AI App Building, From Prompt to Workflow
A 59-second Airtable launch film that turns a spoken AI promise into a sequence of built apps, populated tables, and a fast outcome montage.
Field notes / why it works
The first 10.6 seconds hold on a real office monitor, grounding the promise before the film enters a designed interface. A 20.5-second middle section shows requests becoming populated apps and tables, while the pace jumps to roughly one-second product glimpses at 47.4–51.5 seconds before the brand close.
Plate 01 / The format in one breath
A calm voice names a familiar workflow problem over a real desktop scene, then the camera enters the interface. One natural-language request becomes a working business app; more examples widen the claim from a single feature to a platform. The final seconds compress several outcomes into a rapid montage and a branded promise.
Why it works
- The first 10.6 s stay on a physical monitor while the narration moves from problem to solution, making the opening feel grounded before the graphic sequence begins.
- The 22.2–42.7 s section spends 20.5 s on proof: a request, app creation, then a second data workflow. It shows states rather than listing capabilities.
- The pace accelerates from no cuts in the first 10 s to six one-to-three-second shots around 47.4–54.5 s, turning the closing claim into a visual payoff.
Plate 02 / Format card
| Reference | |
|---|---|
| Platform · aspect · length | X · 16:9 · 58.5 s · 1280 × 720 · 29.97 fps |
| Pace | 13 shots, 12 cuts, 12.3 cuts/min; 12 in-shot changes, 24.6 picture changes/min; mean shot 4.5 s, median 3.0 s. The long 20.5 s proof section contains several state changes. |
| Script | 150 spoken words; 159 wpm overall, 202 wpm while speaking; 13 sentences averaging 11.5 words. No spoken questions. |
| Voice | One clear, confident, animated adult narrator in a medium register; no person appears speaking. |
| On screen | One live-action office-monitor setup, then about 48 s of designed product UI, maps, cards, and logo. No talking head. |
| Captions | No editorial subtitles or kinetic captions. The interface itself carries short prompts, labels, rows, and responses. |
| Sound | Continuous tonal, music-like bed (measured); no clear recurring cut effects. Measured −18.3 LUFS integrated, −2.2 dBFS true peak. |
| Structure | Problem → embedded-AI promise → product name → app-building proof → broader agent/data proof → fast outcome montage → brand lockup. |
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