Nofollow vs dofollow, and why one of them does not exist

The nofollow vs dofollow debate has one small problem: dofollow is not a thing. There is no rel="dofollow" in Google's documentation. A followed link is simply a link with no rel value at all, and a nofollow link is one that carries rel="nofollow". Somewhere along the way the SEO trade gave the absence of an attribute its own name, and it stuck.

Here is the actual comparison, with Google's own pages as the referee.

The two kinds of link, side by side

Google's guide to qualifying outbound links puts it plainly: for regular links you expect Google to fetch and parse, you don't need a rel attribute. That is the whole definition of "dofollow".

The day nofollow went soft

For roughly 15 years, nofollow was a hard stop for ranking. Then on September 10, 2019, Google published Evolving "nofollow" and turned it into a hint. From that day it could use nofollow links for ranking if it chose, and from March 1, 2020 for crawling and indexing as well.

So the clean binary that "nofollow vs dofollow" implies has been gone for years. One side passes credit by default. The other might, and Google will not tell you when.

The same post added two siblings, which made the binary even less accurate:

• rel="sponsored" for ads and paid placements.
• rel="ugc" for comments and forum posts.

Values can be combined, for example rel="ugc nofollow".

Where the dofollow obsession goes wrong

The usual reason people ask about dofollow is that they want to buy one. Google's spam policies are specific here: buying or selling links for ranking is link spam, and that includes paying with goods or services, or sending a free product in exchange for a post with a link. The page also lists text ads that don't block ranking credit and guest posts with keyword-stuffed link text.

The fix Google offers is cheap. A paid link with sponsored or nofollow is fine. A paid "dofollow" link is the textbook case the policy describes.

Things a nofollow link still does

• It sends clicks, and clicks show up in your analytics.
• It can help Google find the page, and the target can be discovered through sitemaps or other links regardless.
• The words around it stay on the page. If that sentence names your product, people still read it, and so does any AI system that reads the page.

What it should not do is appear in your ranking spreadsheet as a win.

The page version nobody means

There is one more nofollow: the robots meta rule, <meta name="robots" content="nofollow">, which Google defines as "Do not follow the links on this page." It applies to every link on that page at once. There is no page-level "dofollow" either. Leaving the rule out is the default.

Settling it with data, not vibes

If you want to know what share of a site's links are plain versus nofollow, the Backlink Finder at seogeoaeo.ai reports, for the 100 strongest linking sites, the share of each site's linking pages that are followed rather than nofollow. One run costs 70 credits and works on any public domain. The numbers are a data provider's estimates, so treat them as a pattern, not proof of what Google counts.

The verdict

Nofollow vs dofollow is a comparison between a hint and a default. Plain links are what you hope to earn. Nofollow links are normal, mostly harmless and occasionally useful. Paid links get sponsored. Everything else is people arguing about an attribute that does not exist.