Agents, Explained Through the Work
A live-action product launch turns AI jargon into concrete approval, policy, messaging and dashboard workflows through fast dialogue and oversized type.
Field notes / why it works
A sparse logo set turns into huge type by 1.4 seconds, then a nine-cut first ten seconds deliberately overplays the jargon before a plain question resets the film at 0:08. Each answer earns its claim with a readable artifact: approval chips near 0:27, a policy document near 0:30, a phone reply near 0:59 and a dashboard held for 8.38 seconds from 1:06.5. A candid work-level payoff at 1:21–1:32 gives the features a human reason to matter.
Plate 01 / The format in one breath
A designed, live-action product launch explainer starts with a deliberately overused industry term, interrupts itself with the audience's basic question, and lets a rotating group of approachable employees explain the product through specific workflows. Big editorial type and restrained UI inserts turn an interview-led film into a graphic launch piece. The viewer gets an answer to what it is, what it does, how it handles exceptions, and why the team cares.
Why it works
- The opening holds one centered set for 3.46 seconds but changes the picture at 1.35 seconds: a logo gives way to huge words; a rapid montage at 3.5–7.5 seconds makes the jargon feel excessive before the question lands.
- A hard visual reset at 8.5–10 seconds asks the viewer's question; later question cards at about 23, 48 and 79 seconds restart attention before new proof sections.
- The claims acquire observable forms: approval chips around 0:27, a policy document at 0:30, a phone reply at 0:59 and a dashboard around 1:07. The final speaker translates those features into relief from repetitive work at 1:21–1:32.
Plate 02 / Format card
| Platform · aspect · length | X · 1280×720, 16:9 · 98.7 seconds at 24 fps |
| Pace | 54 shots; 53 cuts; 32.2–33.4 cuts/min including uncertain changes; 42 picture changes/min; mean shot 1.83 s, median 1.29 s |
| Script | 290 spoken words; 191 wpm overall, 203 while talking; 13 measured sentences, four questions |
| Voice | Several conversational staff archetypes, mixed genders; quick and animated, with distinct question and answer roles |
| On screen | Mostly interview and seated wide shots; intermittent full-screen type, composited keywords, document/phone/dashboard inserts (approximate from sheets) |
| Captions | Small white sentence subtitles in translucent charcoal rectangles near the bottom, distinct from the huge editorial headline type |
| Sound | Quiet tonal bed about 14.7 dB under speech; no consistent cut-hit pattern; −22.2 LUFS integrated, −6.7 dBFS true peak |
| Structure | Jargon cold open → basic definition → approval workflow → policy help → measurement → human payoff → logo card; no 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