Why Expired Domains Fail as a Private Blog Network

A private blog network is a set of sites you own, usually built on expired domains, whose purpose is to link to a site you care about. It is the oldest use of expired domains and the reason the whole field has a reputation problem, and this page ends on don't.

Not primarily because of the policy risk, although that is real and the recovery is unpleasant. Mostly because the thing that makes a link valuable is that somebody chose to give it to you, and a link you gave yourself does not have that property no matter how convincing the site looks. Everything else, the footprint problem, the maintenance, the manual actions, follows from that one fact.

What is a PBN?

A network of sites under common ownership, built to link to a money site, with the ownership concealed. The domains are usually expired ones bought for their existing link profiles, on the theory that a domain with history passes more than a fresh one.

The concealment is the definitional part. Owning several sites is normal. Owning several sites and spreading their hosting, registration details and templates so that nobody can tell they are related is not a technical requirement of publishing. It is only necessary if the arrangement would not survive being seen.

Does Google's policy name PBNs?

No, and it is worth being precise about this because plenty of articles claim otherwise. The phrase "private blog network" does not appear in Google's spam policies for web search. Google writes policies about behaviour rather than about tactic names, which is why they age better than the tactics do.

Two sections cover it between them. Link spam is defined as "the practice of creating links to or from a site primarily for the purpose of manipulating search rankings", with examples running through buying links, exchanging goods or services for them, automated generation, and low-quality directory links with keyword-rich anchors. And expired domain abuse covers a domain "purchased and repurposed primarily to manipulate search rankings by hosting content that provides little to no value to users".

A PBN satisfies both definitions by construction. The links exist to move rankings, and the sites exist to carry the links. There is no reading of either sentence that a network gets out of, which is why the argument for PBNs has always been about detection rather than about compliance.

What has Google actually done?

The clearest record is 2014, and it is worth knowing because it is the last time the enforcement was public.

In March 2014 Matt Cutts announced action against a large guest blog network, reported by Search Engine Land, and the network in question, MyBlogGuest, confirmed the penalty itself. A second network, PostJoint, followed in April. In September 2014, Search Engine Land reported a wider wave of manual actions hitting sites using private blog networks, citing reports from affected site owners. Google's response to that story was characteristic: "We asked Google for a comment and they responded they do not comment if it's about particular manual actions."

What a manual action looks like from the receiving end is documented. Search Console's unnatural links documentation describes the trigger as "Google has detected a pattern of unnatural, artificial, deceptive, or manipulative links pointing to your site", and states that "Buying links or participating in link schemes in order to manipulate ranking in Google Search results is a violation of our spam policies."

Getting out of one is the part people underestimate. You request a review from the Manual Actions report and "Include documentation about the links you've had removed and an explanation of any links you were unable to remove." Which means dismantling the network, evidencing the dismantling, and waiting. Google will revoke the action "If we determine that your site is no longer in violation of our spam policies", on its own schedule.

The enforcement has been quieter since 2014, which is generally read two ways: either it has moved into algorithmic discounting, where links stop counting and nobody tells you, or it has become less of a priority. Both readings end in the same place. Silence is not evidence that the links are working.

Why do they fail?

Three structural reasons, none of which is fixable with better execution.

The footprint problem is unwinnable in principle. Shared hosting ranges, registration patterns, template reuse, publishing cadence, the same outbound link targets, the same anchor phrases. Each one is individually solvable at cost, and the cost compounds with every domain. You are trying to make a set of related things look unrelated to an organisation that specialises in finding relationships in web-scale data, and doing it on a budget.

The content has to be real to work and cannot be, at scale. A network site with genuine readers is a publication, and if you had one of those you would not be running a network. Filling the sites with generated articles lands you in a second policy: scaled content abuse, defined as pages "generated for the primary purpose of manipulating search rankings and not helping users".

The maintenance never ends. Every domain is a renewal, a host, a CMS to patch, and a site to keep alive. Stop paying for one and it lapses, which hands your link source to whoever registers it next.

There is a fourth, less discussed. A domain with a past can take a long time to shake it off. John Mueller, answering someone whose site on an old domain would not rank, said in September 2025 that "Sometimes it just takes a lot of time for the old state of a domain to be shaken off (sometimes that's also the case when it was parked for a while)", and that there is "nothing manual that you can / need to do here". Time is the only lever, and a network needs its domains working now.

A note on the statistics you will see

Searching this topic returns confident numbers: percentages of PBNs deindexed within a given number of months, detection rates for Google's spam systems, average recovery costs. We went looking for the studies behind several of them and could not find any. They are attributed to vendors who do not appear to have published them.

We are not going to repeat them. The case against PBNs does not need invented precision, and a page that cites a fabricated statistic to argue against manipulation has lost the argument it is making.

What about just one site?

The single-site version is the honest edge of this, and it depends entirely on whether the site would exist without the link. A publication you run, that people read, that occasionally links to your product because your product is relevant, is a publication. A site you would switch off the day it stopped serving your main site is a network of one.

Running an expired domain as a resource site sets out the six conditions that separate those two, and is honest about how few plans meet them.

Using the Revised directory for this

We do not want this business and it is worth saying why, plainly. Our directory publishes a listing's referring-domain band, its sources by name, its archived history, its age, its estimated value and a spam and trademark screen, all before you reveal the name. That is the wrong tool for building a network, where the requirement is volume and anonymity, and the right tool for the question a network never asks: is this one domain a genuine fit for a real thing I am building?

The uses that work are the ones where a person ends up served. Redirect a matched domain at a matching page. Rebuild the pages that earned the links. Put one real site on one good name. What to do with an expired domain compares the six options, and what Google's expired domain policy actually says has the verbatim text, including Google's own line that it is "fine to use an old domain name for a new, original site that's designed to serve people first."

For getting links the ordinary way, seven routes for a site nobody has heard of is slower and does not need to be hidden from anyone.

FAQ

Do PBNs still work in 2026? There is no credible public evidence either way, which is itself the problem: the ones that work are invisible and the ones that fail stop being discussed. What is documented is that they satisfy Google's link spam and expired domain abuse definitions by construction, and that the recovery process from a manual action requires dismantling the network.

Does Google's spam policy mention private blog networks? Not by name. It defines link spam as creating links primarily to manipulate rankings, and expired domain abuse as repurposing a domain primarily to manipulate rankings with low-value content. A PBN meets both.

How does Google detect a PBN? Google does not publish detection methods. The commonly cited signals are patterns rather than any single tell: shared hosting, registration and template patterns, outbound links converging on the same properties, and thin content. Treat any specific claim about detection thresholds as speculation.

Is one site linking to my main site a PBN? It depends on whether the site has a reason to exist besides the link. If it would be switched off the day it stopped linking to you, its primary purpose is the link.

What happens if I get a manual action for unnatural links? Your rankings drop and Search Console tells you. Recovery means removing or disavowing the links, requesting a review with documentation of what you removed and what you could not, and waiting for Google to review it.

Is buying expired domains itself against the rules? No. Google's policy targets repurposing a domain to manipulate rankings with low-value content, and its own announcement says using an old domain for a new, original site designed to serve people is fine.