To see old versions of a web page, go to web.archive.org, paste the address and open the calendar of snapshots; pick a blue date to view that capture. To save a page that is not archived yet, use Save Page Now on the same site. If saving fails or the page was never crawled, keep your own PDF copy.
The steps and limits come from the help center and blog of the Internet Archive, which runs the Wayback Machine, checked on October 13, 2026.
This guide is part of the Guide to saving articles to read later, which organizes every guide on the topic. For Keep Analog terms such as “eligible link” and “issue,” see the Keep Analog glossary.
How do I see old versions of a page?
According to the Internet Archive's start-here guide and the Wayback Machine usage help:
- Go to web.archive.org.
- Paste the full address of the page into the search box and browse its history.
- Open the calendar of snapshots and pick a date. You see the page as it was that day.
- Use the Changes view when you want to compare two archived versions side by side.
How to read what you see:
- Calendar dot colors: blue is a good capture (2xx), green a redirect (3xx), orange a client error (4xx) and red a server error (5xx). Most of the time you want the blue ones.
- Date inside the link: the number in the middle of the archived address is year, month, day, hour, minutes and seconds. In
web.archive.org/web/20000229123340/http://www.yahoo.com/, the capture is from February 29, 2000, at 12:33:40. - Every capture of an address: replace that date with an asterisk, as in
web.archive.org/*/followed by the address.
The usage help also notes a lag of about 3–10 hours between a crawl and when the capture appears in the Wayback Machine.
How do I save a page, step by step?
According to the Internet Archive's start-here guide:
- Go to web.archive.org.
- Find the Save Page Now box on the right side of the page.
- Paste the full address of the page you want to save and click Save.
You get a permanent link to the archived version. The Internet Archive blog also gives the direct address of the form, web.archive.org/save, where the button reads Save Page.
What does it save, and what not?
According to the help page on saving pages, Save Page Now saves the page you enter with its images and CSS, but not the links that lead out of it or the rest of the site. The Wayback Machine usage help adds that it saves the page one time: the address is not added to future crawls. For several pages, you save them one by one.
The form at web.archive.org/save also shows options such as Save outlinks and Save error pages. According to the same form, saving a screenshot, adding the page to your own web archive or getting the results by email requires signing in. The Internet Archive says it does not keep your IP address, so your submission is anonymous.
What if the page is not archived, or Save Page Now fails?
According to the usage help, a site may be missing because crawlers never found it, because it was password protected, blocked by robots.txt, or otherwise inaccessible, or because the site owner asked for exclusion. Broken or gray images usually mean those files were not archived; you can check an image URL the same way you check a page URL. The best overview of what was kept for a site is http://web.archive.org/*/www.yoursite.com/*.
The help page on saving pages warns that some sites prohibit crawling and a few have SSL settings that make Save Page Now break. Among the reasons a page is incomplete, the usage help lists robots.txt, password protection, JavaScript that must contact the original server, and owner exclusion requests; as a rule of thumb, simple HTML is the easiest to archive.
The Internet Archive has also published an FAQ on publishers blocking the Wayback Machine: some publishers block archiving out of concern that the material could be reused by AI companies. If a publisher does not allow archiving, or if Save Page Now fails, your own PDF copy is the fallback—see How to save a web page as a PDF on your phone and How to save a web page as a PDF on a computer.
How do I cite an archived page?
The usage help says the Wayback Machine is built to be referenced: you can copy the address of an archived page and use it. For MLA, it reports that MLA told the Internet Archive there is no established format and that it is best to err on the side of more information: cite the page as you normally would, then add the name Internet Archive and the archived address. For the rest of the citation, see How to cite a website article in APA and MLA.
Before you trust a copy: three checks
These checks are an original KeepAnalog suggestion, not steps from the Internet Archive:
- Open the copy right after saving it and make sure the text you care about is there, because a page can be archived incomplete.
- Write down the original and the archived link together, with the title, the author and the capture date.
- For anything you cannot lose, add your own PDF; the phone and computer PDF guides linked above show how.
What next?
To build a reading archive with context, continue with When links disappear: a sturdier reading archive. If you prefer paper, KeepAnalog turns eligible public links chosen by one person (HTML pages or PDFs) into a monthly printed issue. Nothing is printed or charged until that person approves the issue. For now, KeepAnalog's pilot is in Argentina only, with no delivery in the United States.