Where the call to action lives: overlay or in the frame
Two placement help pages describe text the platform lays over an ad: call-to-action buttons, headlines and captions. One generator documents the other half, a final card inside the video carrying brand, product name, tagline and call to action, because the shots themselves cannot be relied on to spell anything. As of 2026-10-13.
| Item | What is published | Detail |
|---|---|---|
| Google Ads, in-feed | Two call-to-action buttons in a Video action campaign | One to the final URL, one to the video |
| Google Ads, mobile Masthead | A headline, description texts and an external call to action | Channel name and icon pulled in automatically |
| SceneMixer, end card | Brand, product name, tagline and call to action on the last card | In the ad's dialogue language; edited without reshooting |
| TikTok, ad caption | Up to four lines taken from the organic caption | Uniform white font; no links, @ symbols or hashtags |
Inclusion rule. Text elements described on a placement's own help page or on a generator's own guide. Visual elements a platform renders without documenting them are not counted. Order. Alphabetical by surface.
1The platform owns the clickable part
Every clickable element on these pages is added by the platform: two buttons on an in-feed ad, an external call to action on the mobile Masthead. None of it is part of the uploaded file, and none of it can be designed by the advertiser beyond a headline and a few lines of description.
TikTok's caption is stricter still. It is taken from the organic post, limited to four lines, set in one font that cannot be changed, and cannot carry a link, a mention or a hashtag.
2The video has to carry the name
What the platform cannot add is the brand inside the picture. A generated ad has a specific problem there: the guide for one generator states plainly that video models do not write legible text, so a logo or a product name asked for inside a shot is unreliable.
Its answer is a card at the end of the cut with four lines, any of which can be left blank, edited after the ad is made without reshooting the shots, and written in the ad's dialogue language rather than in English by default.
3Overlays and a card compete for the same space
TikTok's safe zone shifts with the dimension of the video and the length of the caption. A card whose lines sit near the bottom of a vertical frame is exactly where a long caption lands, so the shorter the caption, the more of the card the viewer reads.
That is a planning point rather than a specification, and it is one reason to keep the card's lines short: a brand and a call to action survive an overlay that a full tagline may not.
- Call to action on in-feed adsIn a Video action campaign, in-feed video ads include two call-to-action buttons, one linking to the final URL and one to the video
- Call to action on the mobile MastheadThe mobile video Masthead carries a customizable headline, description texts and an external call to action, and pulls in the channel name and icon
- Text TikTok lays over an adAd captions come from the organic post, show at most four lines in a uniform white font, and support no clickable links, @ symbols or hashtags
- Where on-screen elements can sitThe safe zone depends on the video's dimension, the caption length and any additional formats used
- Where the call to action sitsThe final cut ends on a card carrying brand, product name, tagline and call to action, editable without reshooting the shots
- Why the card existsThe guide states that video models do not write legible text, so the end card is where the brand name reaches the viewer
- Language of the cardThe card is written in the ad's dialogue language
4Sources
Format requirements come from Google Ads Help and tool limits from each vendor page listed on published limits, both read 2026-10-13. The rule against mixing the two is on method. Nearby: Product photo, TikTok specs, Shape by platform.