gramps
82960d0ca3
New `youTubeTrainingVideos.md`: the feature list worth filming, split into "ready now" and "after the surge lands" so nothing gets filmed before it exists, plus the format decisions settled on 2026-11-01. Format: 16:9 long-form is the library (search intent, walkthroughs, conversion); vertical Shorts are trailers cut from the same pass. A 16:9 video cannot be a Short, so it is a per-video choice. TikTok is a mirror of the verticals, never a strategy. The numbers behind it: Shorts need 1,000 subs + 10M views in 90 days for YPP versus 4,000 watch hours in a year; Shorts RPM is $0.02-0.08 per 1,000 against $3-8 for tutorials; and one $79 sale nets ~$55.30, which is what 700k-2M Shorts views pay. Shorts are an install engine, not revenue. Film first: privacy blur, auto-ducking, webcam blur.
5.0 KiB
YouTube training videos — what to film, and in what shape
The channel (@llamacasty) is the funnel; Steam is the checkout. These are the videos that make the
funnel work, and the format rules settled on 2026-11-01. Money policy lives in
docs/ai/commercial.md; voice and audience in MARCOM.md (gitignored).
Format — settled
16:9 long-form is the library. Vertical Shorts are the trailers. Not either/or.
- A 16:9 video cannot be a Short: Shorts are vertical or square, up to 3 minutes. So this is a per-video choice, not a setting on one file.
- Long-form carries search intent ("how do I…"), holds a walkthrough, earns trust, and is what converts a viewer into an install. This is the priority when only one can be made.
- Shorts are 20–45 seconds, one feature, no setup, ending on "full walkthrough on the channel". Their job is reach, not teaching.
- One production pass, two outputs: record the 16:9 tutorial, then cut a vertical from it — or better, record the vertical separately with LlamaCasty's own 9:16 tier and show the app doing it.
Why Shorts are not the revenue plan
- Eligibility: both count toward YPP, but Shorts need 1,000 subs + 10 million Shorts views in 90 days; long-form needs 1,000 subs + 4,000 watch hours in 12 months. For a niche Windows tool, 4,000 hours is a realistic year; 10M Shorts views is not.
- RPM: Shorts pay roughly $0.02–0.08 per 1,000 views in pooled ad revenue; long-form tutorials run $3–8 per 1,000. 100,000 Shorts views ≈ $2–8. 1,000 long-form views ≈ $3–8.
- The product is the revenue. One $79 Pro sale nets ~$55.30 — roughly what 700,000–2,000,000 Shorts views pay. Shorts are a subscriber-and-install engine, not an income line.
TikTok
Repost the verticals; never build for it. Zero marginal cost once a Short exists, and the odd spike is free reach. But the feed's intent is entertainment, links are buried, and a Windows streaming tool is a hard sell there: it is a mirror, not a strategy.
Cadence
One long-form tutorial plus two Shorts cut from it, per feature. That is roughly the 8–10 asset
launch set already in MARCOM.md.
Features worth filming — ready now
| Feature | The video it earns |
|---|---|
| Five built scenes with art | "Your stream is built before you open the app" — the KISS thesis in 90 seconds |
| Display / window capture with game detection | "It found my game by itself" |
| Webcam placement: drag, resize, mirror, rect/round, border colour + animation | "Make your cam look intentional" |
| Webcam background blur / removal | the most demo-able feature on the list |
| Privacy blur layer (box or circle, covers everything, brand exempt) | "The moment you need to hide something" — nobody expects it |
| Zoom magnifier (anchor or follow, Ctrl to close) | "Look at this part of my screen" |
| Vertical 9:16 tier with locked camera + chat | "Stream Shorts-style from one app" |
| One mic control: volume, mute, meter | the anti-mixer pitch |
| Voice polish: bass, treble, gate, compressor | A/B demo — sells without explanation |
| Auto-ducking | the 20-second clip that converts: loud game, voice on top |
| Limiter | "why your stream never clips" |
| TRAX background music with its own meter | "music without a copyright strike" |
| Mono fold | accessibility, and it is the creator's own story |
| Lip-sync offset | "fix audio drift in one slider" |
| One-click go-live, settings auto-filled | the headline |
| Test stream (private, simulated chat) | "practice without an audience" |
| Chat overlay (fonts, colours, fade, never stored) | privacy + looks |
| Alerts: built-in clip, custom video, sound byte, ticker | "your Super Chat moment" |
| Social bar | "your handles, validated" |
| Scoreboard layer | session tallies, monetization-adjacent |
| Drag/resize with undo, transitions (cut/fade/move/wipe) | polish demo |
| Local recording + rename on stop | "record and keep it" |
| YPP journey tracker | the monetization-awareness hook nobody else ships |
Film first: privacy blur, auto-ducking, webcam blur. The two most demo-able, and the blur is unique to this app.
Features to film after the surge lands
Text layer · media file picking · scheduled streams · stream resilience / reconnect · bandwidth step-down · publish-recording · YPP slice 2 · visibility picker. Filming these early means filming something that does not exist yet.
Channel roles — do not blur them
@llamacasty— training and how-to, promo videos, the quiet "here's the tool I use".- Steam — distribution, wishlists, stats, reviews, the store page.
- GitHub
llamacasty/support— issue tracking only, deliberately behind an account. llamacasty.com— the model (free app, paid edition), the store link, the privacy policy. No price, ever.
Support → content loop
A repeated support question is not a ticket, it is next week's video: answer once by email, record
the answer, and the video answers it forever. Policy in docs/ai/commercial.md → Support.