Editing a screen recording so it can be read
Why they are hard to watch
A tutorial recorded from a screen has a specific set of failures, and all of them have fixes.
Everything is too small. A 27-inch display captured at 4K and delivered at 1080p means a 12-pixel menu label becomes 6 pixels. On a phone it is unreadable.
Nothing tells the viewer where to look. A screen has a hundred elements and the interesting one is a checkbox in the corner.
The cursor teleports. Mouse movement is fast and the viewer cannot follow it.
The pace is the presenter's, not the learner's. You know where the menu is; they are still reading the previous step.
Capture settings
Record at the delivery resolution where you can. Set your display to 1920×1080 before recording, or record a 1080p region of a larger screen. Capturing 4K and downscaling halves the size of every interface element.
On a high-density display, this means changing the scaling so the interface is physically larger and the capture is 1080p of enlarged interface, not a shrunken view of a large desktop.
Record at your timeline's frame rate, constant. OBS Studio is free, cross-platform, and records constant frame rate — unlike most built-in screen recorders, which write variable frame rate and produce the drift covered in the first block.
Turn off notifications before you start, and use a clean user account or profile with no personal bookmarks, no open tabs, no account names, no email addresses on screen. This is a privacy matter as much as a tidiness one.
Record system audio separately from your voice if the software lets you, so you can balance them later.
Punch in, relentlessly
The single most valuable editing technique for screen content: scale into the relevant part of the frame.
A 4K capture delivered at 1080p can be scaled to 200% with no loss, which is exactly the headroom needed. Even at 1080p, a punch-in to 150% is usually worth the softness because unreadable text at full size is worse than slightly soft text you can read.
Do it as an animated move — scale and position keyframes, eased, over about 12 to 20 frames — so the viewer follows the movement rather than being teleported. A cut straight to a zoomed view disorients; a quick push does not.
Then pull back out when the detail is done, so the viewer re-establishes where they are.
The cursor
Most screen recorders can highlight the cursor, add a click indicator, or smooth its motion. Turn these on; they cost nothing and they solve the teleport problem.
If your recorder does not, the editing equivalents: a soft circle or arrow animated to the click point, or a brief flash on the click. Keyboard input needs a visual too — a small on-screen key display is standard in software tutorials and several free tools provide one.
Pace and silence
Cut out every moment of nothing: loading, typing a long path, a menu you scrolled past. This is what the speed ramp is for — 400% over a ten-second file copy, with the audio continuing underneath.
And leave a pause after each step completes. Two seconds of the result on screen before moving on. This feels dead while editing and is where the viewer catches up.
The common failure of screen tutorials is the opposite of most video: they are usually too fast, not too long, because the presenter already knows the software.
Highlighting
When the viewer must look at one specific thing:
- A soft rectangular highlight — the rest of the screen darkened by 30–40%, the region of interest at full brightness. Animated on and off over about 8 frames.
- An arrow or circle, drawn simply, appearing briefly.
- Blur or darken the irrelevant parts, which also handles any personal information you failed to clear.
Do not use a red circle with a bouncing animation. The movement is more interesting than the thing it is pointing at.
Redaction
If personal data, licence keys or account names appear, blur or box them out. Use a solid box, tracked, rather than a blur — a blur applied to text can sometimes be partially reversed, and a mosaic on a short string is not secure. A solid rectangle is.
And check every frame, not just the obvious ones. Browser autocomplete suggestions, notification popovers, window titles and recent-file lists are where this leaks.
Today
Take a screen recording and punch in to 180% on the part that matters, with an eased 15-frame move in and out. Watch it on a phone. That is the difference between a tutorial that works and one that does not.
The one thing to keep
Capture at the delivery resolution so interface elements stay large, punch in with an eased move rather than a cut whenever detail matters, ramp through every moment of waiting, and redact with a tracked solid box because blur on text is not reliably irreversible.
Before you move on
A tutorial is recorded on a 4K display and delivered at 1080p. Viewers on phones say they cannot read the menus. The editor's instinct is to export at 4K instead. Why is punching in a better answer?
Pick the one you would defend. Nobody sees your answer.