Is Analytics Data used to identify networks of sites?

[internetheaven]

Hypothetical situation: ;)

If an SEO manages 25-30 widget sales websites (each owned by a completely separate company) and has all 25-30 in his Google Analytics account - will Google mark those 25-30 as a "network of sites" even though their only connection is being in the same Google Analytics account?

Separate question, though related:

Will Google only rank a couple of sites from an "identified network" in the top 10 results for any given search to avoid monopolies?

For the purposes of this discussion can we keep the two questions related, yet answered separately? (If that makes sense.)

Thanks
Mike

[aristotle]

Is there any interlinking between the sites themselves?

[internetheaven]

None.

[aristotle]

As I understand it, in Google's view there is nothing wrong with the same person owning multiple sites provided there is no interlinking between them.

Also, in this case Google might determine that the sites have different owners from the Whois data anyway.

So in my opinion there shouldn't be any problem due to the Analytics account

[rocco]

I rather believe that they count which backlinks get clicked often and use this data to find a better algo for valueing backlinks.
Take this example:
Site A uses GA, and has thousands of backlinks. Even with this tiny data google could get a clou of the value of the linking pages.
The same topic site B does not use GA, but because of similar backlinks google could estimate the amount of traffic resulting from those backlinks and value them accordingly.
And of course - the tighter that net gets, the more accurate will the estimates become.

[leadegroot]

If I were really worried about this, the first step I would take would be to have each domain 'owned' by a different google account, and have this central account that can see them all just be a 'User with Access to Profile' account.
This should help send a message to G that the domains are separate and the 'central' account is a third party.
If there is no cross site linkage as well ( as said above) then it would be a good start - perhaps enough :)

[Reno]

the first step I would take would be to have each domain 'owned' by a different google account

Good advice but what a shame we have to play these silly games just so the Google-Nun does not hit us with the ruler. How much simpler it would be for everyone if Google would do away with any possible penalty for this very common situation, and just assign zero link value to the crosslinking between commonly owned sites, so crosslinking wouldn't hurt you, and it wouldn't help you (in regards to PR). Yet another example of Google's over-emphasis on penalties making things more complicated (and stressful) than should be necessary.

..........................

[CainIV]

Hypothetical situation: ;)

Hi Mike.

will Google mark those 25-30 as a "network of sites" even though their only connection is being in the same Google Analytics account?

Not necessarily, but by offering that data to them you are giving them clues in the case where they want to manually search into further information such as whois, c class and most importantly linking patterns.

An often overlooked segment of this is the actual email used to obtain the account.

If the email is [e-mail], and that email is from one of the domains listed (and the rest are all about the same topic) then it would certainly lead me to believe that you are the owner of all of those, as it is unlikely a competitor would give you access to their reports :)

Will Google only rank a couple of sites from an "identified network" in the top 10 results for any given search to avoid monopolies?

Not necessarily. The risk here is really about patterns. If each website offers a unique product or service from the rest, and the inbound link patterns graph looks different for each website (different inbound links, at different times, different rates) then it is likely that Google would view those domains as valuable for the visitor.

If all / many / some of the domains, on the other hand, are linked, and the inbound link graph is similar for each, that *might* result in a manual look at the domains, or a subsequent automatic adjustment for those types of networks.

My advice?

-Use a gmail for the network accounts.
-Develop unique websites, even when about the same product or service
-Stay away from cross linking.
-Build a unique inbound link graph for each website.

[BradleyT]

Does anyone know the definitive answer as to whether or not Google can see private whois data? I would think not but I've heard both ways.

If they can't then it's probably a tactic you'll want to use for your empire of sites.

[signor_john]

How much simpler it would be for everyone if Google would do away with any possible penalty for this very common situation, and just assign zero link value to the crosslinking between commonly owned sites, so crosslinking wouldn't hurt you, and it wouldn't help you (in regards to PR).

The problem with not having penalties is that there's no incentive for good behavior and no disincentive for bad behavior.

I'd guess that Google is pretty good at distinguishing between natural and unnatural patterns, and that "this very common situation" really isn't so common (or at least isn't a commonplace occurrence in the non-SEO world).

[CainIV]

Does anyone know the definitive answer as to whether or not Google can see private whois data? I would think not but I've heard both ways.

Google is a domain registrar, so in theory they have access to this information, private or not. Private whois only protects the owner against whois queries; the registrars have the unblocked data.

How they use it, however, is open for speculation and debate. Personally I think the advantage of longevity in the index weights more than the risk of cross linking. At the same time, there are others who do it without repercussion.