How to record an async sales video

01Ninety seconds, and mean it

A sales video is competing with a subject line, not with other videos. Ninety seconds is the point past which completion rates fall off sharply, and a video nobody finishes is worse than the email it replaced, because it also asked for a click.

The discipline that gets you there is deciding what the video is for. One video, one ask. If it needs to establish a problem, demonstrate a fix and propose a meeting, it is three videos or it is a call.

• Say why you are recording for this person specifically, in the first ten seconds
• Show something on screen that they will recognise as theirs
• Make one ask at the end, and make it small
• Do not open with your company name; they can read that in the signature

02Put their world on screen, not your deck

The strongest opening in an async sales video is their own site, their own job posting, their own product, on screen while you talk. It is proof the video was made for them, which is the only thing that distinguishes it from a template.

The Area tab crops the capture to a region you draw, so you can frame one section of a page rather than a browser window with your bookmarks bar and fifteen tabs across the top.

03A face in the corner, not a face filling the frame

The webcam sits in one of eight positions on a 3x3 grid, sized between 5 and 40 percent of frame height, with a corner radius that runs from square to a full circle. A 1.0x to 3.0x digital zoom crops into the camera feed itself, which handles a badly framed laptop camera without moving the laptop.

Small and cornered works better here than in any other kind of screen recording. The point of the face is to prove a person made this; the point of the screen is the argument.

04Sound like you are in a room, not a corridor

Audio is what makes a prospecting video feel cheap. Six voice processing controls handle it after the fact: cleanup, adaptive noise suppression, bass boost, voice clarity, voice boost and a de-esser at around 7 kHz, applied at export rather than baked into the recording. These are Pro tier features.

Microphone and system audio are separate tracks with independent volume, so a notification chime during the take does not cost you the whole recording.

05Cut the thinking time

Automatic silence trimming removes the silent stretches, which on a first take is where most of the ninety seconds went. The Pro tier includes it, and on this kind of video it is often the difference between sending the first take and recording a fourth.

Recording is controlled from the tray and paused segments are stitched back together, so you can pause while you find the page you meant to show rather than starting again.

06Assume it gets watched on mute

A sales video is opened on a phone, often between other things, frequently without sound. Word level captions with progressive highlight keep it followable in that state, and they are transcribed on your own machine rather than through an API. Transcription is a Pro tier feature.

The practical effect is that your first sentence has to work as text. Write it as though it will be read rather than heard, because half the time it will be.

07Export for where you are sending it

Six aspect ratios cover YouTube, TikTok and Instagram, X, LinkedIn and Facebook, plus the source dimensions. A LinkedIn post and an email embed are not the same shape, and a 16:9 recording letterboxed into a vertical feed loses half its height to black bars.

For email, 1080p keeps the file small enough to host anywhere. The Pro tier goes to 4K at up to 120 fps with GPU accelerated rendering when the video is going somewhere that deserves it.

08The research on your screen stays on your machine

Recording, editing, transcription and export all happen locally. The only network calls the app makes are downloading it and activating a licence.

That is worth knowing when the video opens on a prospect's account page, a CRM record or a pricing sheet. Tools that meter their AI features are metering them because that processing runs on someone else's infrastructure, which means your screen and your audio went there.