image play → motion escalation → self-correction → personal message → logo/CTA. |
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