Author Shai Aharony

Dec 17th 2018

Prodding Google’s algorithms just for fun

Google Killed the PBN

Reboot Line

Google killed the PBN

It was early 2000 when we first started building our own PBN. A PBN or Public/Private Blog Network is a collection of websites built on domains that have normally expired with some previous history. The more history in the way of backlinks the domain had, the better it was. 

These domains are normally re-registered or bought at registrar auctions and are then treated to a splash of Wordpress with some cheap content with links to your target site. It worked amazingly well. A few links from a decent PBN and your site would leapfrog the competition leaving them to scratch their heads in wonderment on how you did it.

pbn seo ranking
2014 example of ranking improvments which was possible on a very competative keyword [end of tenancy cleaning]

If you were really clever in those days, you would hide the PBN from the usual bunch of link indexers such as Majestic, Open site explorer and Ahrefs via the robots.txt or .htaccess files, meaning that it would be almost impossible to spot the links from the PBN. This would be sure to leave your competitors infuriated as they watched helplessly whilst their site that they spent years building its reputation and links get overtaken overnight seemingly effortlessly by a domain that had (apparently) in some cases 0 links. 

As the years went by, more people started jumping on the PBN gravy train and  it became harder to find domains that have not been “done” or abused by previous domain owners. It had reached the point where to get a decent domain, you would not only have to check that its backlink profile was clean, but you would also have to delve into its history looking out for any signs that the domain you are about to buy had not been used in an exposed PBN which would normally mean it had been caught and penalized by Google. (otherwise, why would the domain owner let it go?)
Although it involved a little more work, the gravy train continued roaring onwards. But make no mistake about it. It's slowing down rapidly.

In later years, Google would get better at detecting footprints leading to quite a few high profile cases of PBNs being deindexed. Having hundreds of sites hosted on the same IP and all linking to one target made it too easy but there were lots of other such footprints. Google would regularly discover such footprints of PBNs and would use those to wipe them off its index in one fell swoop. Sometimes taking all those that were linked to the PBN with it, causing huge outcries from people who in some cases were not even aware that their “SEO Expert”! was using such tactics. Wave after wave of penalties, highly publicized to raise fear amongst any new players who may be tempted to start their own PBN. But all those in the know, carried on. The good ones have developed ways of reducing the footprint of such domains. They went to lengths to hide the fact that all of the domains were owned by the same entity. Different registrants, different hosts, different code and some even went as far as never accessing the domains from the same IP. The cost of buying the ever-dwindling supply of good domains remaining coupled with the ever-growing cost of maintaining such a network has reduced its use. But it was still going strong. 

PBNs De-indexed News

At that point, there was a lot of talk of Rankbrain and the infamous google AI, as a company we started reading the writing on the wall and came to the conclusion that if we could come up with test cases where we would be able to detect such link schemes then sure enough Google would be able to do so too. So, we retired our PBN but have kept on a small fraction of it for testing and experimental purposes. One section of domains was segregated specifically for the Prod. The Prod was  a method we developed to gauge how effected a new client website is by an algorithmic penalty of some sort. At the time we were quick to realise that some clients, who have had previous SEO companies do the most horrendous type of link building, would never see any improvements until the root cause was fixed. So, to diagnose if such a root cause exists, we “prodded” the site.  In essence, we built links to the site from our segregated PBN. We knew from experimentation data and some educated guesswork, what the reaction to the links should be and gauging the actual reaction and comparing we could quite effectively judge if the site is under a penalty. Some readers would already know that penalties are plenty and act as layers. Some can have a tiny effect while others can be devastating.  Many small penalties (or filters if you prefer) can mount to monumental negative effects on a site’s rankings. Anyway, having that knowledge gave us a huge advantage. We could now effectively show the client, with data, what the issue is and the possible reasons for it. We then dealt with the issue which 9/10 provided very quick results earning the trust of the client who has failed to see such results from any of the previous SEO companies that tried before us.

This brings us to 2018. After this extensive experience with our PBN we can safely announce that the days of the PBN are numbered. We are no longer even able to reliably use our PBN as a prodding method or, if we even wanted to, use them for link building for one good reason. They hardly ever work any more. Yes, you can get some modest gains but only for low competition keywords. I challenge anyone to show me a predominantly PBN linked site that is ranking for anything competitive these days. They just don’t exist. If they do, its normally very short lived. 

As anyone with their ears to the ground would tell you, there has not been any major Google PBN de-indexing or discoveries lately. No high profile PBN network has been taken down, and no waves of users on Black Hat World or other similar forums announcing their rage at the new penalty their target site has achieved due to an over reliance on a PBN. The reason for the silence? Google has decided to do what they always said they will. The are allowing their algorithms to determine the trust of each individual domain and assign the “juice” to the links coming from the site appropriately. 

Examples of such questions we strongly suspect the algorithms are asking of such domains? Especially of domains where fundamental details such as ownership has changed recently.

 

1. Does the site have any meaningful traffic? Most PBNs will have a WP blog slapped on it and lots of poorly written general blog posts with obvious spammy links. Those never receive any real traffic hence the links from it are almost nullified.

2. Does the domain have authority markers? Things such as a confirmed address, contact details, phone number, Terms and Conditions, relevant blog which is updated regularly and an active social media scene are all signals that Google can easily use to determine how much trust it should allocate to links from that domain.

3. Is anyone actually reading the site? Using metrics like bounce rates combines with knowing average length of time spent on page compared to length and pogo sticking around Google could easily verify if the site is offering any value and demote rink juice accordingly.

4. Does the site have GSC or analytics installed? 

I am sure there are more but you get the jist.

Now, all of the above is still fakeable. But just imagine creating, even a modest PBN of say 100 domains. Each one with a different address, phone number. All uniquely hand coded. All have Google search console (from differing IP addresses). All with good readable, genuinely interesting, content that actually pull in traffic. All on their own, completely unique IP address that has to come from a trustworthy host. What a monumental nightmare that is to keep alive.  It literally means that the cost of creating such a network and then the resources needed to maintain it are so out of this world that it just becomes pointless. And here is what Google is good at. Making some tactics pointless. By doing that, they no longer have to spend millions of dollars in searching and exposing such schemes. They don’t need the whoooarrrr publicity. They just need to make the tactic ineffective. Yes, there are some sites out there that can still offer good gains. But we all know they are dying quickly and sooner or later, the dangers of using such sites will render them extinct.

Im sure that as there are still people using Comment spam and reciprocal linking (do you remember that era?) there will still be some using our leftovers in a decade. But in reality, for most real intent and purposes, PBNs are dead. Long live the…[insert next fashionable tactic]