Motion design in avatar reels — cards, numbers and lists placed automatically by meaning
Pick an editing style before you approve a reel — Craft decides where to show a big idea, a number, a list, a before/after or a call to action, and adds camera pushes and transitions. Your face stays in the frame, subtitles match the style.
A plain avatar reel is a talking face with subtitles. With motion design Craft edits it like an editor would: a big hook in the first seconds, text cards on the key ideas, spoken numbers counting up, enumerations turning into lists, the camera pushing in on accent words and changing framing on every new phrase, and quick whip transitions between scenes.
You don't place anything by hand: Craft reads the reel's text and decides which graphics go where. You only pick the style.
⚠️ Motion design is in a closed test right now — not every avatar-reels participant sees the style picker yet.
How to turn it on#
- Open the reel in Content factory → Avatar reels and check the text as usual.
- In the Editing block, pick a style — each one has a sample picture.
- Click Approve. The reel goes through the "Planning the edit" step (usually 5–20 seconds) and is then built as usual.
Your choice is remembered: the next reel opens with the same style.
Styles#
| Style | What it looks like |
|---|---|
| ZamZam | Smoky glass and a gold accent; cards sit on a blurred frame, your face still visible under the text |
| Editorial | Magazine-strict: light paper and black, big type, the previous line fades to grey |
| Tech orange | Dark screen, orange floods on accent words, ticking numbers, a terminal-style line |
| Engraving | White page, serif type, thin lines that draw themselves, a brick-red accent |
| Interface | App-like: a message being typed in a chat, buttons, lime floods |
| Bauhaus | Circles, triangles and squares on a grid in red, yellow and blue |
| Subtitles only | As before: subtitles with no cards or cutaways |
What Craft does in the reel#
- Hook — big for the first 2–3 seconds, on a dark gradient below your face. Subtitles are hidden meanwhile so they don't compete with it.
- Big idea — the key phrase of a passage; lines appear exactly on the words where you say them.
- Accent word — one word or number across the screen.
- Counter — if a number is spoken (money, percent, deadlines), it counts up from zero.
- List, steps, before → after, quote, bar comparison, chat, path from A to B.
- Call to action at the end — if the last phrase asks people to message, follow or save, it becomes a button with an arrow.
- Cutaway picture — occasionally, on a concrete object (money, a phone, a city). A second AI look checks the picture: no lettering and no close-up faces. If it doesn't fit, your face stays on screen there.
On full-screen cards you stay in the frame — in a small window in the bottom corner (with ZamZam, under the blurred glass).
Good to know#
💡 On cards Craft writes short, like a headline, not your speech word for word — so it reads in a second.
⚠️ Graphics take no more than half of the reel: there is always some of your face between cards — viewers trust that more.
✅ All graphics stay inside the safe margins: Reels, TikTok and Shorts crop the edges of the frame, so neither text nor cards come close to them.
⚠️ If the edit can't be planned within a couple of minutes or something goes wrong, the reel is still delivered — with regular subtitles. You don't lose the credits.
- No numbers in the text — no counter; no enumeration — no list. Want more graphics? Write with specifics: numbers, steps, "before and after".
- The reel's price is the one shown before you approve.
From Claude or ChatGPT#
If Craft is connected to your AI, ask right in the chat: "Make an avatar reel from this text in the Editorial style". The AI will show you the text and the price and, after your "yes", build the reel in that style.
FAQ#
Can I adjust where the cards appear? Not yet: Craft places graphics by the meaning of the text. If you don't like it, change the style or the text and build the reel again.
Does it work with an AI character and "Voice only"? Yes. With a character the cards go underneath — the character always stays on top. In "Voice only" there is no face window; graphics go over the original video.
Why does the card say something different from my speech? That's on purpose: the screen shows the short idea, the ear gets the full phrase. Together that holds attention better than repeating the same words.