Dofollow, nofollow, sponsored and ugc: what a directory link is worth

Google’s documentation on qualifying outbound links defines them (Qualify your outbound links):

• rel="sponsored" marks advertisements and paid placements. Google prefers it for paid links, and says nofollow is still an acceptable way to flag them.
• rel="ugc" marks user-generated content, such as comments and forum posts. A site can drop it for members it trusts.
• rel="nofollow" is for when the other values don’t apply and the site would rather Google not associate it with the linked page, or crawl that page from it.

Values can be combined in a space- or comma-separated list, so rel="ugc nofollow" is valid. Google says links marked with any of them “will generally not be followed”, and adds that the linked page may still be found through other means, such as a sitemap or links from other sites.

A whole page can also be set to nofollow. A robots meta tag or X-Robots-Tag header containing nofollow means “do not follow the links on this page” (source). It applies to every link there, whatever each link’s own rel says.

Neither dofollow nor follow is among the rel values that MDN’s reference documents, and Google’s page lists only the three above. A followed link is simply a link without them:

<!-- followed: no qualifying value -->
<a href="https://yoursite.com">Your Product</a>
<a href="https://yoursite.com" rel="noopener">Your Product</a>

<!-- qualified -->
<a href="https://yoursite.com" rel="nofollow">Your Product</a>
<a href="https://yoursite.com" rel="sponsored noopener">Your Product</a>
<a href="https://yoursite.com" rel="ugc nofollow">Your Product</a>

Other tokens show up in the same attribute and have nothing to do with search. MDN describes noopener and noreferrer as controls on how the browser opens the link and whether it sends a Referer header, and external as a note that the target is on another site. A link with rel="noopener noreferrer" is a followed link. So is one with rel="dofollow", because that isn’t a documented value and changes nothing.

When a directory says “dofollow”, read it as a promise that the link will have no qualifying value. Then check, because the word is used loosely.

Nofollow used to be an instruction. On 10 September 2019 Google announced the two new values and changed how it treats all three: they became hints about which links to consider or exclude in Search (Google’s announcement). For ranking, that applied from the day of the post. For crawling and indexing, nofollow became a hint on 1 March 2020.

Google’s stated reason was that links carry useful information, such as how the words in a link describe the page it points to, and that ignoring every nofollow link threw that away. The practical meaning for you is modest. Google may use a nofollow link or may not, it doesn’t say which, and you can’t find out for a given link. Don’t count on it, and don’t write it off.

The same post answered a few questions that still come up:

• Sites don’t need to change the nofollow links they already have.
• More than one value on a link is fine.
• The only attribute you can get wrong is on a paid link. An ad must carry sponsored or nofollow.

That last point matters when you buy a placement. Google’s spam policies treat paid links that pass ranking credit as link spam, and say paid links are fine when they are qualified (spam policies). A sponsored link on a paid listing is the directory following the rules.

In any desktop browser:

1. Open the page that links to you.
2. Right-click the link and choose Inspect.
3. Look at the highlighted <a> element. Read its href and its rel. No rel, or a rel without the three values, means followed.

To list every link to your domain at once, paste this into the Console tab, with your own domain:

document.querySelectorAll('a[href*="yoursite.com"]').forEach((a) =>
  console.log(a.href, "| rel:", a.rel || "(none)"));

Both methods show the page after scripts have run. To see what the server sent, use View Source, or the backlink checker, which reads the raw HTML without running JavaScript and reports each link’s anchor text and rel, plus a page-level nofollow if the robots meta tag sets one. The difference between the two views is the subject of the next section.

Links added by JavaScript

Google says a link inserted by JavaScript is crawlable as long as it is a real <a> element with an href (source). Things that only act like links, such as a <span> with a click handler or an <a> with onclick and no href, are ones it can’t reliably extract a URL from.

There is a timing cost. Googlebot first parses the HTML response for links, then queues the page for rendering, where it may wait a few seconds or longer (source). The same document notes that not all bots can run JavaScript. A link in the server’s HTML is seen by every crawler on the first pass. A link that appears only after rendering is seen by fewer of them, later.

Links through a redirect

Some directories link to directory.example/go/your-product, which then redirects to your site, so they can count clicks. Three things follow:

• The link on the page points to the directory’s own domain. A tool that looks for links to your domain won’t find it.
• Google treats a permanent redirect (301 or 308) as a signal that the target should be canonical, and a temporary one (302, 303 or 307) as no such signal (source). Check which kind the tracker answers with.
• If the directory’s robots.txt disallows the /go/ path, crawlers aren’t allowed to fetch that URL at all (source), so they never see where it leads.

Visitors still arrive. For search, a redirect link is an unknown, and a direct link is better. How to check that a directory really links to you shows how to test one.

• Followed, in the HTML, on an indexed page. The full version: visitors, discovery and a link search engines can count. How much they count it depends on the page it sits on, which nobody outside the search engine can measure.
• Nofollow or ugc. Visitors, and a hint Google may or may not use. Many of the links a new product can add for itself, on profiles and in communities, look like this.
• Sponsored. What a paid placement should carry. Judge it as advertising: by the people it sends.
• Followed, but on a page that isn’t indexed, or added by script, or behind a blocked redirect. Less than the label suggests. This is why “dofollow” on a pricing page tells you little until you look at a live listing.

Google finds most new pages through links (SEO Starter Guide), and people find products the same way. A nofollow link from a page your customers read does more for a new product than a followed link from a page nobody opens. When you compare directories, start with who visits them, and treat the rel as one line in the comparison.

Everything here comes from Google’s documentation. If another search engine matters to your market, read its own webmaster guidelines for how it treats these values.