Archive Your Links · Method · DoingResearch.net

Purpose

When you cite a web page, you’re betting it will still be there, saying the same thing, when your reader clicks. That bet loses often.

In a study published in 2024, the Pew Research Center checked a sample of nearly a million web pages collected between 2013 and 2023. By October 2023, 38% of the pages that existed in 2013 could no longer be reached. Across the whole sample, a quarter were gone. Pew called its test the most conservative possible: a page counted as gone only if it returned one of nine error codes that clearly mean it no longer exists.
The same study found at least one broken link on 23% of news pages and 21% of government pages, and 11% of the reference links on Wikipedia pages no longer worked.
In 2014, Jonathan Zittrain, Kendra Albert and Lawrence Lessig found that more than 70% of the URLs in three Harvard law journals, and about half of the URLs in U.S. Supreme Court opinions, no longer led to the cited material.

Two things go wrong. Link rot: the page is gone. Reference rot: the link still works, but the page now says something different from what you cited, which is worse because nobody notices. An archived copy fixes both. It freezes what the page said on the day you used it.

When you need it

You cite any web page, especially news stories, government pages, blog posts or company pages.
You quote or take a number from a page that could be edited later.
You cite a social media post, which can be deleted or made private.
You’re finishing a paper, report or article and want the links to work in five years.
A page you need has already changed or vanished, and you need to see what it used to say.

How to do it

1Save the page with the Wayback Machine

Go to web.archive.org/save. Paste the page’s address into the “Save Page Now” box and click “Save Page.” When it finishes, you get an archive address that starts with web.archive.org/web/ followed by a date and time stamp and the original URL. Copy it. A checkbox lets you also save the pages it links to (“Save outlinks”). Extras such as saving a screenshot need a free Internet Archive account. Save Page Now saves only the one page you enter, not the whole site.

2Make a second copy elsewhere

For anything important, save it in a second archive too. Archive.today (archive.ph) saves a text copy and a screenshot of the page, with scripts and pop-ups removed, and gives you a short link. Perma.cc, built by the Harvard Library Innovation Lab and supported by libraries, is designed for scholars, journals and courts. It needs an account. Two archives run by different organizations are safer than one.

3Check that the copy actually worked

Open the archived link and look. Is the text you’re citing there? Did the numbers, the table or the chart load? Archives struggle with some pages. The Internet Archive’s own help pages say JavaScript that needs to contact the original server will fail when archived, and pages behind a password or a form aren’t collected. If the copy is blank or broken, try the other archive, or save a PDF or screenshot for your own records.

4Cite both links

In your citation, give the live URL and the archived URL, with the date you accessed the page. The live link lets readers see any updates; the archived one shows exactly what you saw.

5Archive as you go

Save pages when you first use them, not the night before you publish. By then some will already have changed. Make it part of how you take notes: the URL, the archive link and the date, together.

6Find what a dead page used to say

When a link you need is broken, paste the URL into the search box at web.archive.org. A calendar shows every date it was captured. Pick the capture closest to the date the page was cited. Search archive.today for the same URL too.

Try it

Your task

In 2011, in the U.S. Supreme Court case Brown v. Entertainment Merchants Association, about a California law on violent video games, Justice Samuel Alito wrote an opinion concurring in the judgment. Footnote 14 cites a web page at http://ssnat.com. Use the Wayback Machine to find what that page showed when the opinion came out (June 27, 2011), and what it showed about a year later. Don’t visit the live address; the domain has changed hands.

Show a hint

Type ssnat.com into the search box at web.archive.org and look at captures from 2011 and 2012.

Show the answer and the path

The opinion is at 564 U.S. 786 (2011), and footnote 14 starts on page 818. It cites a post titled “Inside the Sick Site of a School Shooter Mod” (March 26, 2011) at ssnat.com, as an example of a violent game whose developer said it might host the game on its own site.

A capture from June 29, 2011, two days after the decision, shows the cited page: the game developer’s own website, with that post dated March 26, 2011, on its front page.
By June 2012 the domain had a new owner. A capture from June 13, 2012 shows a one-page site with a message for the Court: “Aren’t you glad you didn’t cite to this webpage.” It goes on to say the original content was gone and someone else had bought the domain to make a point about how fleeting linked information is.
Later captures show other pages again, including error pages, and today the address redirects to an unrelated landing page.

What it shows: within a year, a URL in a Supreme Court opinion pointed to something completely different, and only an archive kept what the Justice actually cited. The ABA Journal reported the case in 2013 alongside the Zittrain, Albert and Lessig study.

Common mistakes

Archiving at the last minute. The page may already have changed since you read it. Save it when you first use it.
Not checking the archived copy. A capture can be blank, missing images, or show a cookie banner instead of the article.
Citing only the archive link. Readers lose the live page and any later corrections. Give both.
Trusting one archive. Archives can lose content or remove it at the site owner’s request. A second copy elsewhere is cheap insurance.
Assuming a working link means unchanged content. Reference rot is silent. Compare the live page with your archived copy before you publish.
Archiving a search result or a logged-in page. The archive saves what a stranger would see, which may be a login screen.

Limits

Archives miss things. Pages built by JavaScript that loads data from the original server, interactive charts and maps, paywalled articles, pages behind logins, and most social media feeds often come out broken or empty. For those, save a PDF or screenshot for yourself, and describe in your citation what you saw.

Site owners can ask the Internet Archive to exclude their sites, and a site’s robots.txt file can keep crawlers out. An archived copy you made can still become unavailable later.

An archive proves what a page said on a given date. It doesn’t make the page true, and it doesn’t make it citable if it wasn’t a good source to begin with.