Dofollow vs Nofollow Links: What Actually Matters
The dofollow/nofollow split is real. It is also the wrong first question to ask about a link.
A dofollow link is an ordinary link that search engines are free to follow and count. A nofollow link carries a rel="nofollow" attribute that asks them not to.
That is the entire mechanical difference and it fits in one sentence. The reason the question gets asked tens of thousands of times a month is that people want it to answer a different question: which links are worth my time? The rel attribute is a poor answer to that question. It is one bit of information, set by a publisher for their own reasons, describing a relationship that stopped being binary in 2019. Sorting a prospect list dofollow-first feels like rigour. In practice it filters out some of the most valuable placements on the web and waves through a lot of junk.
This piece is the honest version: what the attributes do, what Google has actually said about them, how to check a link in ten seconds, and when a nofollow placement is worth the effort anyway.
The short version
- "Dofollow" is not a real HTML attribute. It is what SEOs call a link with no rel attribute at all.
nofollow,sponsoredandugcare the three values that qualify a link. Google's documentation says links marked with them "will generally not be followed."- Since 2019 Google treats all three as hints, not directives, for ranking. Since 1 March 2020 that applies to crawling and indexing too.
- A backlink profile made only of dofollow links looks engineered, because the real web is full of nofollow.
- The better filter is not "does this pass link equity" but "will a real person click this."
What is the difference between a dofollow and a nofollow link?#
The difference is the presence of one attribute in the HTML. A plain link looks like this:
<a href="https://genlink.so/">GenLink</a>
A nofollow link looks like this:
<a rel="nofollow" href="https://genlink.so/">GenLink</a>
There is no rel="dofollow". It does not exist in the spec, Google does not read it, and if you see it in a CMS setting it is a label the developer invented for a checkbox that removes the nofollow. "Dofollow" is SEO slang for the default state of a link. This matters more than it sounds: when a site tells you they offer dofollow links, they are describing the absence of a thing, and you should confirm it in the HTML rather than take the word.
The reason anyone cares is that links are how discovery and authority both work. A crawler follows links to find pages, and the pattern of who links to whom is one of the oldest inputs into ranking. If you are new to the mechanics, start with what a backlink actually is and come back, because everything below assumes it.
What does rel="nofollow" actually tell Google?#
It tells Google you do not want your site associated with the linked page, or do not want to vouch for it. Google's own documentation on qualifying outbound links defines three values, and it is worth reading their exact framing rather than the folklore version.
| rel value | Google's stated purpose | Typical use |
|---|---|---|
| (none) | A regular link "you expect Google to fetch and parse without any qualifications" | Editorial citations, your own internal links |
rel="sponsored" |
Mark links that are advertisements or paid placements | Paid reviews, ads, affiliate links, sponsorships |
rel="ugc" |
Mark user-generated content links, "such as comments and forum posts" | Comment sections, forums, user profiles |
rel="nofollow" |
"Use the nofollow value when other values don't apply, and you'd rather Google not associate your site with, or crawl the linked page from, your site" | Untrusted or unvetted outbound links |
Two details from that page that people routinely get wrong. First, Google's wording is that these links "will generally not be followed" — not that they are ignored, and not that they are penalised. Second, you can stack values: rel="ugc nofollow" is valid, and multiple values can be space- or comma-separated.
Google also notes that sponsored is now preferred over nofollow for paid placements, though nofollow "is still an acceptable way to flag them." If you are buying or selling anything link-adjacent, this is not optional housekeeping. Google's spam policies list "exchanging money for links, or posts that contain links" as link spam, and then say plainly that buying and selling links "is not a violation of our policies as long as they are qualified with a rel="nofollow" or rel="sponsored" attribute value."
Did nofollow stop being a directive?#
Yes, in 2019. This is the single most under-appreciated fact in the dofollow vs nofollow debate, and it is in a Google blog post most people have never read.
In Evolving "nofollow", Danny Sullivan and Gary Illyes wrote that when nofollow was introduced, "Google would not count any link marked this way as a signal to use within our search algorithms. This has now changed. All the link attributes — sponsored, ugc, and nofollow — are treated as hints about which links to consider or exclude within Search."
The post gives the reason: "Links contain valuable information that can help us improve search, such as how the words within links describe content." And it dates the rollout precisely. The attributes became ranking hints immediately, and "for crawling and indexing purposes, nofollow will become a hint as of March 1, 2020."
So the clean binary that most link-building advice still assumes — dofollow passes value, nofollow passes zero — has not been Google's stated position for years. What replaced it is fuzzier and, for practical purposes, more honest: Google decides, link by link, what to do with the information. You do not get to know the outcome. Which means building a strategy on the attribute alone is building on a variable you cannot observe.
Do nofollow links help SEO?#
Sometimes, and rarely in the way people hope. The right way to hold it is that a nofollow link has zero guaranteed ranking effect and several reliable second-order effects.
The second-order effects are the real argument:
- Referral traffic. A link in a widely-read newsletter or a busy subreddit sends people. People convert. The attribute is irrelevant to a human reader.
- Discovery by other publishers. Journalists, bloggers and researchers find sources through exactly these venues, then cite you from their own editorial pages, which are usually not nofollowed. One nofollow mention seeding three followed ones is a common pattern.
- Citation in AI answers. Assistants that summarise the web do not filter their source sets by rel attribute. Being mentioned in the places they read is its own channel.
- Profile realism. This is the one people miss. A natural profile is mostly nofollow, because the biggest link sources on the web nofollow by default.
That last point is easy to verify yourself instead of taking anyone's word for it. Open Wikipedia's own article on search engine optimization, view the source, and search for rel="nofollow". Essentially every outbound citation on the page carries it, by policy. The most cited encyclopedia on the internet does not pass a drop of link equity to anyone, and it is still one of the most valuable places on the web to be cited. Reddit, YouTube descriptions, most forums and most comment sections behave the same way. If your backlink profile contains none of that, it does not look like a profile a real brand accumulates.
How do you check if a link is dofollow or nofollow?#
Read the HTML. There are three ways, in increasing order of speed.
In the browser. Right-click the link, choose Inspect, and look at the <a> tag highlighted in the Elements panel. If there is no rel attribute, it is a dofollow link. Do not trust a browser extension badge without spot-checking, because extensions read the rendered DOM and some sites inject or strip attributes after load.
In view-source. Press Ctrl+U (Cmd+Option+U on Mac), then Ctrl+F for your domain. This shows the raw HTML the server sent, which is what a crawler sees first.
From the terminal, which is the one to use when you are checking a list of placements:
curl -sL "https://example.com/the-post" | grep -o '<a[^>]*yourdomain\.com[^>]*>'
That prints every anchor tag pointing at your domain, attributes included. If rel is absent, you have a followed link. Run it again a week after a placement goes live, because "we'll make it dofollow" and "it is still dofollow next month" are different claims, and some sites sweep old outbound links into nofollow on a schedule.
One catch that trips up even careful people: the three methods can disagree. The HTML the server sends is not always the HTML that ends up in the DOM, because a script can add or strip a rel attribute after load. Google renders pages before it decides what to do with the links on them, so the rendered version is the one that matters.
Which nofollow links are worth chasing?#
The ones where the audience is the point. The attribute tells you almost nothing on its own; cross it with whether a relevant human reads the page and the answer falls out immediately.
That is the rule. The list below is for the edge cases, because a few placement types are not obvious until you name what you are actually buying with the effort.
| Placement | Usually nofollow? | What you are actually buying | Effort worth it? |
|---|---|---|---|
| Major publication editorial mention | No | Ranking signal, referral traffic and the credibility to get the next one | Yes, the highest value link there is |
| Niche industry blog post | No | A signal from a topically adjacent site, plus readers who already care | Yes, relevance beats raw authority |
| Wikipedia citation | Yes | Third-party validation that journalists and AI assistants both read | Yes, but earn it with a source worth citing |
| Reddit or a busy forum thread | Yes | Referral traffic and discovery by people who write elsewhere | Yes, if you are a real participant in the thread |
| Sponsored post or paid placement | Yes, and must be | Advertising. Nothing else, and Google requires it be labelled | Only on advertising maths, never on link maths |
| Blog comment with your link dropped in | Yes | Nothing. Google names this pattern in its link spam policy | No |
| Social profile and directory links | Yes | A tidy, consistent brand footprint | Do them once in an afternoon, then stop |
| YouTube video description | Yes | Referral traffic from an audience that watched you for ten minutes | Yes, if the video reaches your buyers |
| Low-quality "we accept guest posts" site | Varies | A link on a page with no readers and a footprint Google knows | No, whatever the attribute says |
Notice that the last column barely tracks the second one. It tracks the third almost perfectly, and the third column is only ever answering one question: does a relevant human read this page. That is the filter to run your prospecting on, which is why the work of finding the right sites and the right people matters far more than any attribute audit.
The hard part of that filter is not deciding it, it is doing it across a few hundred pages. That is the job GenLink takes: it searches out the pages in your niche that real readers actually land on, works out who runs them, and puts a specific pitch in front of that person. Start free.
When should you nofollow your own outbound links?#
Whenever money or a stranger is involved. The rule is simpler than the debate around it:
- Anything paid, affiliate or gifted gets
rel="sponsored". This includes the product someone sent you in exchange for a review, which Google's spam policies name explicitly. - Anything a user typed gets
rel="ugc", optionally stacked with nofollow. - Anything you cannot vouch for gets
rel="nofollow". - Everything you are citing because it is genuinely useful gets no rel attribute at all. This is the default and it should stay the default. Sites that nofollow every outbound link to "preserve equity" are hoarding a resource that does not deplete, and they read as stingy to the people whose links they want.
That last habit has a direct cost when you start asking others for links. Editors check whether you link out before they link to you.
What does this mean for how you build links?#
It means you stop treating the rel attribute as a qualification criterion and start treating it as metadata you record after the fact. The sequence that works is: find pages where your audience already is, judge relevance and readership, pitch something genuinely useful, and log the attribute you got. Not: filter a database for dofollow, mail the survivors, and wonder why the reply rate is what it is.
The practical playbook for that sits in how to get more backlinks, and the pitch mechanics in link building outreach. Both start from the same premise as this piece. A link is a decision a specific person made. The attribute is a footnote on that decision, not a substitute for earning it.
GenLink is built around that sequence. It finds relevant sites, identifies the contact, writes the personalised pitch, sends from warmed mailboxes, follows up and triages replies, so the part that takes human hours is the part that gets automated. The judgement about which placements matter stays yours.
Stop grading links by attribute and start grading them by readership. Open a free GenLink agent and point it at your niche.
Frequently asked questions#
What is the difference between dofollow and nofollow links?
A dofollow link is a normal link with no rel attribute, which search engines are free to follow and count as a signal. A nofollow link includes rel="nofollow", which asks Google not to follow it or associate your site with the destination. Google's documentation states such links "will generally not be followed," and there is no actual rel="dofollow" attribute in HTML.
Do nofollow links help SEO?
They have no guaranteed direct ranking effect, but they help indirectly and reliably. They send referral traffic, get you discovered by publishers who then cite you editorially, appear in AI-generated answers, and keep your backlink profile looking natural. Since 2019 Google treats nofollow as a hint rather than a strict directive, so the effect is not necessarily zero.
Is nofollow still a directive or just a hint?
A hint. In September 2019 Google announced that sponsored, ugc and nofollow "are treated as hints about which links to consider or exclude within Search." The change applied to ranking immediately, and Google stated that for crawling and indexing purposes nofollow would become a hint as of 1 March 2020.
What is the difference between nofollow, sponsored and ugc?
They describe why the link is qualified. Use sponsored for advertisements and paid placements, ugc for user-generated content like comments and forum posts, and nofollow when neither applies but you do not want to endorse the destination. Google now prefers sponsored for paid links, though it says nofollow remains acceptable for them.
How do I check if a backlink is dofollow?
Open the linking page, right-click your link and choose Inspect, then read the <a> tag in the Elements panel. No rel attribute means dofollow. Check the rendered DOM rather than only view-source, because a script can add or strip the attribute after load, and Google renders pages before deciding what to do with the links on them.
Should I refuse a nofollow backlink?
No. Judge the placement on whether relevant people read the page, not on the attribute. A nofollow link on a site your buyers actually read beats a followed link on a site nobody visits. The only links worth refusing are the ones that violate Google's link spam policies, whatever attribute they carry.
