|Importance of diverse IP addresses|
I am just looking for some opinions here.
If a network of websites used for link development is geing developed would you reccommend hosting all the websites on unique IP addresses if:
- the sites themselves don't link to each other.
- the sites do link out to a common set of target sites.
Or would you think that it would be fine to host them all on the same IP address?
All comments and opinions most welcome!
If the sites have their own backlinks then they're going to be considered separate. However, you do open yourself up to competitors learning what your other sites are. Bing shows all sites hosted on an IP addy.
Just to be clear:
Sites A,B,C,D are all hosted on the same IP address, but have their own independent backlinks, and don't link to each other.
Site E is on a different IP address, has links from A,B,C,D will this be seen as more suspicious as A,B,C,D are all hosted on the same IP address?
Would it be better to host A,B,C,D on different Class C's, would this lower the risk?
Thanks for being clear-er. There's a pattern there, yes. Will Google penalize it? IMO, probably not, especially if Site E has more than four inbound links.
If the links come from sites on the same IP, Google might devalue them.
@martinibuster, when you say Google may not penalize, would you rule out devaluation as well?
There are many SEO companies that have incoming links from different TLD hosted at the same server, and they rank very well on G.
I have no idea if Google would devalue them or not. Would they penalize the site just for those four links, probably not. Let's think this through and try to hopefully gain a deeper understanding. Feel free to answer or add more questions.
How important is it that a set of links is under control of the webmaster and being used to manipulate the rankings of site E?
Do links from a set of sites within the same IP indicate a probability of an attempt at manipulating PageRank?
Are there situations where links to site E from the same IP but different domains are useful links from a family of related sites that happen to all like site E?
Are there situations where links to site E from the same IP but different domains have a probability of being purchased links?
Are there situations where links to site E from the same IP but different domains have a probability of being syndicated content or some other scenario that can be rationalized as not an attempt to manipulate rankings?
Is it possible that statistical analyses of situations where links to site E from the same IP but different domains indicate a probability of negative intent? And if the intent is there will Google decide to give it an automatic deprecation or, in an effort to avoid deprecating quality sites will use that data as part of a larger computation to correlate it with other signals in order to come to a final conclusion of whether the site/page should be considered not the best answer to a query?
I think the last one makes sense. We have seen in the past where tweaks to the algorithm resulted in collateral damage where false positives resulted in useful sites dropping out of the SERPs. As I recall, the engineers then take that data and use it to refine the algorithm so that the quality sites remain. So in my mind, and I may be wrong, but to me it seems reasonable to use the IP data as part of an overall statistical analysis, not just taking one point (IP data) and reaching a conclusion, but taking a range of points and coming to a conclusion. So one data point in itself might not be enough to put a flag on your site and result in deprecation, but it can when combined with other data points that statistically indicate an attempt to manipulate the algorithm.
For all we know, it could result in an automatic deprecation of the power of that link. I don't have information on that. But to express it in the simplest terms, how natural is it? That's usually a fair yardstick when in doubt. ;)