Wayback Machine YouTube: What Is Saved and How to Find It
The Wayback Machine has been photographing the web since 1996, and YouTube pages are part of what it caught. When a video disappears, that snapshot is often the only thing left of it. This page explains what the archive actually holds for a YouTube URL, what it never held, and how to search it when you no longer have the link.
What a snapshot of a YouTube page contains
A capture is the HTML the crawler received at one moment in time, not the video. From a saved watch page you can usually recover:
- The title, as it read on the day of the capture.
- The description, including the links the author put in it.
- The channel name and id, which is what lets you walk the rest of its uploads.
- The upload date, plus the view and like counts frozen at capture time.
- The thumbnail, saved separately as an image file.
- Captions, when the crawler happened to fetch the subtitle track as well.
That is metadata rather than video, and it is more useful than it sounds: it identifies the video, proves it existed, says what it was about, and with captions it often preserves what was actually said.
Why the player on an archived page is blank
YouTube never served the video from the page you are looking at. The page carries a player, and the player then asks Google's media servers for the stream over an address that is signed and expires within hours. A crawler saving the page does not follow that stream, and even if it had, the address would be dead long before anyone replayed the snapshot. So an archived watch page looks correct and plays nothing.
The media, when it exists at all, sits on a separate path: the archive sometimes holds the raw video response for a given video id, and the Wayback Machine can play that back. This is the rare case, not the normal one. In the results, videos with a saved file carry a Has video mark. See watching and downloading archived files.
"We have not been able to archive the video associated with this YouTube page"
Enough people paste that sentence into a search engine that it deserves a plain answer. It means the Wayback Machine saved the page and did not save the video behind it. The usual reasons are that the video was already gone when the capture ran, that the media fetch was refused or timed out, or that the stream was simply out of reach for the crawler.
It is a report about one attempt, not a verdict on the video. What to try next:
- Look for an older capture. The failed attempt says nothing about snapshots taken years earlier, and those may predate the deletion and carry the full page.
- Check the media path directly. Browsing a snapshot does not ask the archive whether it holds the video file. Paste the link or the video id into the search box and the pipeline will ask for you.
- Look for a copy somewhere else. Deleted videos are often reuploaded; see finding a reupload.
Searching the archive when you still have the link
Paste the full YouTube URL, the short youtu.be link, or the bare 11 character video id. You get whatever the archives kept: title, description, thumbnail, subtitles, and whether a video file exists. Step by step: find a deleted video by its URL or id.
Searching when you do not have the link
This is where the Wayback Machine on its own runs out. Its interface answers one question, "do you have a capture of this exact URL", so you have to know the URL before you can ask. There is no way to ask it which videos from a channel it saved, and no way to search inside what it saved.
That gap is what this site fills. Over 1.5 billion video ids from the Wayback Machine and Common Crawl, indexed so they can be searched three other ways:
- By channel. A channel URL, an @handle or a channel id returns every video we know of from it, deleted ones included, which is the fastest route when the video is gone but the channel is not. See searching by channel.
- By words. Full text search across archived titles and descriptions, for when you remember what the video was called but nothing else. See searching by title and description.
- By what was said. Archived subtitles are indexed too, so a remembered phrase from inside a video can find it.
Does Save Page Now archive a YouTube video?
The Save Page Now button on web.archive.org captures the page. For a video that is still live it sometimes captures the stream as well, with no guarantee and no clear signal about whether it worked. Treat it as a chance rather than a plan: if a video matters to you, save your own copy while it is still up. The archive is a fire escape, not a backup service.
What the archive will never give you
- Videos nobody crawled. If the page was never captured before the deletion, there is nothing to recover. Coverage follows attention: popular channels and linked videos are captured often, a video with forty views may never have been visited by a crawler at all.
- Private and unlisted videos that were never publicly linked. A crawler cannot save a page it cannot reach.
- Most comments. For most of YouTube's life comments loaded after the page did, so they are usually not in the HTML that was saved.
- Anything the archive removed at a rights holder's request.
