Search for advice on backlinks and you will drown in it. Vendors sell them by the thousand, tools score them on a dozen different scales, and a lot of the guides still read like they were written in 2013. The reason so much of it contradicts itself is that most of it rests on five assumptions that either expired years ago or were never true in the first place.
Here is where I land before we start: backlinks still matter, but not as a number you accumulate. They work as evidence of reputation, and the system doing the judging has gotten much better at telling a real endorsement apart from a bought one. Every myth below is a version of missing that distinction.
If you only have one screen’s worth of attention, the table below is the honest summary. Everything under it is the reasoning, because the reasoning is where the money is actually saved.
| Myth | What the evidence actually shows | What to do instead |
|---|---|---|
| More backlinks equal higher rankings | Correlation, not causation. Unique referring domains track rankings more closely than raw link count. | Chase relevance and one strong link over ten weak ones. |
| Nofollow links are worthless | Google has treated nofollow as a hint since 2019, and AI search engines ignore the attribute entirely. | Accept nofollow links from good sources. Stop pricing them at zero. |
| Domain Authority is Google’s metric | DA and DR are third-party scores. Google uses neither. | Use them as rough guides, never as targets. |
| You must disavow every toxic link | Google ignores most spam automatically, and “toxic” is not a Google concept. | Leave it alone unless you have a manual action to clear. |
| Buying links always triggers a penalty | The rules target manipulation and non-disclosure, not the transaction itself. | Vet relevance, and make sure any paid placement is properly qualified. |
Sources for the table: Google’s published spam policies and link-attribute guidance, Backlinko’s 11.8 million-result ranking study, and Ahrefs’ 75,000-brand visibility study. Dates and links are in the text and the methodology note below.
Myth 1: More backlinks mean higher rankings
The most stubborn backlink myth is also the easiest to dismantle. Backlinko’s study of 11.8 million Google results found that the number-one result has, on average, 3.8 times more backlinks than the results ranked two through ten. Vendors quote that line constantly. They leave out the part that makes it interesting.
That study, and every other large one like it, measures correlation. Pages rank because they are genuinely useful, and useful pages tend to attract links. The link count is a symptom of being worth linking to, not the cause of the ranking. Read the same dataset another way and the messiness shows: roughly 95 percent of all pages have zero backlinks, and the number of unique referring domains correlates more strongly than the total count. Ten links from ten different sites beat ten links from one site with a footer widget.
If a report leads with total backlinks, I close it. Sort by referring domains or do not bother. The count is the least interesting number in any backlink tool.
Myth 2: Nofollow links are worthless
This one was true until September 10, 2019, which is presumably why it refuses to die. That is when Google announced it was treating nofollow, sponsored, and ugc as hints rather than hard directives. Google can now choose to use a nofollowed link for crawling, indexing, or ranking if its systems decide the link is a genuine signal.
In practice, it does not promise to count any particular nofollow link, so you should never build a strategy that depends on it. But “worthless” is still the wrong word. A nofollow mention on a site your buyers actually read sends referral traffic, reinforces your brand as an entity, and often sits on a page that later earns you a followed link anyway. A profile that is 100 percent dofollow is a red flag, not an achievement.
And in AI search, the attribute stops mattering almost entirely. Chatbots and AI Overviews retrieve and cite pages on relevance and reputation, not on the HTML of a link tag. Some of the most frequently cited sources across those systems, like community forums and encyclopedia pages, are nofollow by default. Call those links zero and you are arguing with the citation data.
Myth 3: Domain Authority is the number Google uses
Domain Authority from Moz and Domain Rating from Ahrefs are useful estimates built by companies that are not Google. Google does not use either one. This matters because teams routinely set “get to DR 60” as a goal, then optimize for a metric that exists only on a scoreboard they invented.
My rule of thumb is blunt: a link from a domain rating 30 site that is dead-on your topic is worth more than a link from a rating 80 site that has nothing to do with you. Relevance multiplies the value of a link; authority without relevance mostly adds a nice-looking number to a spreadsheet. Use these scores to compare candidates that are otherwise similar, then stop.
Here is the part that stings: chasing authority is how sites end up with links from general news outlets and coupon directories while their actual niche peers never mention them once.
Myth 4: You should disavow every sketchy link
Backlink audit tools are happy to label links “toxic.” Google has no such concept. In March 2026, Google’s John Mueller answered a question on Bluesky about whether disavowing was worth it and said, “Most sites don’t need it, but that’s not all sites.” He has spent years making the same point: if a random tool can flag a bad link, Google’s own systems can almost certainly ignore it without your help.
The disavow file exists for specific situations. You have a manual action for unnatural links, or you personally built or bought links that violate the guidelines and you want them gone. For the everyday site with a normal mix of links, though, the disavow ritual is closer to anxiety management than SEO. (Yes, I have watched people disavow a legitimate industry magazine because its rating looked low. It does not end well.)
Unless you are in one of those two situations, leave your bad links alone. You have better uses for the afternoon.
Myth 5: Buying backlinks always gets you penalized
The rule is narrower than the folklore suggests, and getting it right is where a lot of businesses either burn money or leave growth on the table. Google’s spam policies prohibit links created “primarily for the purpose of manipulating search rankings,” and they explicitly call out buying or selling links that pass ranking credit without being disclosed. The violation is the manipulation and the lack of disclosure, not the fact that money changed hands.
Google says as much itself: buying and selling links is a normal part of the web economy, and it is not a violation as long as the links are qualified with a rel=”sponsored” or rel=”nofollow” attribute. So the honest test is not “paid versus free.” It is whether the placement would exist on its merits and whether the commercial relationship is disclosed.
That test also tells you how to judge anyone offering backlinks for SEO. Ask which sites, ask who decides placement, and ask how links are qualified. A provider that answers those three questions clearly is working with the rules. One that promises “guaranteed followed links on DR 50-plus domains, no disclosure needed” is selling you a future cleanup job.
One more thing worth internalizing, because it changes how you spend: across 75,000 brands, Ahrefs found that unlinked brand mentions correlated with AI Overview visibility at 0.664, against 0.218 for backlinks. A link is one form of being vouched for. In AI search, being named at all is now the stronger signal. That does not make backlinks pointless for classic Google rankings, which still run on them, but it does mean the company you keep matters more than the connections you buy.
Frequently asked questions
Are backlinks still a ranking factor? Yes. Google’s own documentation lists links as a core signal, and large correlation studies keep showing that pages with stronger link profiles rank higher. What has changed is that quality and relevance now do most of the work, and the raw count matters far less than it once did.
Do nofollow links help SEO at all? Indirectly, and sometimes directly. Since 2019 they are hints, not directives, so Google may count a good one. They always bring referral traffic and brand reinforcement, and AI engines do not look at the attribute at all.
Is Domain Authority what Google ranks by? No. Domain Authority and Domain Rating are third-party estimates from Moz and Ahrefs. Google does not use either metric. Treat them as rough comparison tools, not as scores to grind upward.
Should I disavow bad backlinks? Only if you have a manual action for unnatural links, or you know you built links that break the guidelines. For most sites, Google handles spam automatically, and a hand-built disavow file is more likely to remove something useful than to fix a problem.
How many backlinks do I need to rank? There is no number. The useful question is how many credible, relevant sites in your field genuinely reference you. One link from a trusted niche publication can outperform hundreds from directories and low-quality guest posts.
Can you rank without any backlinks? Yes, for lower-competition and long-tail queries, especially if your site has clear topical coverage and clean structure. For competitive commercial keywords, a strong link profile is usually the difference between ranking and hoping.
How this article was put together
I wrote this to answer one question: which commonly repeated claims about backlinks survive contact with Google’s own documentation and the largest public studies available. I read Google’s spam policies and its guidance on qualifying outbound links directly, rather than summarizing them second-hand, and cross-checked the ranking claims against Backlinko’s 11.8 million-result study and Ahrefs’ 75,000-brand AI visibility research. The John Mueller quote comes from his March 2026 comments, as reported by Search Engine Roundtable. I could not independently verify every “correlation equals causation” claim vendors make, so I have treated the link-ranking relationship as directional rather than proven. Figures will need rechecking as Google updates its spam policies and as the AI visibility studies are refreshed. This article contains one commercial reference to a backlink provider, marked inline.
