Camera on or off: when to show your face in a screen recording
A face in the corner adds warmth and trust. It also covers part of the screen and gives you one more thing to get right. Decide recording by recording, and when the camera does earn its place, set it up so it hides nothing.
By the VeoRec team · · 12 min read
In short
Show your face when the recording carries tone or trust: a first message to a client, feedback that could sting, bad news, a welcome. Leave the camera off when the screen carries the message and every pixel matters: bug reports, code walkthroughs, dense tutorials. When the camera is on, keep the bubble small, in a corner the content does not use, lit from the front, with a tidy or blurred background.
- Camera on for tone and trust (clients, feedback, bad news, introductions); camera off when the screen is the message.
- Research on instructional video suggests a face helps engagement more than understanding; showing where you are looking helps understanding.
- Use your cursor, click highlights and on-screen drawing to direct attention, with or without a face.
- Keep the bubble small and in a corner your content does not use; on websites that is often not the bottom right.
- Light from in front, camera at eye level, background tidied or blurred. Check what is behind you before every external recording.
- Never make camera-on a team rule. A clear voice and a well-planned screen do most of the work.
Every recording tool asks the same question before you start: camera on or off? Most people answer by habit, either always on because that is what video means, or always off because they hate seeing themselves. Neither is a good rule. Whether to show your face in a screen recording depends on what the recording is for, and the answer changes from one video to the next.
A better habit is to ask one question before each recording, then, if the answer is "camera on", spend two minutes on the bubble, the light and what is behind you. The research on faces in video helps here, though not quite in the way most people assume.
Decide by what carries the message
Ask one question: is the important part of this recording on the screen, or in how you say it? If it is on the screen, the camera mostly takes up space. If it is in the tone (reassurance, enthusiasm, an apology, a hard piece of feedback delivered kindly), your face carries information that words alone do not.
| Situation | Camera | Why |
|---|---|---|
| First recording to a new client or prospect | On | They are deciding whether to trust you; a face makes the sender a person |
| Design or writing feedback that could sting | On | Expression softens criticism that reads harshly in text or voice alone |
| Bad news, a delay, an apology | On | Hiding makes it feel evasive |
| Welcome or onboarding introduction | On | The point is to meet someone |
| Weekly update to a team that rarely meets | Your choice | Nice for connection, not needed for content |
| Bug report | Off | The developer needs every pixel of the reproduction |
| Code or pull request walkthrough | Off, or very small | Dense text fills the screen edge to edge |
| Tutorial in a help centre | Off, or intro only | Viewers come back to find one step; a face adds nothing to step four |
There is also a middle option: camera on for the opening and the ask, off for the dense middle. You introduce yourself and the point, hide the bubble for the walkthrough, and bring it back for "so, can you approve this by Thursday?" If your recorder cannot hide the bubble mid-recording, recording the intro as its own short clip achieves much the same.
Why the bookends matter most
The opening and the closing are where a face does the most work. In the first seconds the viewer decides whether this video is worth their time and what attitude it comes with: a friendly heads-up, an urgent problem, a formal update. Your expression answers that faster than any sentence. At the end, the ask lands differently when someone is looking at you while making it. "Could you sign this off by Thursday?" from a face is a request between colleagues; from a voice over a spreadsheet it can sound like a ticket being assigned.
The middle, meanwhile, is usually where the screen is busiest and your face is least informative. You are looking down at the code or across at the design, half concentrating, and the bubble mostly shows the top of your head. Losing it there costs nothing and gives the content the whole frame.
What research on faces in video actually shows
Most of the research on faces in video comes from online courses, and two studies are worth knowing because they point in slightly different directions.
In a 2014 study of 6.9 million viewing sessions on edX, Guo, Kim and Rubin found that videos which cut between the instructor's face and the slides held viewers longer than slides alone. They also compared a course filmed with the instructor at his office desk against one filmed in a professional TV studio, and viewers stayed longer with the desk one. That is a single comparison between two different courses, so treat it as a hint, but it is encouraging for anyone recording at a kitchen table.
Engagement is not the same as understanding, though. A 2024 experiment by Schneider and Sung, published in the Journal of Learning Analytics, ran a semester-long online course with four versions of the videos: with or without the instructor's face, and with or without a visualisation of where the instructor was looking. The face on its own had no significant effect on learning. Showing the instructor's gaze did improve conceptual understanding.
A face makes people want to keep watching. Showing where you are looking is what helps them follow.
For screen recordings, that second finding is the practical one. Viewers need to know where to look. You can give them that without a camera: move the cursor deliberately to what you are talking about, use click highlights, and draw on the screen to circle the thing that matters. In VeoRec, drawings stay attached to the element you drew on as the page scrolls, so a circle around a button stays around that button.
Turn the camera on when trust and tone matter
The strongest case for the camera is any message where the same words could land well or badly depending on delivery.
Feedback. "The onboarding flow has too many steps and the copy on step three is confusing" is fine in a calm voice with a face that is clearly on the designer's side. As text, or as a disembodied voice over a screen, it can read as a verdict. If you give a lot of design or content feedback, our guide to giving design feedback covers the wording; the camera covers the tone.
First contact. A prospect or a new client has no picture of you yet. A short recording with your face, explaining a proposal or saying hello before a project starts, does a lot of the work a first meeting would. Our guide to video proposals goes into that use in depth.
Bad news. A delay, a bug that reached customers, a scope change. Delivering it on camera, looking at the lens, signals that you are not hiding behind the screen.
In all three, the face does not need to be big. A small, well-lit bubble with clear eye contact at the key moments does the job.
What the bookends sound like for a delay
Say a client launch is slipping by a week because a payment provider review took longer than planned. A version that works, camera on for the first and last fifteen seconds only:
- Opening, looking at the lens: "The launch is moving from the 14th to the 21st. That is on us for not starting the payment review earlier, and I want to show you exactly where things stand."
- Middle, camera off: the review status page, the two items still open, the revised plan. Screen and voice only.
- Close, camera back on: "If the 21st causes problems on your side, reply here today and I will look at what we can pull forward."
The face carries the ownership and the offer. The screen carries the evidence. A version recorded entirely with the camera off would contain the same facts and still read as more distant.
Leave it off when the screen needs every pixel
Any bubble covers something. On a slide with a lot of white space, that costs nothing. On a code editor with long lines, a spreadsheet, a dense dashboard or a bug that happens in a toast in the corner, it can cover the very thing the viewer needs.
The camera also adds cognitive load for you. You are thinking about the bug, the steps, the cursor and the words, and now also about your face, your posture and the laundry behind you. For a quick bug report or a code walkthrough, that is effort better spent on the content.
Remember where the video will be watched, too. You record on a large monitor, but the viewer may open the link on a laptop in a small player, or on a phone between meetings. Everything shrinks and the text gets harder to read, while the bubble still hides exactly the same part of the content. Viewers on a small screen have the least detail to spare, and they cannot drag your face out of the way. If the content is dense and the viewers are likely to be on small screens, that is a strong reason to leave the camera off.
And the camera is optional in a more basic sense. Some people are uncomfortable on camera, some are recording late at night, some are in a shared space. A clear voice, a well-prepared screen and a cursor that points at the right thing are a complete recording. A camera-on rule for async video is more likely to discourage recordings than to improve them.
Size and place the bubble where it covers nothing
When the camera is on, the bubble should be big enough to read your expression and small enough that nobody notices what it covers. For most walkthroughs, a bubble around a fifth of the frame's height is enough to read your expression. Make it bigger for an intro where you are the content, and use the smallest size your recorder offers for dense screens.
Placement matters more than size. Before you start, look at the screen you are about to record and ask which corner the content does not use. Try it on this mock product screen: drag the bubble around and switch between sizes to see what each position hides.
Some common traps:
- Websites and web apps: the bottom right is often taken by chat widgets, cookie banners and "back to top" buttons, and many apps show toast messages in a corner. The bottom left is often the safer default for a product demo.
- Code editors: many show a minimap or scrollbar along the right edge and a status bar along the bottom. A bubble in the top right can cover the minimap; the bottom left can cover the file tree when it is open. Check the layout you actually use.
- Slides: pick the corner with no text on most slides, and keep slides from putting key content there.
- Anything with a corner notification: if the thing you are demonstrating appears in a corner, move the bubble before you get there, not after.
Treat placement as part of your preparation, like closing tabs. If your recorder lets you drag the bubble during recording, moving it on camera is fine, but it is one more thing to remember while you are talking.
A placement mistake, and the fix
A common one: you record a checkout walkthrough for a developer with the bubble in the bottom right corner. Everything looks fine until the payment step, where the bug shows up as a red "Card declined" toast in the bottom right corner, under your face. The developer watches twice and replies "what error?"
The fix takes ten seconds before the next take. Trigger the error once without recording and note where it appears. Then put the bubble in the opposite corner, or shrink it, or leave the camera off for a bug report altogether. The general rule: find out where the important thing appears before you decide where your face goes.
Light from the front and put the camera at eye level
Most webcams are small sensors that struggle in dim rooms, and the result (grainy, smeared, orange) is what makes people dislike their camera image. Light fixes most of it.
- Face the light Sit so a window or lamp is in front of you, slightly to one side. A window behind you turns you into a silhouette, because the camera exposes for the bright background.
- Raise the camera to eye level A laptop on a desk looks up your nose and shows the ceiling. Put it on a stand or a few books until the lens is roughly level with your eyes.
- Frame head and shoulders In a circular bubble, your face should fill most of it with a little space above your head. Too far back and you become a small figure in a room.
- Look at the lens at the key moments You will look at the screen most of the time, and that is natural. For the opening line and the ask, look at the camera. That is where eye contact happens for the viewer.
Light also helps background effects. Blur and replacement work by separating you from what is behind you, and they do that best when you are clearly lit and the background is not the same colour as your clothes.
Blur or replace the background, and check what is behind you
You have three options for what appears behind you: the real room, a blurred version of it, or a replacement image.
The real room is the most natural and has no edge artefacts around your hair and hands. It needs a tidy, neutral view: a wall, a bookshelf, a plant. The edX finding about informal settings suggests a real home office is fine as long as it is not distracting.
Blur keeps the sense of a real place while hiding the details. It is the most forgiving option: if the software misjudges an edge for a moment, a blurred halo is far less noticeable than a cut-out one.
Replacement gives a clean, consistent look, which suits external recordings and branded videos. It is the most demanding: quick movements, hands near the face and busy patterns can produce visible edges. Keep still-ish, and choose a simple image.
VeoRec's camera bubble offers blur and replacement, both processed locally in your browser. More on how that works on the webcam background blur page.
Score your next recording
If you are still not sure for a particular video, tick what applies. The more boxes, the more the camera earns its place.
Make peace with seeing yourself
A lot of camera reluctance is simply that seeing yourself is uncomfortable. You notice the hair, the double chin, the way you blink when you think. Viewers do not; they are looking at the screen and listening to what you say.
Some things that help: keep the bubble small, so it is not a mirror in the corner of your eye. Do not watch it while you record. Check framing and light once before you start, then forget about it. And remember that a face in a work recording is not a performance. Nobody expects presenter polish in a two minute design review. A relaxed, normal face is exactly what makes a recording feel like a person talking to them.
If you never get comfortable, that is fine too. Use the camera for the recordings where tone really matters, and leave it off for the rest. A recording with good sound and a clear screen is better than one where you spent the whole time worrying about your face. If you want to work on the sound side, see our guide to sounding good on recordings.
Think about viewers who cannot hear well
Faces have one more use that is easy to forget: some viewers who are hard of hearing, or watching in a language that is not their first, follow speech better when they can see the speaker's mouth. A small bubble is not much help for that. If you know your audience includes people who rely on it, make the bubble large enough that your mouth is clearly visible, keep your face lit from the front, and do not cover your mouth with your hand while thinking.
Captions matter far more than the camera, though. They help everyone watching without sound, in a noisy office, at double speed or in a second language. VeoRec adds a transcript and captions to every recording automatically, and their accuracy depends mostly on how clearly you are heard, which is one more reason to get the microphone close. Our guide to screen recording accessibility covers captions, contrast and pace in more detail.
What to do before your next recording
Before the next one, ask what carries the message: the screen or the tone. If it is the screen, leave the camera off and spend the effort on preparing the screen and moving the cursor with intent. If it is the tone, turn the camera on, check the light and the background, drag the bubble to the corner your content does not use, and look at the lens when you say what you need.
And if you are setting norms for a team, write "camera optional" into them. People record more, and more honestly, when they are not also required to be on show. For the full setup of microphone, camera and screen together, see our guide to screen recording with audio and webcam.
Frequently asked questions
Should I show my face in a screen recording?
Show it when tone or trust matters: first contact with a client, feedback that could sting, bad news, introductions. Leave it off when the screen carries the message and every pixel counts, as in bug reports, code walkthroughs and step-by-step tutorials. You can also show your face only for the opening and the ask.
Does showing your face make a video more engaging?
Some evidence says yes for engagement: a large edX study found videos mixing the instructor's face with slides held viewers longer than slides alone. A 2024 experiment found the face alone did not improve learning, while showing where the instructor was looking did. Faces help people keep watching; pointing helps them understand.
Where should I put my webcam bubble in a screen recording?
In a corner your content does not use. On websites the bottom right is often taken by chat widgets and banners, so the bottom left is a common safe default. In code editors and dashboards, check which corner is empty in your own layout. Move the bubble before you reach anything that appears in that corner.
How big should the camera bubble be?
Big enough to read your expression, small enough not to cover content. Around a fifth of the frame height works for most walkthroughs; use the smallest size available for dense screens such as code or spreadsheets, and a larger one for an introduction where you are the focus.
Is background blur better than a virtual background?
Blur is more forgiving: when the software misjudges an edge, a soft halo is hard to notice. Replacement looks cleaner and more consistent but shows edge artefacts more easily with fast movement and busy patterns. Either works best with good front lighting and clothes that contrast with the background.
Should teams require camera on for async video?
No. Making the camera optional gets you more recordings, and most work recordings carry their message on the screen and in the voice. Encourage it for client-facing videos, feedback and introductions, and leave the rest to each person.