A watermark that sits in a fixed corner is a promise with a short life. Every phone ships with a photo editor, every photo editor leads with a crop tool, and a crop is one gesture: drag the edge of the frame inwards by a few dozen pixels and the corner with the logo is gone. The video loses a sliver of height that nobody will miss, and the free tool loses the only payment it ever asked for. This is not an obscure skill. It is the first thing almost everyone tries, roughly a minute after their first marked export, and it works so reliably that a static corner mark is best understood as decoration rather than protection. Tool makers know it, which is why the arms race has escalated in ugly directions: marks dragged into the centre of the picture where no crop can reach them, marks inflated into banners, marks tiled diagonally across the presenter’s face. Each of those survives the crop by ruining the video — the brand is protected at the expense of the only thing the brand was ever attached to.
It is worth stepping back and asking what a free tier is actually for. The maker needs something back from users who pay nothing, and there are only two honest currencies: features or output. Charging in features means crippling the tool — capping exports at 720p, locking the decent codec away, cutting recordings off at three minutes, counting your exports. That approach quietly poisons the evaluation, because what the free user tests is not the product; it is a worse product wearing the product’s name. You cannot find out whether an app fits your work while it keeps refusing to do the work. Charging in output is the opposite deal: every feature works, at full quality, with no meter running — and what leaves the app says what made it. That is closer to word of mouth than to a toll: a small credit in the corner of a video that would not exist without the tool, seen by exactly the people most likely to want the tool. The trade stays fair as long as two things hold. The mark must not ruin the video, and it must not come off in one gesture. Static corner marks fail the second test. Face-covering banners fail the first.
Talk2Camera’s answer is a mark that refuses to sit still. Free exports carry a small app mark that moves between the four corners of the frame, taking a new corner every six seconds and never sitting in the same corner twice running. That single property redraws the economics of removal. Cropping away a static mark cost one gesture and a sliver of picture; cropping away this one means cutting all four corners at once, which is no longer a trim but a visible shrinking of the whole frame — or retouching the mark out frame by frame, which costs more in effort than the paid tier costs in money. The viewer’s experience, meanwhile, barely changes. The mark stays small and keeps to the corners, where eyes do not rest; it moves rarely enough that it never reads as motion, and often enough that it never quite becomes furniture. People who watch the video remember the video. People who try to lift the frame clean discover that they cannot.
The details are where the design earns its keep. The mark opens in the bottom-left, and the choice is deliberate: on the platforms these videos are made for, the bottom-left is the one corner the interface leaves alone. The right edge of a vertical feed is a column of buttons; the lower middle belongs to captions; overlays crowd the top of the screen. A mark that started under any of those would be invisible at best and doubled-up at worst. Starting bottom-left also means the first frame — the frame feeds tend to lift a thumbnail from — looks composed rather than stamped on. And the walk is deterministic: the same take produces the same schedule of corners every time, so the preview on your screen shows exactly the walk the export will carry. That sounds like a small courtesy until you have used tools where it is not true — where the exported mark lands somewhere the preview never showed, over a face or a caption, and you find out after uploading. Here, what you approved is what you published. There is no roll of the dice between the preview and the file.
Two last properties complete the honesty of the scheme. The mark is applied at export, and only at export — the recording on disk is always clean. Nothing is baked in at capture, so the take you shot in January is not hostage to the tier you were on in January: upgrade later and the same recording exports unmarked, because it was never marked in the first place. And the paid tiers do not simply delete the mechanism — they hand it over. The same slot, the same corners, the same six-second walk, now carrying the creator’s own logo instead of the app’s. What protected the free tier turns out to be a feature in its own right: a moving mark is as hard for someone else to crop off your video as it was for you to crop off the app’s. A free tier built this way makes a plain deal — every feature, full quality, a small moving credit — and keeps both halves of it. That is rarer than it should be, and worth preferring, whatever tool you end up choosing, over a free tier that hobbles the work and calls it generosity.