Link Types and rel Attributes

Link types are distinguished by two features: where they lead from and to (internal and external) and how you tell Google about your relationship with the page you link to (the rel attribute). Below: which rel values Google’s documentation describes, what it says about them and what it does not — in particular, it never says that a link “passes” or “does not pass” weight. We also cover the mistakes that really do make paid links a problem.

rel values in the documentation

Values can be combined with a space or comma: rel="ugc nofollow". nofollow was previously recommended for paid links; it is still acceptable, though sponsored is preferred. If you want to recognize trustworthy contributors, you can remove ugc from links by members who have consistently posted high-quality contributions; protection from user-generated spam has its own help page.

What this does — and does not do

The documentation says: links marked with these values will generally not be followed, but the linked pages may still be found by other means — through sitemaps or links from other sites — and thus may still be crawled. It says not a word about “weight”. So the “passes / does not pass” column from popular tables is not a documented fact. Three more clarifications from the same page:

• rel values work only in <a> elements Google can crawl (see the lesson on link equity);
• nofollow also exists as a robots meta value; to stop Google fetching a link to a page on your own site, use the robots.txt disallow rule (the file in detail);
• to keep a page out of the index, allow crawling and use the noindex rule.

Types by location

Internal links

These lead to pages of the same site. Per the links help page, every page you care about should have a link from at least one other page, and links should have clear anchors. Internal links usually need no rel.

External links

These lead to other sites. Google writes that linking to other sites is nothing to be scared of, and citing your sources can help establish trustworthiness; what matters is to give readers context about what to expect.

Image links

If an image is the link, Google uses the alt attribute as the anchor: an empty alt on a linked image is a lost anchor. More on alt is in the lesson on image SEO, and on anchors in the lesson on anchor text.

When paid links become a problem

The spam policies are calm about paid links as such: buying and selling links is a normal part of the web’s economy and is not a violation as long as the links are qualified with nofollow or sponsored. A violation is links created primarily to manipulate rankings. Examples include:

• buying and selling links for ranking purposes, including exchange for goods and services;
• excessive link exchanges and partner pages for cross-linking;
• using automated programs to create links to your site;
• requiring a link in terms of service or a contract without letting the content owner qualify it;
• text ads, native advertising and guest posts with links that pass ranking credit or with optimized anchor text;
• links in low-quality directories, widgets and footers distributed across many sites;
• forum comments with optimized links and signatures.

What “punishment” follows, and whether it does, the documentation does not state; the general approach to spam is described in the lesson on spam policies, and suspicious links lead to the lesson on toxic links. Which link acquisition tactics are acceptable is covered in the lesson on tactics.

How to check your links

Use a crawler to find outbound links on pages with ads, partner content and comments and see which lack the right rel. The anchors and placement of internal links to a page are visible in the internal link context tool.

Checklist

1. Advertising and paid placements carry rel="sponsored" (nofollow is acceptable).
2. Links from comments and forums carry ugc.
3. rel values sit on real <a href> links.
4. Image links have a meaningful alt.
5. Block crawling through robots.txt and indexing through noindex, not through nofollow.
6. After publication at partners, it is checked that attributes were not lost.

Mistakes

• Believing an attribute “zeroes out weight” — the documentation is silent about weight.
• Assuming nofollow guarantees Google will not find a page.
• Using nofollow to hide a page of your own site from crawling.
• Buying articles with optimized anchors and no label.
• Leaving image links with an empty alt.

What next

Next come link acquisition tactics: which ways of earning links stay within the policies and which are a direct road to link spam.