| 7:33 pm on Feb 23, 2005 (gmt 0)|
to those who don't rank for domain.com, how do the rest of the pages fare? none of my pages rank. all are indexed with a feb 15th cache too.
"This seems to be more of a problem with filtering out paid links more than anything. What they did in my opinion was place a flag that would go up if a site received too high of a percent of links from the same term. Basically a way of stopping link buying for major keywords."
[edited by: walkman at 7:53 pm (utc) on Feb. 23, 2005]
| 7:45 pm on Feb 23, 2005 (gmt 0)|
It seems to be sitewide. 16,500 pages indexed by Google all are missing from actual results. To summarize searching for my distinct company name on Google yields 21,500 results. 16,500 are listed using site: command. using safesearch filtered results yields 299 results (opps G maybe should be zero?). Visiting last results page and choosing to show supplemental results homepage IS found on page 22 of results. BUT the title and description is from Feb. 2004 "under construction" first page ever title and description! Following the "cached" link shows the page is freshly cached on Feb 22 2005. Page cached yesterday with a year old title and description? Using the site: command returns the current title and description. This points to a bug in Google rather than a filter, penalty etc? I find outdated title and description when using text from outdated description as search string. This confirms the ranking of my site is based on year old data even though cache is fresh. The title and description being shown for homepage matches what would have been expected one year ago. Using the internet archive wayback machine web.archive.org I see an exact match for what google is displaying for "real" "organic" "keyword" searches. using webmaster type commands like site: etc. shows a current title and description corresponding to actual page cache. However since site seems to be invisible it has been very difficult to even find anything in search results other than webmaster type commands. Anyone have any advice? If it is unique to my site what should I do? Write Google? I would like to get feedback from other webmasters with MIA sites and maybe sandbox victims. Check the wayback machine then search on the text you would expect to be displayed by google in results for the old archive of your homepage. Misery loves company so I hope I am not the only one with this issue. Advice or suggestions would be greatly appreciated.
| 8:09 pm on Feb 23, 2005 (gmt 0)|
walkman, I'm in the same boat. Not doing well on domain.com, AND for tons of phrases that we used to do well on, we now do near zip. We are in the SERPs but not high enough to matter. For all practical purposes, we don't get any Google traffic anymore.
All of this occured Feb 3. Occasionally, I see some recovery on domain.com but it's not happening on enough Google servers to matter.
Just checked one phrase I know we used to do well on and we are #112 now. Too bad for anybody who was interested in that subject!
It has a penalty feel to it. Do they still tell you if you have a penalty? I've haunted these forums for ages and remember Googleguy telling people whether they had penalties or not. How do you determine if you've been penalized?
| 8:39 pm on Feb 23, 2005 (gmt 0)|
bears5122: I think you may be onto something.
Its seems to me from reading various threads in WebmasterWorld and elsewhere that part of the Allegra bungling may have to do with the "ODP connection". Someone listed in the ODP, depending on category, may well have hundreds of cloned or semi-cloned directories + spam directories + 'scrapers' linked to their site always using the same exact keyword(s) as the ODP. If you've been in the ODP long enough, you could easily have over a thousand of these links, all identical. If these folks are not trading or buying links for keyword diversity etc, then what they have is a situation as you describe. Compounded for many by the actual title and domain / url, and the same ODP keyword(s) used there. And for those folks who link back to their ODP category, it probably 'looks' even worse to a mis-guided algorithm. The whole thing sorta looks like a giant link farm to a mis-guided algorithm.
I've also noticed for some of these sites in the ODP that dmoz.org doesn't even show up using operators like link: , which is at least a bit strange, since some of the 'clone' directories do.
| 8:52 pm on Feb 23, 2005 (gmt 0)|
Seems like a reasonable conclusion as to "why" the question is did they succeed? Did it make results more or less relevent. Of course that depends on who you ask.
| 9:01 pm on Feb 23, 2005 (gmt 0)|
It's impossible to judge the overall success of Allegra until its "over"; and all the evidence I've seen and still see, is that this is an ongoing, but still broken, update. :)
| 9:22 pm on Feb 23, 2005 (gmt 0)|
Well let's hope when it is all said and done they get it right.
| 9:47 pm on Feb 23, 2005 (gmt 0)|
The ODP/overlinking theory can't be right. All last week I was running into little informational sites that aren't listed in the ODP being creamed by this Allegra thing. The educational site of the organization I work for is listed in the ODP and is still on top of a Google search for its topic. My personal homepage, which *nobody* links to, has dropped off the Google search for my *name*, which nobody else is using or optimizing for and practically nobody even knows how to spell in the first place.
| 10:15 pm on Feb 23, 2005 (gmt 0)|
The examples you give certainly don't disprove any of the conjecture in the various posts above. They demonstrate there are a variety of reasons and circumstances a site may have fallen or disappeared.
Here's another way: Suppose someone links their Website A to their Blog Website B. Suppose they put one link from A to B,and one link from B to A. This should be no big deal, but maybe it is. If the blog were template driven such that a link on the "main page" also shows up on every post and every archive page, suddenly Google would see hundreds or thousands of nice juicy keyword laden links to your website that materialized overnight - a misguided algo could interpret this as 'unnatural' and penalize site A. I don't know if this can happen - Do you? :)
| 11:43 pm on Feb 23, 2005 (gmt 0)|
I guess anything is possible; but the theory doesn't really explain why some sites which are listed in a zillion ODP clones are unshaken in the Google serps while others have fallen off the deep end.
And it doesn't explain why a site would fail to show up for something like a person's own name, particularly when there appears to be no one else on the entire Internet with the same first and last name combination. If the person's name is in the title of the site, then it shouldn't matter how many links are being discounted by Google; the site should still show up #1 for that name, not #87 underneath dozens of telephone directory pages in which the first and last name are separately listed and never in proximity to each other. Unless there is an error in Google, the opposite should *never* happen, even if the person's site has *no* anchor text at all.
And it also doesn't explain why this particular personal page, which as far as I can tell has exactly two links coming into it (neither of them from the ODP or any other directory), should have fallen prey to the same identical thing.
My own pet theory is that Google has completely 'missed' a whole bunch of domains somehow, which are now being indexed ONLY on the strength of their anchor text, or only on their meta tag, or something. So it's coming up #87 because Google doesn't realize that the first and last name are included in the body of the text at all, and is ranking it like a totally unrelated website that happens to have one link with the right text pointing to it or the name in its meta description or something else weak. It must have screwed up spidering some of these domains somehow. To me, that's the only thing that explains why some of these meek little informational sites are getting bushwhacked while other domains are perfectly fine.
| 12:00 am on Feb 24, 2005 (gmt 0)|
I don't know what it is you are disagreeing with other than a 'straw man' you erected about some theory not explaining everything. No-one has made such a claim.
Look at my msg #66 and #68 in this thread - I noted the allintext correlation with low ranking/ dissapearing websites:
"I am seeing a direct correlation for some sites where allinanchor and allinurl are high, but allintext is inexplicably low (on some DCs), as are the resultant google rankings."
| 12:36 am on Feb 24, 2005 (gmt 0)|
"Even with safesearch off (no filtering) I find outdated title and description when using text from outdated description as search string. "
I ran a test on my domain looking for my business name and my site comes up #1 for my business name. However, as several have noted it has an ancient description. However as i also saw posted in another thread--Goggle is apparently using the description from the google directory listing which comes from DMOZ--which in my case hasn't been updated in a long time.
I suggest we all run over to DMOZ and update our descriptions. That should open someone's eyes.
PS. could this problem with business name not showing up be caused from too many people using their keywords and not their official business name in IBL text. I haven't done that except on inside pages and my site is ok.
| 1:05 am on Feb 24, 2005 (gmt 0)|
My site is not listed in DMOZ. But I can trace the outdated title & description at web.archive.org I am not sure DMOZ is related please see if your outdated info corresponds to old archive. web.archive.org aka wayback machine.... gotta love that name LOL
| 1:21 am on Feb 24, 2005 (gmt 0)|
"too many people using their keywords and not their official business name in IBL text"
We do that too, having a contrived word unique registered name... ranking for your own name has always been "duh" a no-brainer... until now. We don't really have a very aggressive outside link scheme as it were and have always relied on superior content to rank well. (I always looked at the whole idea of PR with disdain anyway.)
But, I have linked to the index with our unique name in anchor from some of our own internal pages the past couple days. The pages all did that anyway, but with a graphic that links to index. I just added some natural text links as the company name occurs in some interior pages we already have. We will see IF that helps out.
| 1:29 am on Feb 24, 2005 (gmt 0)|
I think our sites have penalized or put in the "sandbox" for something, that's why we don't rank at all, not even for our name. Is it "un-natural linking", high % of similarity between each page (templates) or who knows...
If you match what Google thinks is bad, you're stripped of everything, that's why link pages rank higher than you. Think of what Google might've tried to do this update.
| 6:40 am on Feb 24, 2005 (gmt 0)|
I have my company name very unique lost too and all I see is that other sites have my name in a query string
this must look like spam to G. Also these sites all seem to run the same script with a bunch of keywords on the right. My company name seems to be in this scripts list of keywords to spam the engines with. I was wondering if anyone had any info on this script
I wish I could show an example of it but Im sure you guys have seem them just a bunch of keywords on the right or left with google adwords on the top and some search results. these sites are deranking my biz and my hard work too many to contact.
| 8:59 am on Feb 24, 2005 (gmt 0)|
It's gotten so bad now, that the home page won't even show a cache link below it.
| 7:27 pm on Feb 27, 2005 (gmt 0)|
The DCs continue to be actively changing. Some of the folks who have been "lost" might give 188.8.131.52 or other DCs a try. At the moment, some previously lost sites are re-appearing.
| 7:58 pm on Feb 27, 2005 (gmt 0)|
Yes, Buguns, one of my lost sites now comes top for it's own unique name on that DC - others are still buried.
| 8:21 pm on Feb 27, 2005 (gmt 0)|
I think that google have just missed a bunch of sites.
I have see too many well known good established sites go missing, including subdomains of msn.com.au.
hi5.ninemsn.com.au. search for this well know australian kids group with its own tv show using any kws you think relevant.This used to be no# 1 or 2 for hifive
| 8:34 pm on Feb 27, 2005 (gmt 0)|
walkman, I don't want to seem like I'm arguing with you on various threads...I'm not.
With regard to your points in your last thread, though, I'd like to point to a site I did for a real estate title insurance company twenty months ago.
The site does not have a links page, and I didn't solicit links for it. There is absolutely NO duplicate content, either between pages on the site, or pages on other sites.
Up until December or so, if you searched for "title insurance xyz county," or other counties mentioned on the home page, the site would come up in the first five results on Google.
Now, if you even search for the company by its unique name, it comes up #351 in the results.
Everything with the site is by the book.
| 11:02 pm on Feb 27, 2005 (gmt 0)|
Same here on 184.108.40.206 we are back at 1 and 2 for a search by our unique registered name. Don't know what that means. It could be a good sign or that dc may just represent some unfiltered serps. Also, we were seeing 17,500 results on that search with all the scrapers and pseudo directories and this dc has just 500+ results for the same search. It would be nice to think G has the capacity to filter spam and yet return a company in a search for it's unique registered name. It's a little late in my book either way. I already removed G from the company Intranet. ;)
| 12:12 am on Feb 28, 2005 (gmt 0)|
Yup 220.127.116.11 and about ten other datacenters are again showing sites that are pagejacked on other dcs.
| 12:33 am on Feb 28, 2005 (gmt 0)|
on my field the people on top either have hidden text (via visibility hidden) or about 4-5 huge paragraphs of nonsense with $keyword text $keyword2 text blah blah keyword on the bottom.
This one is easy to prove, so I'm not just saying because I'm essentially out. I just had the relevant info that people wanted and look what that got me. Time to design sites for google ;)?
On another note: it seems like G realized that it screwed up and a few blog sites are shwoing for their name again. Maybe there's hope.
| 2:38 am on Feb 28, 2005 (gmt 0)|
boy i hope these DC results cross over...
| 5:27 am on Feb 28, 2005 (gmt 0)|
I saw some funny result today.
#1 mycompanybrandname.com (my IT related website)
#2 An adult website, title starts with mycompanybrandname.com - ....
The URL is key1-key2-key3.....keyN.com/key-key-key.html
#last result of page 1 is the same adult website, title starts with mycompanybrandname.com....
| 10:46 am on Feb 28, 2005 (gmt 0)|
Yep - Googles just been broken for the past 3.5 weeks. Lots of companies can't get found for their unique company names in Google. They rank everywhere else - but not in Google.
Nothing we can do about it- except use something else.
After all - if 70 or 80 PHDs can't work out what they broke when they implemented Allegra - then what chance has anyone else got working from the outside?
| 11:31 am on Feb 28, 2005 (gmt 0)|
Our site is now showing up again on:
Anyone else seeing a return on these?
| 11:41 am on Feb 28, 2005 (gmt 0)|
On the first DNS address you gave, I move up several slots from 3rd to 2nd page
for my main keyword, as compared to the results for the default www.google.com.
I don't know what the DNS is for that, I guess it changes. - Larry
| 12:12 pm on Feb 28, 2005 (gmt 0)|
Something is happening, our site is now showing on these also:
| 12:43 pm on Feb 28, 2005 (gmt 0)|
Has anybody noticed if their site turn up when using this Tbar feature [toolbar.google.com]?
| This 192 message thread spans 7 pages: < < 192 ( 1 2 3  5 6 7 ) > > |