A viewer who clicks on a talking-head video spends the first seconds on a question that has nothing to do with your topic: who is this? Not hostilely — just automatically. A face has appeared and started making claims, and the brain files claims by their source before it weighs them. On television that question never has to be asked, because an answer is already printed across the bottom of the screen. Online it usually goes unanswered. The name is in a channel title somewhere, the credentials are in a description nobody opens, and the video itself carries a stranger. The gap matters most in exactly the videos where the speaking is the product — advice, analysis, teaching, opinion — because the same sentence lands differently from someone whose name and role you know. A lower third is the smallest possible answer to that question: a line of text, present for a few seconds, saying who is talking and, implicitly, why they may be worth listening to. It is not decoration. It is the difference between a claim with a source and a claim without one.
The convention is older than the web, and it was built for exactly this problem. Television news had seconds to establish a stranger’s authority — a correspondent in the field, an expert in a studio chair — and solved it with a caption keyed over the picture: name above, role below, gone before the answer began. The graphic took its name from where it lived, the lower third of the frame, safely under the face and above the edge of the screen. Decades of that discipline trained audiences in a quiet piece of grammar: text at the height of a person’s chest is not part of the scene, it is the broadcaster vouching for who the person is. That training did not expire when video moved to phones. A viewer who has never once thought about broadcast design still reads a name card as institutional shorthand — someone put this here on purpose, someone stands behind this person. Borrowing the convention costs a few seconds of screen time and buys the association wholesale, which is why it has survived every redesign of television news for seventy years.
What makes a lower third work is mostly what it leaves out. Two lines at most: the name, and one line that says why this person on this subject — a role, a handle, a credential, never a paragraph. Set it large enough to read on a phone held at arm’s length, because that is the screen it will mostly be read on, and keep the contrast honest against whatever your background is doing in the opening shot. Then mind the timing, which does more of the work than the design does. A card that appears on the very first frame fights the viewer’s first look at your face; one beat later, it is read as an answer instead of a title screen. And it must leave. A lower third that stays becomes wallpaper, and wallpaper stops being read; the same line that earns trust in its first seconds reads as clutter by its thirtieth. Arrive after a beat, stay long enough to be read twice, go while it is still information. The viewers it was made for will not consciously notice it. That is what a lower third looks like when it is working.
This is an export-time job, not a recording-time one, and that is how Talk2Camera treats it. Type your name once, add a handle if you want one, and the app remembers both across takes — the card is not something you rebuild for every video. At export it burns the card over the opening seconds of the video: it fades in gently, holds for anywhere between two and eight seconds — whatever you set — and fades back out. Like captions, it is applied to the export rather than to the take, so the recording itself stays untouched; re-export the same take for another platform and the card can change length or disappear entirely without anything being re-shot. It also keeps out of the way of the other thing that may live at the edge of the frame: the card sits clear of the corner where the free tier’s mark lives, so the two never collide. The mechanics are deliberately boring. The decision that matters — what the one line under your name should say — is the part the app leaves to you.
The restraint is the point, and it extends to the question of when to use one. If your audience already knows you, a card can feel redundant — but “your audience” is a description of the people who subscribed, not of the people the algorithm hands each video to. Feeds are built out of first meetings: every video is somebody’s introduction to you, and the card is for that somebody, at a cost of a few seconds to everyone else. Use one card per video, at the front, and let it be enough; a name that reappears every few minutes reads as insecurity, not authority. Resist the temptation to make it move — sliding, bouncing and glowing all shift attention from what the card says to the card itself, which is the opposite of its job. The test of a good lower third is that nobody mentions it. Viewers will not tell you that your name card earned their trust, the same way nobody praises a well-hung door. They simply stay a little longer, argue a little less with the person on screen, and could tell you afterwards who was talking. That is the whole trick, and it is enough.