ChatGPT Work's task-to-deck launch film
A quiet capability announcement turns one spoken workplace request into visible research, a finished deck, and desktop app work.
Field notes / why it works
The capability is readable by about 0:02.5, then the film shifts from dark announcement graphics to bright product proof. One 16-word request at 0:10 becomes a finished deck and a visible slide edit near 0:34.8, the top replay peak; a second task around 0:41.2 creates another replay peak before the spreadsheet and document close-ups.
Plate 01 / The format in one breath
A 16:9 product launch film begins with a dark, star-like field and a staged two-part announcement. It switches to bright, nearly full-frame product UI: one short spoken request turns into source gathering, a finished artifact, and a second delegated task. The film sells a sequence of outcomes, with very little narration.
Why it works
- The product and new capability are legible by 2.5 seconds, while the first cut waits until 3.4 seconds. The restrained hook gives the claim room to land.
- One request at 0:10 leads to a deck preview by 0:28.9 and an edited slide around 0:34.8, the strongest replay peak. Viewers can inspect the result rather than take a voiceover's word for it.
- A second task arrives at 0:41.2, near another replay peak at 0:41.6–0:42.2. This extends the story from generating an artifact to operating across work apps.
Plate 02 / Format card
| Platform · aspect · length | YouTube · 1280×720, 16:9 · 56.0 s at 30 fps |
| Pace | 10 measured shots, 9 cuts (9.6/min); 9 internal changes, about 19.3 picture changes/min; mean shot 5.6 s, median 3.0 s. The 26.9 s main demo contains six internal changes. |
| Script | 16 measured spoken words; 32 wpm across the whole film, about 145 wpm while speaking. One task request, then interface and headline text. |
| Voice | Brief, medium-register adult workplace requester; conversational and purposeful. No continuous narrator. |
| On screen | Designed motion backdrop and kinetic type, then composited desktop/product UI, document previews, and app windows. No presenter. |
| Captions | No continuous subtitles. Large center headline cards; task text appears inside the composer. |
| Sound | Continuous tonal/music-like bed, roughly 121 BPM; integrated −16.1 LUFS, true peak −1.8 dBFS. Cut accents are not consistently measurable. |
| Structure | Announcement → benefit → mode switch → task input → source and artifact proof → busywork claim → follow-on action → document proof → dispersed end frame. No explicit spoken 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