A Campaign Lead Delegates to AI Agents
A 33-second product film turns campaign delegation into a human-agent exchange, mixing real UI steps with literal visual cutaways.
Field notes / why it works
The wordless opening establishes a human and an agent gallery in two cuts before speech begins at 0:02.82. Concrete UI requests from 0:03 to 0:24 make the capability legible, while three short visual metaphors reset attention; the moving fire truck at 0:24.4 turns the final alert into a memorable gag. A brief agent acknowledgment at 0:26.3 closes the exchange before the 4.7-second brand card.
Plate 01 / The format in one breath
A polished, 33-second product film makes delegation feel like a tiny workplace story. A professional starts a task in the interface, adds increasingly specific follow-ups, gets a helpful agent response, then the brand resolves the partnership claim. The spoken lines sound like everyday work, while the interface and literal b-roll carry the explanation.
Why it works
- The first 2.8 seconds show a person and an agent gallery before anyone speaks; two cuts establish the human-plus-software relationship without exposition.
- Product capability is demonstrated through a sequence of concrete instructions at 0:03–0:24: plan, flag, nudge, escalate, and involve another team. It feels more tangible than a feature list.
- The visual metaphors at 0:09.5, 0:15.8, and 0:24.4 punctuate long UI holds; the fire truck makes the final contingency line memorable.
Plate 02 / Format card
| Platform · aspect · length | YouTube · 16:9 · 32.9 s · source 640×360 at 25 fps |
| Pace | 10 detected shots; 9 confirmed cuts, 16.4 cuts/min. Four changes around 0:24.8–0:25.8 are uncertain, giving a possible upper bound of 23.7 cuts/min. Four internal graphic changes; about 23.7 confirmed picture changes/min. Mean shot 3.29 s, median 1.92 s. |
| Script | 59 transcribed words; 128 wpm across full runtime, about 221 wpm during active speech; 12 short sentences averaging 4.9 words. One question, no direct “you.” Some product terms in the automatic transcript appear inaccurate. |
| Voice | Conversational professional, medium pitch, animated inflection; spoken workplace dialogue, no presenter to camera. |
| On screen | Approx. 45% product UI/graphics, 30% illustrative live-action b-roll, 20% person with laptop, 5% logo card (estimated from shot durations). |
| Captions | No conventional spoken-word subtitles visible. Short editable commands appear inside translucent UI prompt boxes and agent cards. |
| Sound | Continuous, fairly prominent tonal music-like bed beneath dialogue; no clear cut-synced effects measured. |
| Structure | Wordless human/product setup → task request → layered contingencies → comic visual peak → acknowledgment → brand end card. |
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