For our main kw some questionable site remains on 1 and 2 since ages with 2 urls that stick like glue.
60% of the top twenty are MFA link collections. All have even a lower PR than the only very few real autority sites for that kw.
The same sites also rule for tons of kw variations covering one of the biggest new industries in the world.
For another generaly important kw one site with nothing but PAID LINKS remains solidly on No. 3 right next to yahoo, myspace, google etc..
Perhaps publishers will have to produce also black hat editions of their sites in order to secure visibilty at Google?
An obviously successful trick seems to be having 40+ internal links to a folder with the name of your most important kw.
Just some observations during April 2007 ...
|An obviously successful trick seems to be having 40+ internal links to a folder with the name of your most important kw. |
I'm not following, I'm not exactly clear on what they're doing there. You mean they've got 40 anchor text links all to the same folder with the keyword in the filepath? In the navigation or what?
internal links to a folder named my keyword <a href="mykeyword/bestmykeyword.htm">
|Today my website disappeared again |
Any of you stabilized his website in the index?
What action did you take?
night707 .. that's exactly what we did last summer which did appear to help, however we still have our peaks and valleys of traffic, but not anywhere as severe as it had been.
links and page names like:
that junky MFA link collection which i analyzed today is linking to more than 40 pages but inside
I'm seeing this junk everywhere:
There is something about these that seems to escape the 950 issue as well, and it isn't links from the localset.
only saw that on one page, however our whole industry is captured globaly by a bunch of smart tricksters who sit nicely on top of Google since ages with MFA link collections and other dubious products.
As said before, they even have a lower PR whilst many long time industry leaders with real contributions and good content are miles behind.
IMO, Google should not only rely on their Algo developers.
Instead big G could easily afford to hire a few people with real eyes and common sense to weed out at least their top pages.
Human eyes normally don't scale very well; for humans to see bad results and explain in biz analyst fashion WHY the results are bad, then the algo is changed to see if it results in better SERPS.
Still though, what we're both seeing on the domain/keyword+secondarykeword.html stuff could be related also to what Martinibuster noticed on the exact match keyword phrasing.
My goal is more to see how this all fits in with the 950 situation, since those sites seem to be a bit more secure...the best I can tell has to do with getting links to various sections of the site versus the just the index, but I'm not certain yet (which is helping the index in several cases). Perhaps it mimics what an authority resource structure might look like with spread links, even to pages that otherwise add no value. Several sites that I watch at the end of the re-ranking are smaller in size, with the logical inbound pointed to the index...hmm.
[edited by: JoeSinkwitz at 6:50 pm (utc) on April 24, 2007]
"The URLs in the SERPs are all URL-only entries and they are all disallowed in robots.txt. Google did a good job in removing more than half a million last year. It's just the last few hundred / last few thousand that are the problem now."
I remember corresponding with you on this issue last year. I recall your mentioning your site in question, at the time, was a forum. I'm not sure if you are referring to the same site in your most recent post.
If you recall, I too had a similar problem, but with a content site. I had some 99,000 phantom pages listed in Google serps because of a technical problem with an open source cms (and I made an "I'm and idiot" mistake with an .htaccess file for about 4 weeks;).
I finally got the 99,000 phantom pages number down to about 8,000 pages listed in the Google serps. I bounced the majority of phantom urls from the Google serps by getting them to return either a 404 or 410 code.
At the time, I also guessed (and that's all it really is) that sooner or later, the urls for my site that were disallowed in robots.txt (because they were duplicates), but still technically available to be in the index, would become problematic.
I found a way for my site to finally return either 404 or 410 codes for the previously disallowed pages from robots.txt. Now I am only seeing urls for my site indexed in the SERPS that actually exist on my site (plus no duplicates and only one url/page).
Based on what you've written here, might I suggest your trying to figure out a way to figure out how to return a 404 or 410 code for the remaining pages that you have blocked in robots.txt?
Slowly but surely that technical approach has worked for my getting my site properly listed (and non-supplemental) in the Google SERPS.
.... I can tell has to do with getting links to various sections of the site versus the just the index, but I'm not certain yet (which is helping the index in several cases)...
We have a bunch of urls on Google No.1 and 2 pages since ages and zero links to any internal pages which even carry the noindex metatag.
Perhaps an outdated strategy, but visitors are prefered to arrive at the main gates where all external link juice is going.
greennature : The "pages" actually exist, but return no useful functionality unless you are logged in. The content seen when logged-in is not something that would ever need to be indexed; in fact it should never be indexed. The correct method would be to add a noindex meta tag on those pages, and while easy to do, it isn't going to happen with that site. It is interesting to see what happens to it.
Other sites that do have the noindex meta tag were already cleaned up many months ago. I can verify that that metod does work.
"The "pages" actually exist, but return no useful functionality unless you are logged in. ...Other sites that do have the noindex meta tag were already cleaned up many months ago. I can verify that that method does work."
hey GD, thanks for that tip...it sounds like you have a good grip on the problem...also thanks for the noindex tip...I only use the tags on my error pages currently, and I believe the tag is useful in many other situations.
Yes, they should be used on any "page" of a forum that returns "you are not logged in" for visitors who have not logged in.
That includes URLs for "start thread", "reply", "report post", "send PM", "list PM's", and so on.
Well we have been replaced by a site that is unreachable. Very funny..
|Well we have been replaced by a site that is unreachable. Very funny... |
Yeah, Google search engeneers ar getting better and better with driving traffic to their adwords clients.
Good search results could drive too much traffic to non adsense sites. Also Google makes 100% this way rather than having to pay some portion to adsense publishers.
well still unreachable ..
URL is in format widget.example.com doubtlessly it has widgets2.example.com widgets3.example.com etc etc
the www.example,com of that site is down too... incidently it seems to be at the same provider as us. Nameresolving is still active for that site.
|Good search results could drive too much traffic to non adsense sites. |
They are, according to the forum members who were gnashing their teeth over Wikipedia's rankings a month or two ago.
"Google favors sites with AdSense ads because they make more money that way."
"Google favors sites without AdSense ads because they make more money that way."
Which (if either) is the reality? The conspiracy theorists can't have it both ways.
WP is the wild card to dillute the reality. ;)
It would be so very easy for Google to weed out at least the most obvious junk on top ranks.
Any user will go for relevant adsense whenever there is obvious junk on top.
Any user will go for good results on top and much less for adsense.
Google makes more money when not having to pay adsense publishers.
Google is not loosing customers with partly poor results and Yahoo, MSN market shares are further shrinking.
Pls. correct me if any of these observations might be wrong from your point of view.
Conspiracy is certainly a wrong word. But i am just not certain that Google dislikes that state of their search quality.
Some months ago I had asked search support, why they don't want to see the regular authority sites on top for a certain popular KW on instead of MFA linkcollections that violate all sorts of their webmaster rules.
There had never been an answere and one site with fully irrelevant content sits nicely one top with two urls since ages.
How come, that paid link collections get a PR of 8 and subsequent sites with nothing but paid links get PR 7 or at least 6.
2.)Site with adsense
3.)Site with adsense
4.)Site with adsense
5.)Site with adsense
6.)Site with adsense
7.)Site with adsense
8.)Site with adsense
9.)Site with adsense
10.)Site with adsense
None of these are necessarily spam.
As an adsense publisher myself it looks more and more as the algo ranks with in adsense sites. We had falling CTR because we improved server speed so that we had about 4 PV per visitor. SO more pageviews but the same people clicked the same amout of ads which led to falling PR.
Our competition now shows server on maintenance, doubtlessly as it crashed with all the traffic (We had the same problem on that spot and built up to three squids and a db server). Slow server means more money as users can't continue on your site and are more likely to click on an ad as your server takes ages to load.
My experience anyway.
Sites with adsense, well if they are very good sites, unique content with adsense its natural they will be on the top. Good publishers work hard!
A good publisher will gain "trust rank" with google that way. Google knows who owns the site, knows everything about the site and there is some trust gained there.
The search term includes "Free"
1.) Site selling questionable software
2.) Same site as on No.1
3.) link collection made for adsense
4.) link collection made for adsense and adbrite
5.) link collection made for adsense
6.) Latest Mozilla browser always crashes
7.) site does not show up
8.) link collection made for adsense
9.) link collection made for adsense
10.) link collection made for adsense
That is what people find when searching for a free popular media service with global relevance.
10 - 20 offers 2 real services that people are looking for and most of these links collections violate G's webmaster rules extensively.
Geez, who seaches for the keyword "free" only? Thats a very vague keyword. Go look at the stats for that keyword on google trends:
It sure is not searched in the US that often at all.
I'm new to search engine promotion and got asked by a friend why his site had not yet been indexed 2 weeks after submitting for inclusion. After looking at the site < domain name removed > and seeing it was totally flash with no inbound links, I thought googlebot may be struggling to find and index it. Is this the case? How long does it generally take to get indexed? Or is googlebot just having a slow crawling month?
Thanks in advance,
<Sorry, no specifics.
See Forum Charter [webmasterworld.com]>
[edited by: tedster at 7:16 pm (utc) on April 25, 2007]
pls read with care term INCLUDES "free ... " certainly there is more :-))
My website is in and out of the SERPS like many others here. It is currently out from about a week ago. However, when I search using uppercase it is back in 5th position. Whats this all about? The number of results is exactly the same as the lowercase search too, 40+ million, the only difference being that my site is missing from the lowercase search.
It is against the rules here to post your website.
Having said that, search engines do not read flash, so there is nothing to index.
Back to Watching,
|search engines do not read flash |
I'm sure Google reads flash, though not sure how well it reads it. Apparently not to well. Always best to have some html outside the flash object unless of course its a gallery or game, whatever.
Thats a common trip-up, i think a lot of webmasters have walked into it when flash sites became popular.
[edited by: walrus at 7:33 pm (utc) on April 25, 2007]
Reading a Flash file and trusting its content for ranking purposes are two different things. There are so many ways to create deceptive content in a Flash file, that I doubt we'll see much on this front for quite a while. Certainly not in the April SERP changes, and probably not in May either.
| This 42 message thread spans 2 pages: 42 (  2 ) > > |