Video compression shrinks file size by removing data your eye cannot see

Compression works by having your video software analyze each frame and discard information that makes almost no difference to how the video looks. A video file is enormous because it stores color and brightness data for every single pixel in every single frame. Compression software finds patterns — areas where neighboring pixels are nearly identical, or frames where almost nothing changes between one moment and the next — and stores those patterns instead of storing every pixel separately.

The trade-off is always the same: smaller file size versus visible quality loss. You choose where on that scale you want to land. A video compressed to 10% of its original size will look noticeably softer or blockier than the original. A video compressed to 70% of its original size may look nearly identical to someone watching on a phone or laptop screen, even though a pixel-by-pixel comparison would show the difference.

Key Takeaways

  • The two main compression methods are changing the video codec (the format itself) and lowering the bitrate (the amount of data stored per second), and most compression tools let you do both.
  • H.264 is the most widely compatible codec and produces good quality at smaller file sizes than older formats, while H.265 compresses further but may not play on older devices.
  • Bitrate is measured in megabits per second (Mbps), and dropping from 25 Mbps to 10 Mbps typically cuts file size in half with visible but acceptable quality loss on screens smaller than 27 inches.
  • Free tools like HandBrake and FFmpeg can compress video on Windows, Mac, and Linux, while paid software like Adobe Media Encoder and DaVinci Resolve offer more control and faster processing.
  • Always compress a copy of your original file, never the original itself, so you can re-compress with different settings if the first result is not what you wanted.

Understanding codecs and bitrate — the two levers of compression

A codec is the format your video is stored in. Common codecs include H.264 (also called AVC), H.265 (also called HEVC), and VP9. Each codec is a different mathematical approach to storing video data. H.264 has been the standard for over a decade and plays on nearly every device. H.265 is newer and compresses to about 50% smaller file sizes than H.264 at the same visual quality, but older phones, tablets, and streaming devices may not play it.

Bitrate is how much data the video stores per second, measured in megabits per second (Mbps). A 1080p video shot on a phone might be 25 Mbps. The same video compressed to 10 Mbps will be noticeably softer and may show blockiness during fast motion, but it will still be watchable on a laptop or phone screen. Dropping to 5 Mbps produces visible quality loss that most people notice. The relationship is roughly linear: cutting bitrate in half cuts file size in half.

Most compression tools let you change the codec, the bitrate, or both. Changing codec alone (H.264 to H.265) typically shrinks the file by 40 to 50% with no visible quality loss. Lowering bitrate shrinks it further but introduces visible softness or blockiness. The best approach is usually to switch to H.265 first, then lower bitrate only if you need the file smaller still.

How to compress video in HandBrake (free, all platforms)

HandBrake is free software that runs on Windows, Mac, and Linux. Open HandBrake, click the folder icon in the top left, and select your video file. The software will scan it and show you the current resolution, codec, and bitrate.

In the right panel under "Summary", click the dropdown next to "Format" and select "MP4" or "MKV" (both are containers that hold video data; MP4 is more compatible). Under "Video" on the left, click the "Video" tab. Change the codec dropdown from "H.264" to "H.265" if your devices support it. Leave "Constant Quality" selected unless you have a specific bitrate target. A quality value of 20 to 22 produces visually lossless video; 25 to 28 produces quality loss you may notice on a large screen but not on a phone; 30 and above produces obvious softness.

Click "Start Encode" at the top right. Compression takes time — a 10-minute video typically takes 5 to 30 minutes depending on your computer's speed and the quality setting. When finished, HandBrake will save the compressed file in the same folder as the original.

How to compress video in Adobe Media Encoder (paid, Windows and Mac)

If you already subscribe to Adobe Creative Cloud, Media Encoder is included. Open the software, drag your video into the queue on the left, and a row will appear showing the file name and current settings.

Click the blue text next to "Preset" to choose a compression profile. "YouTube 1080p HD" and "Vimeo HD" are reasonable starting points. Click the blue text next to "Output File" to choose where to save the compressed version. Then click the green play button to start encoding.

Media Encoder encodes faster than HandBrake on most computers because it uses your graphics card to speed up the process. The trade-off is cost — you must pay for a Creative Cloud subscription. If you are already editing in Premiere Pro, Media Encoder integrates directly with your project timeline, so you can compress without exporting first.

How to compress video using FFmpeg (free, command line, all platforms)

FFmpeg is a command-line tool that is more powerful than HandBrake but requires typing commands instead of clicking buttons. If you are comfortable opening a terminal or command prompt, FFmpeg gives you the most control and often the fastest encoding.

After installing FFmpeg, open a terminal or command prompt and navigate to the folder containing your video. Type this command to compress to H.265 at quality 28:

ffmpeg -i input.mp4 -c:v libx265 -crf 28 -c:a aac -b:a 128k output.mp4

Replace "input.mp4" with your actual file name and "output.mp4" with what you want to call the compressed version. The "-crf 28" sets quality (lower numbers are higher quality; 28 is a good balance). The "-b:a 128k" sets audio bitrate to 128 kilobits per second, which is fine for most uses. Press Enter and FFmpeg will begin encoding.

Choosing quality settings based on where the video will be watched

If the video will be watched on a phone or tablet, you can compress more aggressively. A bitrate of 5 to 8 Mbps at 1080p resolution looks good on a 6-inch screen. If it will be watched on a 27-inch monitor or larger, aim for 15 to 20 Mbps to avoid visible softness. If you are uploading to YouTube, Vimeo, or another streaming platform, check their recommended bitrate — YouTube recommends 8 to 16 Mbps for 1080p uploads, and the platform will re-compress your video anyway, so compressing too much before upload wastes your effort.

For archival — keeping a copy for yourself — use the highest quality setting your tool offers. You can always re-compress later if you need a smaller file for sharing. Compressing twice degrades quality more than compressing once to the final target, so get it right the first time if possible.

Common mistakes that waste time or ruin quality

The biggest mistake is compressing the original file instead of making a copy first. If you compress too aggressively and the result is unusable, you cannot undo it — you have lost the original data. Always compress a copy, and keep the original untouched.

The second mistake is lowering resolution (changing 1080p to 720p) when you only need to lower bitrate. Lowering resolution makes the video smaller but also makes it blurrier. Lowering bitrate at the same resolution keeps sharpness and only softens fine detail. For most uses, lower bitrate at the same resolution looks better than lower resolution at higher bitrate.

The third mistake is choosing a codec your target devices cannot play. If you compress to H.265 and try to play it on a 2015 laptop or older Android phone, it will not play at all. Stick with H.264 unless you know your audience has devices from the last five years.

Frequently Asked Questions

How much smaller will my video be after compression?

That depends on the original bitrate and your target quality. Switching from H.264 to H.265 at the same quality shrinks the file by 40 to 50%. Lowering bitrate from 25 Mbps to 10 Mbps cuts the file in half. If you do both, you might end up with a file that is 25% of the original size. The only way to know is to compress a short test clip first and check the file size.

Will compressed video look bad on a large TV?

Yes, if you compress too much. A bitrate below 10 Mbps at 1080p will show visible softness and blockiness on a 55-inch TV. For TV playback, aim for 15 to 25 Mbps at 1080p, or use H.265 at quality 22 to 24. If the video will only be watched on phones and laptops, you can compress more aggressively.

Can I compress a video that is already compressed?

Yes, but quality degrades each time. If you compress a video to 50% of its original size and then compress that result again, the final video will look worse than if you had compressed the original once to the same final size. Always compress from the highest-quality version you have.

What is the difference between lossy and lossless compression?

Lossy compression removes data permanently — you cannot get it back. All video compression is lossy because video files are too large to store without removing data. Lossless compression removes only redundant data and can be reversed, but video files stay huge. For practical purposes, use lossy compression and accept that some quality will be lost.

How long does compression take?

A 10-minute video typically takes 5 to 30 minutes to compress, depending on your computer speed, the codec you choose, and the quality setting. H.265 takes longer than H.264. Lower quality settings encode faster. Using a graphics card (available in Adobe Media Encoder and some other tools) speeds up encoding by 2 to 4 times compared to using only the CPU.