Every outbound link on your site tells Google something about a relationship. Some links are editorial endorsements you stand behind. Others are ads, affiliate placements, or something a stranger dropped in your comments. Google gives you three rel attribute values to describe that difference: nofollow, sponsored, and ugc. Used well, they keep you clear of link-scheme trouble and give Google honest signals. Used carelessly, especially on your own internal links, they quietly work against you.
Key takeaways
- Google introduced
sponsoredandugcon September 10, 2019, to describe paid links and user-generated links more precisely thannofollowalone could. - Since March 1, 2020, Google treats
nofollowas a hint for crawling and indexing, not a strict directive. For ranking, all three attributes became hints in September 2019. - Use
sponsoredfor ads and paid or affiliate links,ugcfor comments and forum posts, andnofollowwhen none of the others fit and you would rather not vouch for the page. - Do not use these attributes on your own internal links to shape how authority flows. Google has said internal PageRank sculpting with
nofollowis a waste of time. - To keep a page on your own site out of Google, use a robots.txt disallow rule or a
noindexrule, not a link attribute.
What each attribute actually means
By default, a normal link passes signals to the page it points at. In Google’s words, for regular links that you expect it to fetch and parse “without any qualifications, you don’t need to add a rel attribute.” The three attribute values below exist for the links where you want to qualify that relationship, and Google documents all of them in its guide on how to qualify your outbound links.
rel=”nofollow” is the original. Google launched it in 2005 to fight comment spam, then expanded it as a way to flag advertising-related or sponsored links. Its job is to say you would rather not associate your site with, or pass ranking credit to, the linked page. Google’s current guidance is to reach for nofollow “when other values don’t apply, and you’d rather Google not associate your site with, or crawl the linked page from, your site.”
rel=”sponsored” and rel=”ugc” arrived together on September 10, 2019, when Google announced it was evolving nofollow into a small family of attributes. sponsored marks links that exist because of advertisements, sponsorships, or other compensation agreements, such as paid placements and affiliate links. ugc is recommended for links inside user-generated content, such as comments and forum posts. The point of splitting them out is precision: a paid link and a reader’s comment are not the same thing, and now you can tell Google which is which.
One practical detail from Google’s documentation: you can combine values. A single link can carry rel="ugc sponsored" or rel="nofollow ugc", as a space- or comma-separated list, which is useful when, say, a sponsored link also appears inside user-generated content.
When to use which
The decision is almost always about the nature of the link, not a guess about SEO impact. Ask what the link is, and the right attribute follows.
| Attribute | When to use it | Typical example |
|---|---|---|
rel="sponsored" |
The link exists because of advertising, a sponsorship, an affiliate deal, or other compensation | A paid product link inside a sponsored review, or an affiliate link in a roundup |
rel="ugc" |
The link sits inside content your users created | A URL a reader adds in your blog comments or forum thread |
rel="nofollow" |
None of the above fit, and you would rather not endorse or pass credit to the page | A source you cite for reference but do not vouch for |
| No attribute (a plain link) | An editorial link you trust and are happy to stand behind | A citation to a primary source or a helpful reference you chose freely |
A few honest notes. Google now says sponsored is the preferred value for paid links, though it confirms that using nofollow on them is still acceptable. And you are not obliged to re-tag old links: when the new attributes launched, Google was explicit that “there’s absolutely no need to change any nofollow links that you already have,” while suggesting you switch sponsored content to rel="sponsored" if and when it is convenient.
The 2020 “hint” change, explained
Here is the part that trips people up. Before 2019, nofollow was effectively a hard rule: any link carrying it was not counted at all in Google’s algorithms. The 2019 announcement changed that to a two-stage model.
For ranking purposes, all three attributes became hints on September 10, 2019. In Google’s own words, “all the link attributes, sponsored, UGC and nofollow, are treated as hints about which links to consider or exclude within Search.” For crawling and indexing, the change came a bit later: as Search Engine Journal reported, beginning March 1, 2020, Google began treating nofollow as a hint for crawling and indexing too, rather than as a directive that guaranteed a link would not be followed.
Why make the change at all? Google’s reasoning was that a hard directive threw away useful information. “Links contain valuable information that can help us improve search, such as how the words within links describe content they point at,” the company said. “By shifting to a hint model, we no longer lose this important information, while still allowing site owners to indicate that some links shouldn’t be given the weight of a first-party endorsement.” A hint model lets Google study anchor text and linking patterns to spot link schemes, while still respecting your signal in most cases.
In practice, Google told publishers not to expect significant changes to search results, and said that for most sites it would generally treat these links as it did before and not consider them for ranking. The word “hint” matters less as a day-to-day worry and more as a mental model: you are giving Google guidance it usually honors, not issuing a command it must obey.
There is one important consequence of the crawling-and-indexing change. If your only reason for adding nofollow was to stop Google reaching a page, that was never a reliable method and is even less so now. Google notes that pages behind these attributes “may be found through other means, such as sitemaps or links from other sites, and thus they may still be crawled.” For genuine crawl or index control, use the right tool, which we come to below.
The internal-link mistake to avoid
The single most common misuse of these attributes is putting nofollow on your own internal links. The old theory, called PageRank sculpting, was that you could hoard authority by nofollowing links to low-priority pages so more flowed to your important ones. That has not worked for a long time. Google’s John Mueller has said internal PageRank sculpting with nofollow is a waste of time, and Google closed that loophole back in 2009, when authority assigned to a nofollowed link began evaporating rather than being redistributed to your other links. Nofollowing an internal link does not concentrate authority. It just stops that link from doing its job.
Internal links are how you pass authority and context between your own pages, so treat them as first-class signals rather than something to throttle. If you want a fuller picture of how to route that authority deliberately, our guide to technical SEO covers site structure and the mechanics of how crawlers move through it, and our explainer on how search engines work shows why those internal signals matter for crawling and ranking.
So what do you use when you genuinely want to control a page on your own site? Google is direct about it. To stop Google fetching a link to a page on your own site, use the robots.txt disallow rule. To keep a page out of the index, allow crawling and use the noindex robots rule. Link attributes are for describing relationships with pages you link to, not for managing your own crawl budget or index coverage.
Examples in the wild
The clearest real-world example of a blanket attribute policy is Wikipedia. Because anyone can edit it, Wikipedia applies nofollow to its external links so it cannot be gamed by spammers seeking ranking credit. Search Engine Land uses exactly this case to explain the stakes of the hint change: “all external links on Wikipedia are nofollowed, if those links start counting, and you have a lot of links from Wikipedia, you might see your rankings improve.” It is a neat illustration of why a site with massive volumes of user contributions defaults to nofollow rather than vouching for every destination.
The same logic maps onto the newer attributes. A publisher running paid placements marks those with sponsored, which is Google’s stated preference for advertising and affiliate links. A platform with comment sections or forums marks reader-submitted links with ugc. Google even notes a nuance for ugc: if you want to recognize trustworthy contributors, you may remove the attribute from links posted by members who have consistently made high-quality contributions over time.
Frequently asked questions
Do nofollow links still have any SEO value?
Since the attributes became hints, Google can choose to consider a nofollowed link, but it says that in most cases it treats them as it did before and does not use them for ranking. Treat any ranking benefit as incidental, not something to rely on. The value of a nofollow tag is honest signaling and staying clear of link-scheme problems.
Should I use sponsored or nofollow for affiliate links?
Google prefers sponsored for paid and affiliate links, and says nofollow remains an acceptable alternative. Either keeps you compliant. If you are setting things up fresh, use sponsored.
Do I need to update all my old nofollow links to sponsored or ugc?
No. Google stated there is no need to change nofollow links you already have. You can migrate sponsored content to rel="sponsored" when convenient, but nothing breaks if you leave existing tags in place.
Can I use nofollow to keep a page out of Google?
No. Since March 2020, nofollow is only a hint for crawling and indexing, and a page can still be discovered through sitemaps or links elsewhere. To block crawling of your own page, use a robots.txt disallow rule. To keep it out of the index, use a noindex rule.
Written by Matthis Duarte, a senior SEO and organic growth expert with 10+ years of experience driving organic growth for international brands across highly competitive verticals. He is the founder of Knownful, an independent publication on SEO and organic growth featuring in-depth guides, best practices, playbooks and original analyses, including a free monthly study of which brands AI actually recommends across 10 industries.
