A nofollow link is a hyperlink carrying a small instruction in its code that tells search engines not to treat it as a full endorsement. When you ask what a nofollow link is, who uses it, and why it exists, the short answer is control. Site owners add it to links they cannot fully vouch for, and search engines read it as a signal about how much trust to pass along.
How does it work? A normal link quietly passes authority from one page to another. A nofollow link still takes a visitor to the destination when clicked, but it asks crawlers to discount that connection. Where does it appear most? In comments, forums, sponsored placements, and user profiles.
This guide stays deliberately mechanical. It explains what the attribute is, how crawlers interpret it, where it came from, and how to spot it in page HTML without guessing.
At code level, nofollow is not a separate kind of link. It is a value inside the rel attribute of an otherwise ordinary anchor tag. The destination and the visible anchor text stay exactly the same.
A standard link is an anchor tag with an href pointing at the destination. A nofollow version adds one thing: rel="nofollow" inside that same tag. That single word is the entire mechanism.
The rel attribute can hold several values at once. A link marked rel="nofollow sponsored" is still a nofollow link, just with extra context about why it was marked.
For most of its life, nofollow was treated as a directive. A crawler that saw it excluded the link from ranking calculations, and the linking page passed no authority through it.
That treatment later changed to a hint model. Under the hint model, described in Google guidance on qualifying outbound links, the engine reserves the right to consider the link anyway. It may use it for discovery or as a weak signal, depending on context.
A directive means the engine promises to ignore the link for ranking. A hint means it will usually discount it, but it makes the final call itself. Three practical consequences follow:
What did not change is the visitor experience. A nofollow link clicks and loads exactly like any other link. The attribute speaks only to crawlers, not to people. For broader background, the Seovaro blog covers related link topics in the same plain style.
Nofollow was introduced as a defence against comment spam. Blog comments had become an easy way to drop links on thousands of sites, and those links manipulated rankings because engines counted each one as a vote. Letting blog owners mark comment links as untrusted removed the reward without removing the conversation.
From that narrow start, use cases widened. Paid links, advertorials, press releases, forum signatures, and any link a site published but did not editorially choose all became candidates. The logic stayed the same: if you did not genuinely endorse the destination, you could say so in code.
Two companion values were later added for precision. The sponsored value marks advertising and paid placements. The ugc value marks links inside user generated content such as comments and forum posts. Nofollow remains the general purpose option when neither specific label fits.
Because the system grew in layers, you will meet all three values in the wild, sometimes combined on one link. Sponsored declares a commercial relationship. Ugc declares the source of the content. Nofollow declares a withholding of endorsement.
You need no specialist software to check a link. You need the source code behind the page and the ability to read the anchor tag itself. Guessing from where a link sits is unreliable, because identical looking links can carry different attributes.
rel attribute, if one is present, and note every value it contains.No rel attribute at all means a standard followed link by default. There is no dofollow attribute in HTML, despite how often the word is used. Followed is simply the absence of a qualifying value.
Extensions and audit tools can flag attributes across a whole page at once, saving time on long pages. The underlying check is identical: every tool reads the same rel value you can read yourself in the inspector.
Clear link signals work best on clearly described pages, so try the AI SEO meta description generator to draft accurate descriptions for your own content and keep every signal on your pages clean and deliberate.
Under the hint model an engine may choose to consider it, but plan as if it passes little or none. Its reliable value is referral traffic and discovery, not ranking authority you can count on.
No. There is no dofollow value in HTML. People use the word informally to mean a normal link with no qualifying rel value. Followed behaviour is the default state of a link.
Yes. The rel attribute accepts multiple space separated values, so a link can declare both a commercial relationship and a withholding of endorsement at the same time.
Not reliably. Crawlers may still visit and index the destination, especially under the hint model. Use robots controls, not link attributes, when you need to manage crawling itself.