Google now reports manual webspam actions in WMT

[travelin cat]

View manual webspam actions in Webmaster Tools

We strive to keep spam out of our users’ search results. This includes both improving our webspam algorithms as well as taking manual action for violations of our quality guidelines. Many webmasters want to see if their sites are affected by a manual webspam action, so today we’re introducing a new feature that should help. The manual action viewer in Webmaster Tools shows information about actions taken by the manual webspam team that directly affect that site’s ranking in Google’s web search results. To try it out, go to Webmaster Tools and click on the “Manual Actions” link under “Search Traffic."

[googlewebmastercentral.blogspot.com...]
.

[edited by: Robert_Charlton at 8:04 pm (utc) on Aug 8, 2013]
[edit reason] Added title in Google quote [/edit]

[asusplay]

I used a third party tool to create the disavow file and to work on identifying and contacting the sites to get links removed. Still, those links dont appear on any service I check backlinks with...like you say I have a hard way to go, looking at starting over on another site to be honest.

[asusplay]

I meant I used a third party tool as well as WMT,

[Shepherd]

Submitted a disavow and rr for another site last week. Received response today (labor day here in the US) noting the the site still had unnatural links pointing to it. This response gave 2 examples of bad links.
The first example was from a blog. We had disavowed the blog post url but not the domain and somewhere else on the blog the post was listed, so OK, we disavowed the whole site. (This link was one that was built on our behalf)

The second link sample was not a link that we had built. This link was on a professional's website in their "recommended links" section. This is a good and natural link but I think there are a couple of innocent issues which are setting off automated flags for google. First, the section is named "recommended links" and second on the page there is a link to "submit your link request". I have no doubt that those things are throwing a red flag for google BUT if a normal person looks at the page and the site they would know this is not a spamy site (feel free to pm me to see the site). The links on the page are curated by the site owner, on topic, relevant, freely given. The "submit your link request" link goes to their contact us page, not one of those automated link sharing deals.

Anyway, I resubmitted a RR and let google know that we respectfully disagree with their thought that the link provided as an example was a spammy, unnatural link and that we would not be adding that link to our disavow file as it would not be fair to the site that was linking to us.
I'm not going to be expecting a response...

[deeper]

Hm, I don't think the two factors you mentioned are enough to create a spam link.
Anchor? How many links of this sort in total (usually they give typical examples and no single cases)? Reciprocal?

May be you should post this example in the google forum.

[Shepherd]

I don't think the two factors you mentioned are enough to create a spam link.

Knowing that google does everything they can programmatically it may only take a few "flags" to cause a link to be labeled spamy.

[deeper]

Why should a recommended links section be flagged?

Million sites have such pages or sections. Quite natural and common to recommend several other sites at a special place.

[Shepherd]

Why should a recommended links section be flagged?

I don't know that it should. If I were google and on a witch hunt to find unnatural links though one place I think I would look is on any page with "links" in the title and/or the phrase "submit your link" on page.

Fact: a link from a page with those 2 attributes has been labeled as unnatural. The link, along with the other links on that page, couldn't be more natural. Something had to trigger a flag to make the machine think these are unnatural links.

[1script]

I am curious how accurate this "feature" is. I have a site on which all Google traffic literally stopped on August 8th (ironically, the day this thread has been started). It's still indexed by G, so you would think there's definitely going to be a manual penalty. Have you ever heard of an algo change that kills the entire site? Anyway, almost a month since, and still "No manual webspam actions found." Go figure!

[SilviuFera]

Hey,
I received last year a google notification in Webmaster Tools for unnatural links, I tool action and clear some of my backlinks, after I submitted a reconsideration request and received a not so clear message, don't know if the penalty has been lifted or not, because the site has been performing as it was dead. Now when I check in Webmaster tools for a manual action, I can't find any, this means I have no penalty anymore? Thank you.

[Shepherd]

Another example of google reporting a link as unnatural by mistake:

[seroundtable.com...]

[Robert Charlton]

Why should a recommended links section be flagged?

deeper - For years before Penguin, these were considered to be usual suspects, superficially resembling free-for-all directories. For a while Google was graying out Toolbar PageRank on some legit directory pages, and also on pages that looked like poorly curated lists of links to other sites.

Pages that were well organized and edited, with unique descriptions... many of the same quality signals that Google would apply to directories... generally didn't receive this treatment. I saw no evidence that links from these were discounted or the parent site was hurt, and assume it was a generic FUD technique to discourage link buying from such pages.

One very early rumor, never really confirmed IMO, was to avoid the word "Links" in 'recommended links' pages. I doubted this, as many public libraries and the like tended to innocently call them Links pages.

In any event, there were onpage factors that caused these pages to get flagged well before Penguin. I assume that with Penguin, there's also more sophisticated analysis going on... and there's probably a point system of sorts.

I'm thinking that, with Penguin, the combination of several factors, including inbound and outbound linking neighborhoods, would raise enough flags that a page would be considered tainted... and that the statistical model is strong enough that there aren't that many false positives.

[Perren]

Interesting case Shepherd. But what anchor did your link have? Was it an anchor many other sites used?