Do PBN Links Still Work? How Google Neutralizes the Credit
Key Takeaways
- Backlinks remain strongly correlated with rankings, so the answer cannot rest on "links stopped mattering" — it rests on what happens to these particular links
- Google's stated response is neutralization: the credit an unnatural link would pass is discounted to nothing, and no notification is sent to either site
- Because the failure is silent, a flat ranking curve after buying links looks identical to not having bought enough — which is what keeps the market alive
- In March 2024 Google made expired domain abuse its own named policy violation, putting the cheap input that makes networks affordable inside the rules
Backlinks are not a spent force. A 2020 Backlinko analysis of 11.8 million Google search results, run with data from Ahrefs, found that the #1 result carries an average of 3.8 times more backlinks than positions two through ten, and that site-level link authority correlates strongly with where pages land. That finding is why the question in this title keeps getting asked. If links move rankings that much, wondering whether cheap links move them a little is a reasonable thing to wonder.
So do PBN links still work? For the ranking outcome the buyer is paying for, no — and the mechanism matters more than the verdict, because it explains why the failure takes so long to detect. Google's documented response to unnatural links is to neutralize the credit they pass rather than to punish the site receiving them. The links stay live. They keep appearing in every third-party backlink tool. They carry nothing.
That gap between what the dashboard shows and what the ranking algorithm counts is the whole story. A penalty would at least be informative. Neutralization returns a flat line, and a flat line is exactly what under-investment looks like too.
| What the purchase assumes | What Google's documentation describes | |
|---|---|---|
| The linking site | An independent publisher with earned authority | Detectable as a site "used for the purpose of passing outgoing links" |
| The credit | Authority flows to the target page | Credit passed by unnatural links is "lost" |
| The feedback | Rankings confirm whether it worked | No notification is sent for algorithmic neutralization |
| The risk being managed | A penalty, which you would hear about | Silence that reads the same as having done nothing |
Links Still Move Rankings, Which Is Why the Answer Needs a Mechanism
Any answer that starts with "links don't matter anymore" loses the reader who has watched a competitor climb after a link campaign, and it should. The Backlinko study measured several factors and did not find uniform support for all of them — it reported essentially zero correlation between keyword-in-title-tag and first-page position, and no correlation between page load speed and rankings. Against that, the link findings held up: more referring domains tracked with better positions, and domain-level authority tracked strongly.
"Pages with lots of backlinks rank above pages that don't have as many backlinks. In fact, the #1 result in Google has an average of 3.8x more backlinks than positions #2-#10."
— Backlinko, analysis of 11.8 million search results, 2020
Correlation studies like this one describe what winning pages have in common, not what caused them to win — pages that earn links also tend to be the pages worth linking to. The number is still useful for narrowing the question. Links, in aggregate, are doing work. So the honest form of the question is not whether links function but whether a specific class of link gets counted, which turns the answer into a matter of detection rather than a matter of principle.
Where PBN Links Still Work: Your Backlink Tool, Not Google's Scoring
In December 2022 Google published the December 2022 link spam update, describing how its AI-based spam prevention system had been pointed at unnatural links. The scope described in that announcement is the part worth reading twice, because it names both ends of the transaction.
"Besides using it to detect spam directly, it can now detect both sites buying links, and sites used for the purpose of passing outgoing links."
— Google Search Central, December 2022 link spam update
A site whose function is to publish articles containing outbound commercial links is described by the second half of that sentence with no interpretation required. The same announcement states the consequence in plain terms: "Ranking may change as spammy links are neutralized and any credit passed by these unnatural links are lost."
Two properties of that outcome shape everything downstream. The first is that nothing gets removed. The link element stays in the HTML, the linking page stays indexed, and crawlers from Ahrefs, Semrush and Majestic keep reporting the link, because those tools measure the web rather than Google's internal scoring. A backlink dashboard showing forty new referring domains is describing reality accurately and telling you nothing about whether Google counted any of them.
The second is that no message arrives. Manual actions come with a Search Console notice naming the problem, a mechanism covered in more depth in the companion piece on how private blog networks work. Algorithmic neutralization sends nothing. The buyer's experience is a set of live links and a ranking curve that did not respond, which is indistinguishable from having bought too few — and that ambiguity is a large part of why the market persists. Forum threads arguing that networks still work and threads reporting they stopped working can both be describing accurate observations, because neither side can see whether credit was ever passed.
In 2024 Google Made the Cheap Input Its Own Violation
Networks are affordable because they do not build authority, they buy it. Expired domains carry backlinks from a previous working life, and auction prices are set by that residual profile. Remove access to cheap residual authority and the model stops penciling out.
On 5 March 2024 Google announced three new spam policies alongside a core update, and one of them addressed that input directly. Expired domain abuse is defined as buying a lapsed domain and repurposing it "primarily to manipulate Search rankings by hosting content that provides little to no value to users." Google added a sentence ruling out the accidental reading: "Expired domain abuse isn't something people accidentally do."
The significance for this question is that the violation now attaches to the acquisition pattern rather than only to the links. A network domain can be flagged on the basis of what was done with the domain itself, separately from any assessment of who it links to. Google also maintains a running log of spam updates, which is the practical reason detection tends to move in one direction: each pass raises the cost of staying undetected, and none of them lower it.
| Cost centre | Network link | Earned link |
|---|---|---|
| Up front | Domain acquisition priced on residual authority | The work that made the page worth citing |
| Recurring | Hosting, registration and content on every site, for as long as the link needs to stay live | None after publication |
| Direction of travel | Depreciates as detection improves; the acquisition pattern is now a named violation | Unaffected by spam enforcement |
| Failure signal | Silent — no notice that credit stopped being passed | Visible; the host removes the link or the page changes |
| Cleanup if it goes wrong | Removal requests, disavow, reconsideration if a manual action landed | Nothing to clean up |
Set against that, the case for earned links is unglamorous and holds up arithmetically: a link given because a page was worth citing costs nothing to maintain and carries no enforcement exposure. It is slower. It also does not need to be defended every time a spam update ships.
One question worth settling before treating a plateau as a link problem: whether the constraint is external at all. Ranking stalls caused by blocked crawling, duplicate canonicals or unrendered content produce the same flat curve, and those causes sit inside your own templates where they can be measured. MendMySEO crawls the site, scores it across six technical dimensions, and hands engineering the exact markup to change. Join the waitlist.
Frequently Asked Questions
Does PBN still work?
For passing ranking credit, the documented answer is no: Google states that credit from unnatural links is neutralized, and its spam detection explicitly covers sites used to pass outgoing links. What still "works" is the appearance of progress — the links go live, backlink tools count them, and reports look healthy. The gap between those two things is where the money goes.
Are backlinks still relevant in 2026?
Yes, and treating that as contradictory misreads the mechanism. Google discounts links it classifies as manipulative; it does not discount links generally. Editorially given links from sites with their own audiences continue to correlate with rankings, which is the finding the Backlinko study measured. Neutralization is targeted at a class of link, not at the signal.
How do I know whether a link I bought was neutralized?
Directly, you cannot. No report exposes per-link credit, and no notification is sent. The available evidence is indirect: whether rankings for the targeted terms moved in the weeks after the links went live, whether the linking pages get organic traffic of their own, and whether they are indexed at all. A linking page missing from Google's index is a strong signal, since a page that is not indexed cannot pass anything.
What should I do about network links already pointing at my site?
Start by checking Search Console for a manual action. Without one, Google's position is that most sites need no action, because unnatural links are usually handled algorithmically. Where links were bought and can be removed, ask for removal. Where they cannot be removed and a manual action is in play, the disavow links tool exists for exactly that case — it is intended for known unnatural links you are unable to take down, not as routine maintenance.
Do network links work on new domains or in low-competition niches?
This is the strongest form of the counter-argument, and it holds partially. In a query with few competing pages, small signals decide positions, so movement after a link campaign is easy to observe and easy to attribute. Two things undercut it as a strategy. Rankings in thin competition can move on almost any change, which makes attribution unreliable. And the position is held on an asset that depreciates with each spam update, on a domain history that is public — meaning the cheapest wins are the ones with the least durable foundation.