Wikipedia put our site to spam list, Google reacted immediately!

[miozio]

We have a site launched about 3 months ago which was quite stable in Google for some time and i tried to put 3 links to Wikipedia on 3 related pages. The articles there had no information on what our resource actually has. One of the contributors removed those links and i opened a discussion on why he did that. After that he let me repost those links and i did it.

Some 2 hours later the big shot A.B. came in and submitted the site to the spam watch which was picked up my multiple pages on Wikipedia. Google crawled those pages and the next day the site disappeared from Google.

I wrote to A.B. contributor to review his decision but he never replied.

Just to tell you one thing, we have started this site as a review resource actually travelling ourselves and reporting with videos and articles...

I am trying to fix this issue as i am not a spammer and hate spammers myself.

I was wondering if anybody has the same story...

[edited by: tedster at 11:03 pm (utc) on Apr 25, 2010]
[edit reason] removed specific details [/edit]

[internetheaven]

I submited the RR, let's see what the real problem is if they respond.

They won't tell you. The most annoying thing about it really. All you get back is "we've looked at your request" ... and then you sit and wait ... for nothing ...

[miozio]

I guess, what I'm trying to say is that sitewide is the easiest way a blogger can link to other site.

Thank you for your support 1 script. I do think the same way and i developed 15 sites without a problem with sitewide back links.I contacted the sites in question that we did editorial work for and they were also not in favor of such a penalty type. But you never know what G is up to these days.

I have never submitted a RR, it happened a few times that sites disappear for a few days, maximum 2 weeks. But they all came back and remained stable ever since, ofcourse with a little movement up and down...

The thing that worried me now is those big FAT spamwatch pages on wiki shown of the www.domain.com quiery. Now shown from a wiki user's page.

[edited by: miozio at 4:31 pm (utc) on Apr 26, 2010]

[tedster]

do I hear people here saying that blogroll links - the cornerstone of the blogosphere as we know it - a poison?

I think it can be - if you get added to 5 different blogrolls at one time and some of those blogs are already known to Google as selling dofollow advertising. Of course we didn't start out talking about blogroll links, but "5 related very respectable sites" that gave run-of-site links, which sounds even more suspect.

[miozio]

I also see several other sites that copy wiki content with pages as "Wikipedia talk:WikiProject Spam" With our site in the description

[engine]

But you never know what G is up to these days.

Indeed, that is 100% correct, however, experience indicates a pattern which can be explained and shared, as many experienced members have done.

I too would shy away from generating such a volume of links.

What would have been of value would be one good link from that special site you mentioned earlier.

As far as wikipedia is concerned, I've found that the information must be of value to the topic, and if it doesn't look like it's a link for the sake of a link, it's been accepted.

[caribguy]

miozio I feel your pain.

About 18 months ago, I registered for WP and started adding content to a page about a certain geographic area that I am very familiar with. My company maintains a highly regarded informational resource about this area, and I believed that a link to our site would be appropriate (in addition to my other contributions). The link sat happily through nine months of changes. At the same time, I and a handful of others kept improving the WP page in question (enriching the content, removing spam and undoing vandalism).

Until... a WP spam cop landed on the page and started removing "commercial" links. I stupidly undid the revisions without looking who had made them and was immediately written up for vandalism. Got a second opinion from whoever was responsible for the topic project page. Referring to the "letter of the law" rather than its spirit, he or she did not back me up - and so I wished them good luck and a future full of merit badges. It's a discussion that a casual contributor can not win.

About 30 visitors each month will now have to find our resource through other means. It represents a fraction of one percent of our traffic, shrug...

Regarding your drop out of the index, I my guess is that the two are not related and I hope that your reconsideration requests will work out.

[caribguy]

engine: yes, but only until the spam police shows up.

[gouri]

I think this situation can help to analyze if a link from Wikipedia helps in the SERPs?

I know links from Wikipedia are nofollow but I think Google has said that there are times when nofollow links will provide some value in the SERPS.

After your links were put into the spam list, did you notice a drop in the SERP for terms that the page Wikipedia linked to were ranking for?

[caribguy]

Define "some value" - in terms of SERP, absolutely NO effect. In terms of traffic, depends on the topic - in our situation, not worth losing any sleep over. In terms of reinforcing our competitive advantage vs sites that are not listed, possibly. If another resource would take its place in the future, more so.

[TheMadScientist]

Treading carefully here trying not to trip over anyone's ego.

I can't decide between: Ego here? Nah... OR Good thinking! (LOL)

One of the things involved (almost always with G IMO) is patterns, and if the links were 'nofollowed' or not?

1.) If they were nofollowed, then IMO there would not be an issue.

2.) If the links were added to only blog rolls IMO they would be severely discounted if counted at all, even if not nofollowed.

3.) Adding them to a blog roll is certainly IMO a way to automate link additions. Isn't it? Just because someone enters an link in a script and lets it do the work doesn't take away from the fact the addition of the links was automated. IOW: They didn't go and add 1000 links to static HTML pages by hand they used an automated system to add the links.

4.) If the links were naturally picked up on 5 blog rolls because of the quality of the site and information, unless there were only 5 sites in the niche, doesn't it stand to reason the site should have continued to gain links in the same manner from other bloggers over time? IOW: Patterns... If the site was added to 5 blogs in a day or two naturally, IMO, the pattern should continue across more blogs (even if not at the same rate) over time, rather than being added ('noticed') by 5 blogs and then not being added ('noticed') again.

Anyway, IMO blog rolls are automated ways to add links. They AFAIK are usually nofollowed, would be severely discounted if they were 'followed' and IMO if they are natural to a really great quality site the pattern of addition should continue over time, not 'skyrocket' then completely stop.

Hope this makes a bit of sense.

[tedster]

I've made some edits to this thread because many of the comments were getting heated based on incomplete understanding. I'd rather see a cleaner record left behind for "future generations."

Just to help clarify - it IS common to pick up sitewide links when a blogger adds your site to their blogroll. And that is apparently what happened here, not running some spammy link-building script. The question is still how Google's algo would see that when five blogs add one new site to their blogroll in a short time. It could well have triggered a false positive.

I certainly don't rule out the Wikipedia spam report as a factor, and I hope that was part of your reconsideration request. It's definitely worth mentioning, and I wish you a good outcome.

[miozio]

I'd rather see a cleaner record left behind for "future generations."

:) thanks for your time guys

[1script]

@TheMadScientist:
I guess it all comes down to how you define a naturally acquired link. Well, make it "how Google defines a naturally acquired link". Aside from the fact that there is nothing "natural" about HTML links whatsoever, the closest to "natural" I can think of is a SEO-unaware webmaster (a dying breed) goes around the Net, finds your site, thinks:"that's a cool site, why don't I link to it?" and links to it. Since many of the SEO-unaware webmasters are also HTML-agnostic, they would not directly edit the code and use the WP "New Link" feature instead. If they don't use WordPress, they should have something similar to "New Link" feature in their CMS.

In order to have rel="nofollow" added to your links you need to either install a plugin or edit your template but both would mean that you have to know at least something about SEO else you would not worry about rel="nofollow" in the first place. That eliminates you from the "natural" crowd. Again, I'm not talking about people that chose to not wear any clothes but rather those that chose not to worry about what Google thinks (at their own peril, indeed) and as such are not trying to game any silly algorithms.

I can't think of any other "natural" way of "acquiring" links - can you? You always have to ask for the link or at least initiate the process somehow - submit into a directory for example, which makes it progressively less "natural" all the way down to outright buying a link.

Also, if a "normal" webmaster knows a thing or two about SEO, it may actually push them towards demanding a compensation for the link. I just had a conversation going like this: "well, so what if your link has been on our site for two years? We can't keep an eye on every site we link out to; what if you are up to no good anymore and that will in turn harm us? We need you to make it worth our while - $$$$/year will do fine, thank you" (I didn't take the offer - too darn expensive).

Anyways, sorry for the long-winded reply that isn't even on-topic here, I just thought that we ought to be careful with assuming what's "natural". In the end, only randomness is natural but that assumes that there is a chance that 5 x 1000-page sites (blogs, as it happens) may read about you someplace, like you and decide to link to you on the same day, and then nothing else will happen for months - that's randomness for you: everything is possible. How probable - is a different story.

[1script]

I'd rather see a cleaner record left behind for "future generations."

Sorry, tedster, did not see your reply before posting the last one :)

[tedster]

That's fine. Many bloggers are essentially SEO unaware and I can see them adding a new site to their blogroll quite the way your described:

...goes around the Net, finds your site, thinks:"that's a cool site, why don't I link to it?" and links to it.

The idea of a blog was, originally, to allow the creation of websites by the technically unsophisticated. In fact, when social media enters the picture, all kinds of natural linking of the old type becomes potential.

It's just that some of us oldsters are used to all kinds of trickery, we've seen WP used for autogenerated spam networks or blog comment scripts and all that malarkey. and of course, so has Google ;(

[TheMadScientist]

I can't think of any other "natural" way of "acquiring" links

Agreed, and I'm going to try to keep this as short as I can, but will post because I think it could have some bearing on understanding what may have happened, other than the original question, which may not itself singularly be the cause.

In the end, only randomness is natural but that assumes that there is a chance that 5 x 1000-page sites (blogs, as it happens) may read about you someplace, like you and decide to link to you on the same day, and then nothing else will happen for months - that's randomness for you: everything is possible.

This is where I think patterns come into play, because IMO a pattern would give a better 'signal' about what is going on than a 'single event' possibly could, unless the pattern was a single event...

If the pattern is SPIKE -> NOTHING -> NOTHING -> NOTHING IMO it says one thing where if the pattern is SPIKE -> delay -> SPIKE -> delay -> SPIKE over a period of time it would IMO say something else.

The thing about Google is IMO almost everything they do almost has to be pattern based to get better 'signals' related to a single event or series of events and know what to do with those signals more accurately.

IMO There are some cases where a 'single event signal' could be a good indication of something, but there are many times when a 'single event signal' should be evaluated on an 'over time' basis and related to other 'single event signals' to make a better determination of what each 'single event signal' might really be saying.

When looked at singularly the second example above could easily lead to a 'penalty' 3 times, but when looked at over time IMO it looks more like a natural 'discovery pattern' of behavior... A few people see the site... A few people like the site... Links spike... nothing happens for a bit... Some new people visit the site(s) linking and the site itself... A few of them like the site... A few of them link to the site... Links spike... (IMO You could also reasonably expect if the linking and liking was natural and wide-spread the delay between linking and spiking should shorten.)

When looked at singularly IMO the first example could easily lead to 'penalty' and 'over time' a 'sustained penalty' because the pattern is more a pattern of 'odd growth' or 'manipulation of links' over time, so it (IMO) looks more like the link additions were manipulated, whether they were or not... See above to understand why I think the first example would be treated differently over time than the second, because IMO the algo is highly pattern based.

[miozio]

There is one more thing here.... One of our editors submitted 150 new image pages. e.i he took 18 strong separate review pages which had about 8 image thumbnails each, previously linked directly to the image on the server. So he made 8 new pages for each review page, copying the same name, and url ending: example.com/name/page00021/, name/page/00022, etc... The titles of each additional image page resembled the main one with the eddition of "Pic1", "Pic2" These 150 new pages had only one image on them.

When i noticed that, i removed all 150 from the server and from Google index. The webmaster tools removed them in a matter of a few hours...

I did not mention this as i thought, it was a little problem which was fixed anyways...

I now think this might also be the cause.... The site is growing rapidly and its hard to catch up on whats going on :)

[Musicarl]

Miozio-

Can you clarify how put in the Wikipedia links? I'm wondering if you added them as external links at the bottom or wrote some content and added the links as references.

I have a feeling that Google does factor in Wikipedia links, and probably looks at anything they flag as spam as well.

[kevsta]

sorry coming back to sitewides again, I really can't see it being those.

say for example I am a respected web designer and I make a really good WP theme, with my link on the bottom.

and say a few big powerful sites see it and switch to it, so I pick up lots of links very suddenly.

is that a good reason for G to bomb my site?

we get (and lose again) lots of sitewides in various genuine (non-paid) ways and Ive never seen anything negative from them.

only difference is theyre not on sites that sell links, so maybe that's the differentiating factor, otherwise I dont think sitewides hurt you at all.

if anyone wants to test this, ie sitewide links from big sites I'd happily let them point them at my site (as long as I get to choose the anchor text ;)

[TheMadScientist]

is that a good reason for G to bomb my site?

Could be if it was a single event where links spiked and then did not continue to grow, but my guess is that's not the reality of what happened with your site, so IMO you're really comparing apples to oranges. Personally, my 'guess' is links to your site followed a pattern close to what I suggested in my second example above, because...

we get (and lose again) lots of sitewides in various genuine (non-paid) ways...

[youfoundjake]

It sounds like a lot of actions are being performed on the site, both internally, and externally. You may have to give it a little time and see where the dust settles before jumping to conclusions and making changes that may end up harming you in the long run. You might also want to check out the yo-yo effect previously discussed...
[webmasterworld.com...]

Purchased links are generally frowned upon by the major search engines as its indicitive of SERP manipulation. They like natural IBLS and have ways to determine if links are payed for.
Concerning natural IBL's, where you aren't paying for the sitewide link, as long as you don't link back to the site, it should not affect yours negatively.
If it did, then it would be to easy to mess with your competitors rankings by just doing a sitewide link to them from your site.
You may not get full credit for all the links to you, but you shouldn't get dinged for it.
[webmasterworld.com...]

I posted that here over 3 years ago (wow) and I still stand by the concepts held with in it. Reciprocal linking doesn't help, and sitewide links can't harm. (generally).