To find old tweets on the Twitter Wayback Machine route, paste a profile or tweet address (for example twitter.com/username or x.com/username/status/123) into the Wayback Machine at web.archive.org and open a capture from before the tweet changed or disappeared. That works only if the Internet Archive’s crawler saved that page while it was public. This guide shows the quickest ways to check, how to list every archived tweet URL for an account, how to save what you find, and why captures from recent years are often harder to read.
How the Twitter Wayback Machine search works
There is no separate “Twitter Wayback Machine”. It is the same Internet Archive Wayback Machine, used on Twitter (now X) addresses. The archive stores copies of public web pages that its crawlers, or people using Save Page Now, visited at a given moment. If a tweet page or profile page was captured, you can open that copy. If it was never captured, the archive has nothing to show, however famous the account.
Three kinds of address are worth checking, because the same content was published under several of them over the years:
twitter.com/usernameandx.com/username(the profile timeline)twitter.com/username/status/IDandx.com/username/status/ID(a single tweet)mobile.twitter.com/username(older mobile pages, often simpler to read in the archive)
Step by step: find an archived tweet
- Get the address. If you have a link to the tweet, use it. If not, start with the profile address.
- Search the Wayback Machine. Open web.archive.org, paste the address and press Enter. The calendar shows the dates it was captured.
- Pick a date before the change. If a tweet was deleted in March, open captures from February or earlier.
- Try the other domain. If
x.comshows nothing, try the same path ontwitter.com, and the reverse. - Try the mobile address for older periods if the desktop capture does not show tweet text.
- Look for replies and quotes. If the original tweet was never captured, a captured reply or quote tweet may still show its text.
List every archived tweet URL for an account
Clicking through calendars is slow if you are researching a whole account. The Wayback Machine’s CDX index can list every captured tweet address in one request. In a browser, open:
https://web.archive.org/cdx/search/cdx?url=twitter.com/username/status/*&fl=original,timestamp,statuscode&collapse=urlkeyReplace username with the account name, then run it again with x.com. Each line shows a tweet URL, the first capture time and the HTTP status. Lines with status 200 are the ones most likely to show content. To open a capture, combine them: https://web.archive.org/web/TIMESTAMP/URL.
For large accounts the list can be long, and the archive limits heavy traffic, so send requests one at a time. Our guide to the Wayback Machine CDX API explains the parameters, and command-line tools such as waybackurls can produce the same list from a terminal.
Why recent captures often show a login page or nothing
Twitter has changed how it delivers pages many times. Older tweet pages were mostly plain HTML, so the archive captured the text directly. Newer pages load tweets with JavaScript after the page opens, and in recent years the site has frequently asked visitors to sign in. A crawler that cannot run those scripts or sign in may save a shell page, a loading spinner or a sign-in prompt instead of the tweet.
In practice that means:
- Captures from around the early 2010s are often easy to read.
- Later captures may need several attempts on different dates before one shows content.
- Some captures exist but are empty. A capture in the calendar is not proof that the tweet text was saved.
Our guide on what to do when the Wayback Machine is not working covers the general replay problems, such as blocked scripts and broken pages, that also affect tweet captures.
Images and videos in archived tweets
Photos attached to tweets were served from Twitter’s media servers (addresses beginning pbs.twimg.com). If you can see an image address in the archived page source, paste it into the Wayback Machine on its own; the image may have been captured separately even if it does not display inside the tweet.
Video is much less likely to be archived. Video files are large and were delivered in pieces, so the crawler often saved only the page around the player. Our guide on downloading video from the Wayback Machine explains how to check whether a video file itself exists in the archive.
How to save archived tweets you find
Once you find a capture that shows the content you need, keep your own copy so you do not depend on finding it again:
- Record the archive address. The full
web.archive.org/web/TIMESTAMP/URLlink is the citation. It shows exactly which capture you used and when it was taken. - Save the page. Use your browser’s save function, or print to PDF, for a readable copy with the archive banner showing the capture date.
- Save the raw file. Adding
id_after the timestamp (/web/20140101000000id_/URL) returns the page as it was captured, without the archive’s toolbar. This is useful for keeping the original HTML. - Download images separately from their own archived addresses, as described above.
- For many tweets, combine a CDX list with a careful downloader, one request at a time. The Wayback Machine Downloader guide shows the basics; keep the request rate low.
Other social networks on the Wayback Machine
The same method works for other public profiles, with the same limits. Searches such as “wayback machine for Facebook” are common, but Facebook pages usually require login and load content with scripts, so archived copies often show a login screen. Public pages that did not require login, such as old company pages, forums and blogs, tend to archive much better than social feeds. Always check several dates. Two platforms have their own guides: the Instagram Wayback Machine method and the Reddit Wayback Machine method, which relies on old.reddit.com captures.
Your own tweets: use your X data export first
If you are trying to get back your own posts, the Wayback Machine is not the best first step. X lets account holders request a download of their account data from the account settings, and that archive is far more complete than anything a crawler saved. Use the Wayback Machine for public tweets from other accounts, or for posts from accounts you no longer control.
Limits to keep in mind
- Coverage is uneven. Accounts with many followers and links were crawled more often. Small accounts may have only a handful of captures, or none.
- Protected and suspended accounts were not public, so their tweets were generally not crawled while protected.
- Exclusions exist. Some content has been excluded from the Wayback Machine at the request of rights holders.
- Accuracy. A capture shows what the crawler received at that moment. For research or legal use, cite the archive URL and capture date rather than a screenshot alone.
When you need a whole website rather than tweets
If the tweets you are researching point to an old website that has gone, the same archive may hold that site too. Our free archive check shows how much of a domain was captured, and the Wayback website download service can deliver the captured site as a clean HTML ZIP.
Frequently asked questions
Can the Twitter Wayback Machine show deleted tweets?
Only if the tweet, its profile page or a reply showing it was captured before it was deleted. If no capture exists from that period, the archive cannot show it.
Do I search twitter.com or x.com in the Wayback Machine?
Both. Captures are stored under the exact address the crawler visited, so check twitter.com, x.com and mobile.twitter.com versions of the same path.
Why does an archived tweet show a sign-in page?
The site asked visitors to sign in or loaded tweets with scripts the crawler could not run, so the archive saved the prompt instead of the tweet. Try other capture dates or the mobile address.
Can I download archived tweet videos?
Sometimes, but rarely. Video files were often not captured. Check the video file address directly in the Wayback Machine.
How do I see all archived tweets from one account?
Use a CDX query with twitter.com/username/status/* (and the x.com version) to list every captured tweet address, then open the captures with status 200.