How to compress a video without losing (visible) quality

Why phone and camera videos are so large

Cameras record at high bitrates – often 15–50 megabits per second – so the footage survives editing and color grading. They also record at high resolutions (4K) and frame rates (60 fps) by default. That is sensible for production, but wasteful for watching on a phone, in a chat or on a web page.

The size of a video is essentially bitrate × duration. A 60-second clip at 20 Mbps is about 150 MB; the same clip at 4 Mbps is about 30 MB. Almost every compression technique is a way to lower the bitrate needed for a given quality.

1. Lower the bitrate sensibly

The bitrate a video needs depends on its resolution, frame rate and content. As a rough guide for H.264, clean results need about 4–8 Mbps for 1080p at 30 fps, 2–4 Mbps for 720p, and 1–2 Mbps for 480p. Fast motion, confetti and grain need more; talking heads and screen recordings need less.

Never raise the bitrate above the original – it only makes the file bigger without improving anything. Good tools cap the new bitrate relative to the source.

2. Reduce the resolution

Resolution is the biggest lever. 4K has four times the pixels of 1080p, and 1080p has 2.25 times the pixels of 720p. On a phone screen, a 1080p or even 720p video is indistinguishable from 4K, so downscaling alone can cut the size by 50–75% with no visible loss.

For vertical video, think in terms of the shorter side: “720p” means 720×1280 for a vertical clip.

3. Lower the frame rate when motion allows

60 fps looks smoother for sports and gaming, but for people talking, tutorials or scenery, 30 fps (or even 24 fps) looks natural and needs noticeably less data. Only lower the frame rate – never raise it.

4. Choose the right codec

The codec is the compression method used inside the video file:

• H.264 (AVC): plays on everything. The safe default for sharing.
• H.265 (HEVC): about 35% smaller than H.264 at similar quality; plays on modern Apple devices and many others, but not everywhere.
• VP9: similar efficiency to HEVC, excellent in browsers, used in WebM files.
• AV1: the most efficient royalty-free codec, often 40–50% smaller than H.264, but slower to encode and not supported by older devices.

5. Trim and drop what you don’t need

The simplest compression is sending less video. Trimming the first and last seconds, removing a silent audio track or cutting a long recording into the relevant part often saves more than any encoder setting.

Audio is small compared with video, but at tight size limits it matters: 128 kbps is plenty for most content and 64–96 kbps works for speech.

Hitting a specific file size

When a platform has a hard limit – such as an email attachment limit or a chat app’s upload cap – work backwards: divide the size by the duration to get the total bitrate, reserve some for audio and container overhead, and use the rest for video. If the resulting bitrate is very low, lower the resolution too; a clean 480p video looks better than a blocky 1080p one.

Asset Compressor’s “By file size” mode does this calculation for you, picks a sensible resolution, and re-encodes once more if the result overshoots.

Checklist

For most videos that need to be smaller:

• Keep the original file until you have checked the result.
• Downscale 4K to 1080p (or 720p for chat apps).
• Use H.264 in MP4 for maximum compatibility, HEVC or AV1 if you know the viewer supports it.
• Lower 60 fps to 30 fps unless motion is fast.
• Trim unnecessary footage and remove audio you don’t need.
• Compare the result with the original on the screen it will be watched on.