Clicking is the usual way to check a link: if the page opens, the link passes. But Editors Canada’s standards ask for more than a click. The copy editing standards tell editors to correct or query information “that should be checked for accuracy (e.g., quotations, URLs).” The proofreading standards go further: “For web content, ensure that all links lead to the intended source.” The key word there is intended.
When I audited every external link on one business-to-business software blog, I went through 2,234 links in 580 posts, pointing to 796 domains. Most of the links were accurate and led to the intended source, but some needed updating and correcting.
Auditing one blog’s links
My work is off-page SEO, so I spend most days looking at links between websites. In this audit, I sorted every linked domain as “clean,” “review” or “toxic.” Of the 796 domains, only a few were problematic: two domains looked toxic and 26 needed manual review. This is an example from one company’s blog, so it’s not representative of every site.
What surprised me was that a click alone wouldn’t have caught most of the problem links. A few tutorials used made-up example addresses, and someone had turned them into live links! Those domains exist, but other people own them. That certainly doesn’t pass the standard of leading to the intended source. In other instances, the anchor text told the reader nothing, which made it hard to gauge accuracy. Still others pointed to sites that sell links, which made them neither accurate nor leading to the intended sources.
Why clicking isn’t always enough
By itself, a click tells you only whether the page loads today. This has some use, because pages do vanish. Pew Research Center looked at this phenomenon in 2024, and, of the web pages that existed in 2013, 38 per cent couldn’t be reached a decade later.
However, a click doesn’t tell you whether the page still says what the anchor text claims it does, whether the page is still owned by the same people or whether the page has turned into sponsored content. A webpage can still open but say something different now. In a 2014 paper, Lawrence Lessig, Jonathan Zittrain and Kendra Albert called this “reference rot,” and they found it in 50 per cent of the URLs in U.S. Supreme Court opinions. Sometimes the link’s ownership has changed and clicking now pays someone. Google’s Search Central guide recommends “sponsored” tags for paid links, but often the URL is the only hint you get.
You don’t need SEO tools to check links beyond the click: a careful read of each page and its URL will uncover most problems, as long as you’re thinking critically.
Links checklist for copy editing
To help you get started, here is a list of six questions that I ask about each link:
- Does the link go to the original source, not a recap?
- Is the number or quotation from the sentence actually on the linked page?
- Does the link redirect, or does the URL carry tracking or affiliate codes?
- Is the linked page’s domain still owned by the same people, and is it still a real site, not spam?
- Can the reader accurately predict from the link’s anchor text where they’ll land?
- If the source matters a lot, is there an archived copy somewhere, such as the Internet Archive’s Wayback Machine?
If a link fails one or more of these tests, I’d query it the same way you’d query any other claim lacking adequate support. To me, a link is just a citation the reader can click, so it gets the same care as the rest of the content.
—
The Editors’ Weekly is the official blog of Editors Canada. Contact us.
Discover more from The Editors' Weekly
Subscribe to get the latest posts sent to your email.