Compress Video

How Video Compression Actually Works

Video is enormous because it is really a rapid sequence of full images — 24 to 60 of them every second. Compression shrinks that firehose of data using two big ideas: throwing away detail the eye won't miss, and storing only what changes between frames instead of repeating the whole picture each time. Getting a smaller file without an ugly result is about applying those ideas intelligently, and knowing the levers helps you pick the right quality setting.

Bitrate is the main dial

The single biggest factor in a video's size is its bitrate— how many bits are spent per second. Lowering the bitrate is what actually makes the file smaller; the quality presets here (low, medium, high) are really choosing how aggressively to reduce it. Modern codecs like H.264are remarkably good at hiding this reduction, which is why a medium setting can cut 40–70% off a file while looking almost identical to the source.

Keyframes and interframe compression

Instead of saving every frame in full, video codecs save an occasional complete image (a keyframe) and then, for the frames in between, store only the differences — the pixels that moved. This is why a static talking-head clip compresses far better than fast, chaotic action footage: when little changes between frames, there is little to record. It also explains why the same settings produce very different file sizes on different videos.

Practical ways to get a smaller file

If one compression pass isn't enough, stack a few strategies. Trimdead air off the start and end — fewer seconds means fewer bits. Resize to a lower resolution if the video will only be watched on a phone. And choose a stronger quality level for footage headed to email or messaging, where a hard size cap matters more than pixel-perfect fidelity. Combined, these usually clear even strict upload limits.