Forum Moderators: Robert Charlton & goodroi
It is totally nonsense for me to worry about TBPR when I was badly hitted from a -950 penalty (look! my PR raised in almost all pages ... but as I said who cares?).
Please all -950ers come here and join this thread to group possible causes.
Here are mine:
1) E-mail to Adsense team about an account creation with domain name
2) Too many adsense boxes
3) Midly Over-optimized pages
4) Too similar titles
5) Some directory links (as almost all my competitors though)
I add that in last months no big changes were done!
Join -950ers power :-)
[edited by: tedster at 9:08 pm (utc) on Feb. 27, 2008]
I am about 4 or 5 days away from making the next set of changes and these changes might well be the ones I do for this week unless a better suggestion comes along.
Also, I forgot to mention that the site has been in 950 for at least 4 months. That's when I first noticed it. I bought it about 3 weeks ago. Nothing on the site has changed or updated for the past year, except for the structural changes I made after buying it.
I might try to completely remove all descriptive anchor text sitewide and replace them with "here" anchors like "Read more here".
kdobson99 - Don't throw the baby out with the bathwater on this one. Cleaning up the site may take some time, and if the site is as weak as you describe it, it may not be a good test.
Essentially the left sitewide column was a bunch of article links with spammy anchor text.
I've been watching this carefully, and I've noted that repeated keywords in a list of links don't seem to be affecting two kinds of sites adversely...
a) sites on which the repeated phrases are below a certain threshold and simply aren't extremely competitive (I posted about this earlier in these threads)...
b) high PageRank "trusted" sites. Even on extremely competitive phrases, repeated anchor text doesn't seem to hurt them.
I would love to hear some confirmation or disagreement of these observations from others. I'm only seeing a limited part of the web.
My guess, with regard to your test site, is that it does need an infusion of good links, and that means some links from "trusted" sources. Google is looking for good content, not simply clever optimization. It's hard to say whether Google has gotten to the point where it can tell the difference, but they're definitely looking for what they call "quality signals," and I think they've been in an ongoing process of re-evaluating inbound links, if not also the pages receiving them.
One thing I would try, is to rename the urls to non descriptive urls....
FWIW, I've seen a -950 on a non-descriptive url and would have considered renaming it to make it more descriptive, if I believed that page filenames had that much influence, which I don't. ;)
The keyword-rich url is not the only cause of 950, but in combination with other factors, it is a sure way to get 950.
A test for down the road after some other less radical changes are undertaken would be to simply replace all interior pages with new pages having the same content but with different url's some descriptive and some non descriptive.
I have always wondered what would happen if you just gutted and removed your entire internal site except for your homepage and started fresh. Surely you would escape the penalty, I would think. I don't want to try it, but there's good chance it's a quick fix for a 950 problem.
The pages were not removed, they were simply renamed, and the old urls were 301d to the new urls - the 301 is critical to make it work.
I think you're on to something. I suspect the first wave of 950 penalties led folks who were hit by it to seek inbound links to boost their sites back into SERPs. (Which has been suggested here.)
That, in turn, led Google to reevaluate incoming link quality, including, of course, footer links. Clearly link exchanges would be expected for those sites trying to get outside help ("950 relief"), and the footer is a popular place to add the links.
So although the footer link target was not the original purpose of the 950; it is the evolution of it.
I've seen the same thing you mentioned. High PR sites still get immunity from spammy anchor text. One PR6 site, for example, is protected. Still full of repetition in navigation links, but still #1 for an extremely competitive keyword.
I've removed all spammy anchor text from one hit PR4 site (over 1,000 pages), and the 950 penalty still remains. All footer links, too, but no immediate recovery.
There are, however, a few slight indications it may recover. One of the hit pages with higher than average internal links is showing some progress. But it's only one.
I remain conservatively optimistic.
We need to remember the dynamics of a penalty. Google looks at our reactions to penalties and can change them. 950 v. 1.0 could now be 950 2.0.
Algorithms are in flux.
p/g
noindex, follow the entire site, except initially main page and site map pages (so all pages drop from the index but get crawled)
leave title and description elements the same, but competely change the text on some pages
on other pages leave body text the same, but changes titles and descriptions
on other pages change everything... titles, description, body text
leave some pages alone (but still noindex-ed)
See what happens.
I think you'll find that is a cause not equalling effect assumption.
More likely, its the page <title> from which the url is usually auto created if using a cms, or by hand if you don;t use one that is the problem.
I'd be suspicious of repeating a phrase over and over in <title> tags
eg
Green widgets guide, comparisons and cleaning instructions
Blue widgets guide, comparisons and cleaning instructions
Yellow widgets guide, comparisons and cleaning instructions
It is the url with a combination of other factors like, the page over-optimized (titles, keyword repetition, H tags, alt tags, etc. ), plus links into the pages with perfect anchors.
Another important factor is, the majority of the pages on the site are going after a certain keyword/phrase or a variation of it.
So, if you have one page like /nice-widgets.htm, you may be ok, but if your site has many pages like /green-widgets.htm, blue-widgets.htm, red-widgets.htm, etc., and they do what I mention above, you are asking for trouble.
Try the above, and let me know if you do not get 950. I'll bet you have a 80% chance of getting 950.
If you have a strong site with plenty of quality links, you may not. This is why, it may be difficult to 950 a competitor that has a site over-optimized, but if it is a new site, it would be too easy.
It's also obvious at this time that "over-optimization" does not cause this penalty alone as 98% of pages "over-optimized" in the way described don't get hit with the penalty.
The point is that adding a simple delay mechanism can render testing various manipulations of on-page or off-page factors quite difficult.
From what I've noted in a few cases it seems that IF this is in play, to date, the timeouts have been very basic in application (e.g. 30 days, 60 days, 180 days, and not like 47 days).
Just some things I've sensed from low double digit examples I've watched or been involved in.
A few days later, during which time the content wasn't changed, it 950'd the page (on the same search for its title).
My site recently had SERP#60 or so for one competitive search phrase. (It used to be #2 before it was 950d last October.) So I stripped all footers which had that search phrase in it, according to the 950 penalty theory here, when I removed all the footers entirely. Now the ranking has tanked even further to #101.
In other news, be careful about default parking pages on new domains. I'd got a new domain (which I'd considered using to replace the 950d site), and did nothing with it. Didn't change the servers, didn't put up any content, not even a landing page.
But the registrar automatically puts up a parking page with links to get typein traffic clicks. You all know about that.
So Google in its great wisdom, without being asked, decided to crawl that page and promptly 950d the "site" on search of its domain name keywords. Don't know if this poisons the domain's history and damages its future potential. Do I build a new site on it or is it already 950d!? (For good.)
The saga continues.
p/g
[edited by: Robert_Charlton at 6:37 pm (utc) on Jan. 25, 2008]
1.5 - 2 years now and for the life of me I can't figure it out. 6 years we have done well before this mess. One day the site disappeared when big daddy rolled out wiped us clean. 1 year later we get back in and 950 from then on out. Patience...there is none left with Google. Never sold links, never link traded, never black hat, never cloaked, never anything. Heck I see so much spam crap in the results much worse than any optimization we have ever done. Same sites that have been there for years. I just don't know what we did THAT bad to get what we got.
We had lost most of our rankings since August, but there were a few that were never dropped. I'm not sure it was the 950 penalty. but it was definitely some sort of penalty for highly relevant terms that we used to rank within the top 5 results for.
Our site navigation included all of our services pages, down to sub-categories. It is a CSS-based drop-down navigation, so all the html links appeared on every page. It was something like a main category of Widgets. With 4 sub-categories that could be something like widgets by shape, widgets by color, widgets by material, widgets by size. Then under those categories there were more widget phrases such as red widgets, blue widgets, green widgets. This was never intended to be spammy, it was completely user-friendly navigation from our perspective. We wanted visitors to be able to go exactly where they wanted from any page on our site. We had this navigation for more than 2 years without any negative effect on our rankings.
So, after reading this thread I decided to modify the navigation as a test. I kept Widgets as the main category. Then kept widgets by color. Then all the pages within that category I changed to just red, green, blue, etc. Basically I removed all the extra widgets used in the navigation as long as the main category already used it. I still think this makes the navigation less clear for users, especially if they hit a sub-page from a search engine. But Google likes it. All my rankings came back to exactly where they were before.
I did not change any page names, page text, or titles or meta tags. I only changed the phrases used in the global navigation structure.
At first (in about 2 days) I saw about 25% of the rakings bounce back & I was able to verify that those were from pages cached after the navigation change was made. I was able to track for the next few days which pages were cached next & I could then predict the new search rankings based on those newly cached pages. The cause & effect was pretty clear. After about 5 days, all the important pages had been re-cached & our rankings all bounced back. I really can't belive this navigation issue was the cause of our ranking penalty since it didn't occur to me to be spammy in any way. I feel like I almost lost my job over it. But at least I finally figured it out.
