Loom AI's guided feature tour
A 38-second feature launch that turns a shareable video page into a guided tour of AI-generated context and team actions.
Field notes / why it works
The product appears by 0.25 seconds and a human webcam bubble by 0.5 seconds, giving the spoken promise immediate visual proof. Internal UI changes at 15–28 seconds show generated content, tasks, and transcript cleanup while maintaining a coherent page; a question at 31.7 seconds resets attention for the final capability and CTA. The two measured shots contain eight internal step changes, so the tour feels active without relying on frequent cuts.
Plate 01 / The format in one breath
A warm, fast feature announcement: open on the product already in use, show an assistant-like feature working inside a shareable video page, then move from content creation to team follow-through and close with one extra capability. The promise arrives in the first spoken sentence at 0:00; the product proof starts by 0.25 s.
Why it works
- The first second shows a product window and a human webcam bubble, so the broad opening claim has visible proof before the viewer has heard the full feature name.
- The composition changes within long continuous shots: browser card at 0.25 s, dashboard at 1.0 s, page context at 4.3 s, AI panel by 7.1 s, task and transcript tabs at 22.25–28.2 s. The viewer keeps getting a new UI destination without a rapid hard-cut montage.
- The feature list escalates from generated context (11.3–20.5 s) to assigned tasks (21.3–26 s) to cleanup and multilingual reach (26–37.1 s), with a question at 31.7 s before the last reveal.
Plate 02 / Format card
| Reference | |
|---|---|
| Platform · aspect · length | X · 1280×720, 16:9 · 38.4 s · 29.97 fps |
| Pace | 2 measured shots, 1 certain cut at 8.55 s; 1.6–9.4 cuts/min including uncertain changes; 8 internal step changes; 14.1 picture changes/min; mean measured shot 19.2 s |
| Script | 89 spoken words; 144 wpm across runtime, 169 wpm while talking; 8 sentences averaging 11.1 words; 1 question; roughly 6.7 uses of “you” per 100 words |
| Voice | Upbeat medium-register promotional narrator, conversational and animated |
| On screen | About 90% designed product interface and gradient motion, 10% human presence in small circular webcam overlays (estimated from sheets) |
| Captions | No burned speech subtitles; readable text is UI copy, tab labels, and a short feature title |
| Sound | Continuous tonal, music-like bed (measured); exact instrumentation unverified; −15.4 LUFS integrated, −1.9 dBFS true peak |
| Structure | 5 beats: immediate claim → UI proof → AI-generated context → team workflow → surprise feature and 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