Why can moving from one codec to another never be lossless?

Because codecs such as H.264 and VP9 perform lossy compression; to shrink the data they discard the detail the human eye is least likely to notice. The picture in your MP4 has already been through that filtering once. When you re-encode with VP9 the picture is decoded and filtered a second time according to different rules. Because discarded detail never comes back, the total loss accumulates, and there is no technical way to reset it.

Why does encoding take so long?

To achieve better compression, newer-generation codecs search for the most efficient representation by trying far more possibilities. Splitting the picture into blocks of different sizes, tracking motion in finer detail and using complex prediction methods each add computational load. That is why VP9 is slower than H.264, and AV1 is markedly slower still than VP9. The price of better compression is paid directly in processing time.

If I convert my video to WEBM, can I delete the MP4?

We recommend you do not. The WEBM version is a copy produced from your source that has lost some quality; if you throw away the MP4 you no longer have the higher-quality version. And because WEBM does not open in many places outside the browser, you will need the MP4 later if you want to upload the video to social media or watch it on a phone. Use WEBM for web publishing and keep the MP4 for archival.

Do I really need WEBM for my site?

For most small sites it is certainly not essential. MP4 plays without trouble in every modern browser; serving MP4 only is a perfectly valid and widespread approach. WEBM's gain is bandwidth and load time, and that gain only becomes meaningful as scale grows. If you have a handful of videos on your site, it may not be worth the extra encoding and hosting effort.

What Does MP4 to WEBM Conversion Mean for Web Video?

Let us start with a question: why do sites serve two files?

When you look at a web page's source code you sometimes see two different versions of the same video defined — one .webm, the other .mp4. At first glance it looks like pointless duplication. Why is the same content prepared twice and hosted twice?

The answer describes the fundamental tension in web video: efficiency and compatibility do not meet in the same file. MP4 opens everywhere but carries more data for the same quality. WEBM can do the same job with less data but is not much use outside the browser. Serving two files is the way to resolve that tension.

MP4 to WEBM conversion exists for exactly this reason. In this article we explain what that conversion does behind the scenes, and what it gains and costs you.

Not a container swap, but a genuine reproduction

Format conversions come in two kinds, and the difference between them is enormous.

Some conversions only change the container. The video and audio data is taken out of the old box and placed into a new box as it is. Going from MOV to MP4 is usually like this. Because the picture is never touched, there is no quality loss and the process is fast.

Going from MP4 to WEBM does not belong to that group. The reason is this: an MP4 usually contains H.264 video and AAC audio. The WEBM container does not carry those; the codecs it carries are VP8, VP9, AV1 and Opus, Vorbis. In other words, nothing the target container accepts is present in the source file.

In that situation the only path is to decode the picture and compress it from scratch. The process works like this:

The H.264 data in the MP4 is decoded and raw frames are obtained.
Those frames are recompressed with VP9 (or AV1, depending on your choice).
The AAC audio is decoded and re-encoded as Opus.
The result is written into a WEBM container.

Why do you lose some quality?

This is a subject that needs to be discussed honestly.

H.264 and VP9 are lossy codecs. They compress by discarding the detail the human eye is least likely to notice. The picture in your MP4 has already been through that filtering once — the discarded detail is not in the file.

When you re-encode with VP9, the picture is unpacked and filtered a second time, this time according to VP9's own rules. Because two different codecs consider different things unimportant, some detail preserved by the first can be discarded by the second. The result: loss accumulates and cannot be undone.

How visible is this in practice? With a good quality setting, most viewers will not notice the difference. The loss shows up most in fast-moving scenes, on sharp edges and in smooth colour gradients. But if your source file is already heavily compressed, those artefacts can become more visible — because you are compressing an already weak picture for a second time.

The practical lesson: do the conversion from the highest-quality source you have, and do not delete the original MP4.

The gain side: why is it worth it?

So why do people go through this conversion despite all that loss?

Less data, the same picture. VP9 comes from a newer generation of compression than H.264. It can split the picture into more flexible blocks, track motion more finely and use more complex prediction methods. As a result it can carry the same visual quality with less data.

That translates directly into load time. On a web page, video is the heaviest component. When the file gets smaller the video starts sooner, a user on a mobile connection waits less, and there is bandwidth saving on the hosting side.

An open format. WEBM is an open standard with no patent-licensing headaches. If you are building your own infrastructure, that can be a practical advantage.

Transparency support. WEBM can carry an alpha channel. If you are going to use video with a transparent background on a page — for example overlaying an animation on top of page content — MP4 cannot generally do this, while WEBM can. In that scenario the conversion is a necessity.

The second part of the price: time

Alongside quality loss, the second price paid is processing time, and this is usually underestimated.

VP9 encoding is markedly slower than H.264. AV1 is slower still than VP9. The reason makes sense: to achieve better compression, the codec searches for the most efficient representation by trying far more possibilities. Every possibility tried means computation.

In practice that means converting a long, high-resolution video can take minutes. This is not a slowness problem, it is the nature of the algorithm. Our MP4 to WEBM converter does this work on your behalf but it cannot shorten the time — nobody can.

Do you need this conversion?

The short answer: only if you are publishing on the web.

If your video is going to be watched on your own site, in a browser, WEBM offers a real gain. The bandwidth difference becomes meaningful especially if you have a large number of videos or high traffic.

If your video is going to be uploaded to social media, appear in a phone gallery, play on a television or go into an editing program, do not do this conversion. WEBM most likely will not open there, and you will be left with a file you cannot use.

For small sites the honest answer is this: serving MP4 only is a perfectly valid approach. Every modern browser plays MP4 without trouble. WEBM is an additional optimisation, not a necessity.

In summary

MP4 to WEBM conversion is a deliberate price paid for efficiency in web video. Behind the scenes the picture really is decoded and recompressed; that is why the process is slow and why it brings a small but real quality loss. In return you get a smaller file and a faster-loading page. That trade usually makes sense for sites publishing on the web, and is unnecessary in every other scenario. The question to ask when deciding is simple: will this video leave the browser? If it will, stay with MP4.