Internal links and anchor text

1Overview

Google discovers most new pages by following links from pages it already knows, and links are among the ways it judges what a page is about and how important it is [1][2].

Internal links are the part of this a site owner fully controls. They determine which pages crawlers can reach, how many paths lead to each page, and the words used to describe it.

2History

In 1998 Google’s founders described PageRank, a method of ranking pages using the link structure of the web, in which a link from one page to another counts as a kind of vote [3]. Google still lists link analysis and PageRank among its ranking systems, alongside many others [4].

In January 2005 Google introduced the rel="nofollow" attribute so that sites could mark links, particularly in blog comments, that they didn’t vouch for [6]. In September 2019 it added rel="sponsored" and rel="ugc", and announced that all three would be treated as hints; from 1 March 2020 that applied to crawling and indexing as well as ranking [7].

3Anatomy of a link

4Crawlable links

Google can reliably follow a link only if it is an <a> element with an href attribute containing a resolvable URL, absolute or relative [1].

For single-page applications, Google recommends real URLs with the History API rather than fragment identifiers to load different content [15].

5Anchor text

Google recommends anchor text that is descriptive, reasonably concise and relevant to the page it points to and the page it is on. It should make sense out of context, and avoid generic phrases such as “click here” or “read more” [1].

Google also warns against stuffing anchors with keywords or making every link on a page use the same text; anchors should read naturally [1].

For image links, the image’s alt attribute acts as the anchor text [1].

6Internal linking and site structure

Google’s starter guide recommends linking to relevant pages within the site so readers, and crawlers, can find related content [16].

A page with no internal links pointing to it is often called an orphan page. Google can still find it through a sitemap, but sitemaps are recommended especially when pages are not well linked, which underlines that links are the primary path [10].

Breadcrumb navigation gives every page a link back up its hierarchy, and can be marked up with breadcrumb structured data so Google shows the trail in results [11].

7Broken links

A broken link points to a URL that returns an error, usually 404 Not Found. Google has said that 404s for pages that genuinely don’t exist don’t hurt the rest of a site’s rankings [12].

The cost is on the linking page: the visitor reaches a dead end and the link passes nothing. Pages that return 404 or 410 are dropped from the index over time, and pages that return 200 for missing content are treated as soft 404s [13].

Links that point to a redirect still work, Googlebot follows up to 10 hops, but each hop is an extra request, so internal links should point at the final URL [13][14].

8Link attributes

Values can be combined, for example rel="ugc nofollow"[8]. Using nofollow on internal links to “sculpt” authority is not something Google recommends; the attribute exists to describe a relationship [7].

9Link spam

Google’s spam policies prohibit link schemes intended to manipulate rankings, including buying or selling links that pass ranking credit, excessive link exchanges, and automated link creation [9].

10Common misconceptions