Following the permanent algorithm changes, the issue of thematic link algorithms came up in Google. One of the most common types of thematic link building is creating one’s own mini-sites. It would be unprofitable to build 20 sites, and to receive link juice as from a single site. The questions immediately arise:
- Will several mini-sites on the same IP address have an effect as compared to the sites located on different IP addresses?
- What if we take several IP addresses within the same subnet?
- What is the efficiency if we place the source sites at different IP addresses, which will differ not only in last digits (DDD), but in a subnet as well (ААА.BBB.CCC.DDD), i.e. in “ССС” unique values (the so-called class C)?
The questions do not end there. One can look up the IP addresses, as well as the domains at Whois and see the owner. In other words, if allegedly different IP addresses of class C were bought, they can be unified on the basis of a common attribute, such as a netname.
- Does Google regard the netname of IP addresses (if they are few)?
- Is there a threshold, for example, of 15 source sites with the same netname, after which the links either get filtered out or work in a negative way?
We also might be unaware of some other factors.
At the beginning, the satellite sites are located on different hosts.
- Let’s create for comparison the sites on totally different hosts, where the percentage (%) of coincidence of some attributes is minimal.
There is one more hypothesis: in order to solve the problems with IP addresses of mini-sites, there are various so-called SEO hosts that comprise a set of 10, 20, 50 IP addresses with different class C.
- Is it more effective to have a separate hosting for each mini-site or is it possible to get along with IP addresses of classes С or D of a certain SEO hosting?
- Is it true that Google knows about SEO-hosts and filters out their IP addresses, therefore lowering their effect?
The sites are tested in real time. Their positions are tracked once every week and a half. It is logical that the search engines need time to identify the sites’ network. You have an opportunity to monitor the sites in the long run. The sites of experiment are in English, USA region.