Every year we hear about how SEO techniques are becoming outdated and no longer working. We also hear about how Google has supposedly caught on to all of the gray hat techniques and how they no longer work anymore. We’ve actually heard this so much within the past year that we decided to put together some case studies on this topic.
The main thing that we hear is that PBNs are dead and that they are no longer a ranking Factor, or that they only bring negative ranking to websites.
We built a few websites that we ranked for both very easy keywords, and very tough keywords, and built some PBNs differently to determine their effectiveness.
The first thing we noticed was that all of our results seem to be quite random. Some of the websites took an immediate hit after receiving links, and some of the web sites saw a substantial boost. We then tried to figure out what might be causing the hits and the boosts and try to isolate some of those factors.
The first thing we saw was that PBNs with low or thin content seem to have a negative effects towards websites. PBNs with spun content on them always negatively affected the websites that they were pointed at.
PBNs with more than a couple thousand words on them, that were using non spun content seem to fare better, though there was still a little bit of discrepancy.
The next thing we noticed was that PBNs that were set up quickly and used to link to our test websites immediately severely hurt our test websites rankings.
We tried using PBNs with high quality content on them and immediately linking out to our dummy test websites to see if it was a quality issue, and still had similar effects though it was not quite as bad.
We figured that this might be a new algorithm Google had rolled out, since these domains we were using for PBNs had been expired and then re registered.
We then decided to make new PBNs and let them sit with quality content on them for 1 week, 2 weeks, 3 weeks, a month, and then two months before linking out to our test websites.
Almost all of the PBNs that had only sat for 1 to 2 weeks negatively affected the rankings of our test websites.
The ones that had sat for 3 weeks seemed to sometimes help and sometimes hurt, but the effects weren’t that great in either direction.
The ones that sat for a month almost always showed a positive impact on our test sites.
The ones that sat for two months gave a positive ranking effect 95% of the time. This seems to suggest that Google has rolled out a new algorithm change which is watching for expired domains immediately linking out to websites.
The next thing we tested for was the nice relevancy of PBNs.
We set up five PBNs with quality niche related content waited 2 months and then linked to our test website which was in the same niche. We made sure that all of these PBN domains were perfectly clean and had great metrics. Each one of these PBNs gave a powerful boost to our test site which was in the same niche.
The next thing we did was set up 5 PBNs with high quality mixed content and mixed niches. We let these sit as well for 2 months, and then linked to our non niche related test sites. We did not notice any negative impact from the links, however the ranking Factor was not quite as strong. If we had to sum it up, we would say it was probably only 20% as effective as the niche related links.
While there are many factors that could have affected the outcomes and all of the situations above, it is still somewhat of a sign that Google is factoring in niche related content and has a filter for expired domains which are linking out and being used as PBNs.
We are definitely preparing to launch a larger scale test and to hopefully isolate more of the variables so that this can be a more accurate study. In the meantime this does shed a little bit of light on why people are saying that PBNs are dead.
We highly recommend that for now, if you’re thinking of using expired domains as PBN links, that you let them sit for a couple of months after setting them up.
This is definitely not meant to be very scientific and elaborate, just some simple tests which we found interesting.
Feel free to comment and let us know your experiences.