| 4:09 am on Sep 1, 2011 (gmt 0)|
I think it's just an organizational change within WebmasterTools, Donna. At least until other evidence rears its head, that's my take, anyway.
In general, WebmasterTools has been on its own wavelength compared to Google's actual ranking factors. For example, links will be listed even if they are 100% devalued by the ranking algo, nofollowed by the actual page or marked "ignore" in Google's own back end for some reason.
| 4:15 am on Sep 1, 2011 (gmt 0)|
I feel that this is only intended to make the reports show the right numbers. The interlinking between sub domains of hubpages would anyway have been treated the same as internal links by its ranking algo.
What would prevent hupages from transferring (sharing) the sub domain ownership to its writers, yet retain control over it?
| 4:19 am on Sep 1, 2011 (gmt 0)|
But this change at a time when sub domains are reported as a possible solution for panda is interesting.
| 2:04 pm on Sep 1, 2011 (gmt 0)|
I still seeing my own domain show as "links coming from other sites".
| 2:24 pm on Sep 1, 2011 (gmt 0)|
tedster, thanks for being the voice of reason, as usual!
| 3:01 pm on Sep 1, 2011 (gmt 0)|
Tedster is right. We absolutely have to be rational about this. For me, it's only logical that pages on subdomains of your own domain would be considered to be internal links. Because you own the site, it's your site.
But just because it's being handled that way in Google Webmaster Tools does NOT mean that internally, as a part of the Google algorithm, that's the way it's handled.
The guys and gals who work on the Google Webmaster Tools team at Google are different than those who work on the organic algorithm. And I would bet that they don't have access to each others' information. Two completely different departments, as I recall.
| 3:45 pm on Sep 1, 2011 (gmt 0)|
I have noticed that that our subdomains results are now mixed in with the SERP link block for the main site. Interestingly, subdomains SERP listings remain pure, and not showing links from the main site, or from our other subdomains.
For us, this might be an issue. It reduces the number of product relevant site links shown to visitors. I am guessing this will make the searcher do one additional click before arriving at their desired destination.
| 6:55 pm on Sep 1, 2011 (gmt 0)|
Who cares - I dont, and am certain there's no benefit here that will increase my profits. And that's all I care about at the end of it all. This is Google making you feel you're getting some huge favour thing.
- whats the value for these sites in their index?
| 7:54 pm on Sep 1, 2011 (gmt 0)|
A lot of us have websites with mobile subdomains like m.mywidgets.com for mywidgets.com. Does this affect us? If subdomains links are now being internalized are subdomain pages as well? Do we need to be careful about our intra-linking strategy between our subdomain sites and main sites?
| 8:33 pm on Sep 1, 2011 (gmt 0)|
Do there still exist sites where subdomains do not belong to the same person? Does g### know who they are? Like the former virtualave.
| 8:47 pm on Sep 1, 2011 (gmt 0)|
According toe Hubpages CEO Matt Cutts himself suggested the subdomains. it seems to be a good solution for both Google and Hubpages allowing Google to penalize authors that produce junk and reward authors that produce good material.
sooo i'm with tedster on this one
| 9:16 pm on Sep 1, 2011 (gmt 0)|
Guys, the devil's in the details. It ONLY how Google Webmaster Tools is reporting the links, we have no word that this has any effect on rankings.
| 11:43 pm on Sep 1, 2011 (gmt 0)|
They also now report any deep links from another site as a TLD link.
GWT is highly untrustworthy since what you see is tailored to take care of the basics, such as learning about 404 pages, but will NOT give you insight into ranking improvements.
The keyword list is somewhat helpful since it tells you which keyword your site can pass the most value for in links.
| 1:12 am on Sep 2, 2011 (gmt 0)|
|I'm thinking this is in response to the hubpages "panda fix". |
There are no coincidences.
| 8:40 am on Sep 2, 2011 (gmt 0)|
This change is only for the report on Webmaster Tools. It has not changed an algorithm of web search.
I heard this just now from John Mueller of Google on video chat of Google+.
| 12:17 pm on Sep 2, 2011 (gmt 0)|
I see no reason to give more weight to a site that use wildcard subdomains (thousands of pages with product.site.com). This is clearly a kind of spam.
Changing in WMT can be a first step.
| 5:28 pm on Sep 2, 2011 (gmt 0)|
@suzukik can you post the chat transcript or screens or anything to prove that? lots of webmasters trying to figure this out right now.
| 6:10 pm on Sep 2, 2011 (gmt 0)|
I can say, I'm seeing speculative evidence, though no hard proof that this is more than just in WMT... After Panda I started several new domains, and one subdomain to organize posts into genres, and the subdomain just recently blew out of the sandbox, while the other domains are still in it.
| 6:48 pm on Sep 2, 2011 (gmt 0)|
@Dot_Media Sorry I have no transcript or screenshot because he just spoke so to me over the video chat (not text chat).
Here is the video chat I joined.
John Mueller - Google+ - John Mueller hung out with 19 people. [plus.google.com]
| 12:56 pm on Sep 3, 2011 (gmt 0)|
I don't know whether anyone had been following this but I do track 'links to site' regularly.
The no. of external links is now showing new numbers which seem to be more accurate, but this was how it was before Feb 2011. I noticed a sudden massive increase in the number of external links in late Feb and wasn't sure why it was so, then.
it looks like what they had done in late February is to consider 301/302 redirects as external links.
|Google has recently made improvements to how we gather and display link data (for example, link counts now include links redirected using 301 or 302 redirects). As a result, you may see a change in the number of links displayed for your site. |
Source - [google.com...]
I believe that the www to non-www redirects and vice-versa were reflected as external links.So if you do interlink pages on the site, tis would have resulted in an increase in the number of reported external links. But I am not sure why they made this change and how it improved the reports.
This current change seem to have brought back sanity in those external link reports.
|Most people think of example.com and www.example.com as the same site these days, so we’re changing it such that now, if you add either example.com or www.example.com as a site, links from both the www and non-www versions of the domain will be categorized as internal links. |
But should those redirects be treated as internal links?
| 1:04 pm on Sep 3, 2011 (gmt 0)|
Just back from holiday, checked stats and saw a decrease of about 30% of traffic on the domains on which I use subdomains. The subdomain pages contain user generated content, some maybe with spam and some with good content.
So I can confirm, that it isn't just a change in WMT only!
| 1:58 pm on Sep 3, 2011 (gmt 0)|
Correlation is not causality.
| 4:22 pm on Sep 3, 2011 (gmt 0)|
but it could be :)