Identifying a PBN

PBNs are the soup de jour at the moment, with every man and his dog and (his dogs brother) building a PBN and buying expired domains like they're going out of fashion...they're the talk of this fine forum therefore it's logical to assume that they are the talk of the team at Google tasked with fighting black hat SEO, many of whom logically hang out here. They know what we know.

There's been a lot of talk lately about PBNs getting de-indexed with an ever increasing frequency. We all see the rules of building an undetectable PBN touted here often enough and there's plenty of good advice on how to avoid footprints e.g different hosting accounts, faking whois etc

One thing that I don't see mentioned often and probably the one thing that I would use against blackhatters building PBNs is their link pattern.

Google inevitably knows which sites link to others. It seems it would be almost trivial to query their database and find out which websites link to the same websites. Lazy/new/public/oblivious PBN builders may use their entire network of PBNs to link to all of their money sites. With an algorithm that scans sites and determines that the sites linking to it have a link similarity of >5 (it links to 5 identical sites - 5 is an arbitrary number here), this could easily raise a huge red flag for further inspections, via whois or manually.

This is another flag for public blog networks too, say I have a network of 10 sites (small) and 20 people pay me to put links there, it's inevitable that the outward link profiles of these sites start to look the same.

Now I don't know what the safe threshold is for a similar link profile and sure perfectly legitimate sites may occasionally, although I would imagine quite rarely beyond a certain number, have links to the same sites but maybe like anchor text diversity, links to other sites dilute down this similarity but this may be purely speculation.

The endpoint of all this speculation for me is not re-using PBN sites. While I'm not using 1 site to 1 PBN network as I've seen others mention, I'm certainly not overlapping on a regular basis. If I link one site to two money sites, I make sure that no other site on my PBN contains those same two links. Just something to bear in mind which seems too logical to ignore.

I completely agree with you.

Assuming a PBN has been built taking every possible precaution and looks no different to any other money site, the only thing left to identify it is the link pattern.

Google would have worked that out fairly quickly and once they did it would be easy for them to come up with an algorithm that analyses the patterns - after all, they are the best company in the world at doing exactly that.

If that's what they have done it doesn't mean the end of PBNs though - it just means they will take a temporary hit until the innovators assess the damage and adapt.

Trepanated said:

I completely agree with you.

Assuming a PBN has been built taking every possible precaution and looks no different to any other money site, the only thing left to identify it is the link pattern.

Google would have worked that out fairly quickly and once they did it would be easy for them to come up with an algorithm that analyses the patterns - after all, they are the best company in the world at doing exactly that.

If that's what they have done it doesn't mean the end of PBNs though - it just means they will take a temporary hit until the innovators assess the damage and adapt.

Click to expand...

The thing is, 99% of people don't have the money/time/inclination/knowledge to avoid this huge footprint. So if I'm right, it's mission accomplished for Google anyway and you're right, the innovators and adaptors will go on making money like they always have.

I don't see chaos, I see a ladder to climb.

myopic1 said:

While I'm not using 1 site to 1 PBN network as I've seen others mention, I'm certainly not overlapping on a regular basis.

Click to expand...

I'm one of those who follows the principle of building a PBN for a single money site and I do think that it is a good approach to take (although I admit that overlaps of two or possibly three sites isn't necessarily a bad thing) but there is a simple reason people leave massive linking footprints as you say - cost. Most of the people here will try to hedge their bets and have lots of money sites in lots of niches in the hope that one or two of them pay off. Building PBNs dedicated to each money site gets expensive.

Maybe it's a case of people focusing their efforts on one key site rather than spreading themselves (and their PBNs) too thinly. That's probably a bit too much of a white hat thought around here though.

SteveWaller said:

I'm one of those who follows the principle of building a PBN for a single money site and I do think that it is a good approach to take (although I admit that overlaps of two or possibly three sites isn't necessarily a bad thing) but there is a simple reason people leave massive linking footprints as you say - cost. Most of the people here will try to hedge their bets and have lots of money sites in lots of niches in the hope that one or two of them pay off. Building PBNs dedicated to each money site gets expensive.

Maybe it's a case of people focusing their efforts on one key site rather than spreading themselves (and their PBNs) too thinly. That's probably a bit too much of a white hat thought around here though.

Click to expand...

I completely agree, cost is a huge factor in this.

How has your network been and how many sites do you operate? Mine is fairly modest in size compared to numbers I've seen thrown about here...and I've seen no impact whatsoever operating under this principle. Although I'll admit absence of evidence isn't evidence of absence...it's a small sample size.

Thanks for the great suggestion OP. This is how I would catch PBNs if I were assigned this task. There is a definite link pattern for public blog networks. And I think it's easy for g00gle to come up with an algorithm for it.

With that being said though, it is going to increase the cost of PBNs a huge factor and many people would still avoid it. Same reason why people still use cheap hosting though it is a big red flag.

I think the only way to make your PBNs 100% safe is to use this structure.
For this you can't make it but you've to use PBNs from other people (that can have to udestand this graphic and give only 1 link to 1 money site).

The only problem with using other people's networks is you get mixed up in their footprints and taken down as collateral, even if you're only buying one link from them.

myopic1 said:

I completely agree, cost is a huge factor in this.

How has your network been and how many sites do you operate? Mine is fairly modest in size compared to numbers I've seen thrown about here...and I've seen no impact whatsoever operating under this principle. Although I'll admit absence of evidence isn't evidence of absence...it's a small sample size.

Click to expand...

I operate well over 100 sites and have had approximately 12 hit so far. These were clearly hit because of the hosts that I was using for those sites (4 different hosts - no two PBNs for any one money site on each host).

If Google manages to drive 75% of the people away from PBNs with these scare tactics then they will have partly achieved what they set out to do but a) I don't think this will happen and b) the guys who are still in the PBN game are likely to be the hardcore group who continue to successfully do churn and burn, etc.

SteveWaller said:

I operate well over 100 sites and have had approximately 12 hit so far. These were clearly hit because of the hosts that I was using for those sites (4 different hosts - no two PBNs for any one money site on each host).

If Google manages to drive 75% of the people away from PBNs with these scare tactics then they will have partly achieved what they set out to do but a) I don't think this will happen and b) the guys who are still in the PBN game are likely to be the hardcore group who continue to successfully do churn and burn, etc.

Click to expand...

What about the hosts do you think got penalised?

If you mean because they were on cheap hosting alone, I find it hard to believe that cheap hosting is getting de-indexed across the board because there are too many legitimate sites hosted on them, while Google certainly catches innocent sites in their net, I find it too blunt a tactic.

It is all designed to make seo ever more expensive and difficult to predict.

I have read a couple of threads on How to build and maintain a PBN but it would be better if someone would share a method that passes current G algorithms.

If any of you found something like this ,please do share.

Thanks,
Castiel

myopic1 said:

What about the hosts do you think got penalised?

If you mean because they were on cheap hosting alone, I find it hard to believe that cheap hosting is getting de-indexed across the board because there are too many legitimate sites hosted on them, while Google certainly catches innocent sites in their net, I find it too blunt a tactic.

Click to expand...

Ok, I'm not saying that it's purely hosting related in my case but I think that's one starting point they use. I have checked some other sites on the same host (importantly, not just the same IP but the same nameservers too) and plenty are still indexed in Google. But the fact that several of my network sites that link out to different money sites got hit at exactly the same time with only hosting linking them together is not something I can overlook.

So they probably have a set of multiple red flags that they look for with hosting being one of them.

Randomize. Randomize.

Agreed! Tons of money and work to build out a pbn properly but if you can do it, you have a great advantage!