Welcome to WebmasterWorld Guest from 184.108.40.206
So with all of this discussion about what's happening, has anyone heard anything about what G is "trying to accomplish"?
I had heard that they were headed for "all inclusive" search where they wanted to bring images, videos, books and "normal sites" into a single listing. Not to say that I think that's a good thing but is this the beginning of that? Yahoo and MSN surely aren't trying this.
I've never ( 5 years experience ) seen anything like what is showing up. Craigs list entries, books.google.com entries, entries that I can't even see a KW. My site has been Page 1 for 4 years of over 5 million pages and still is on Yahoo #7 with 7 million and MSN #2 with whatever. I can't see that there is a solution for this right now and have decided to ride out the storm ( which I believe started in Feb and got worse in May and just got worse now ) AND this is ONLY happening for single KW, my phrases have not been touched. Even the guy with www.KW.com which used to rank really high ( not a great website but he ranked high ) has joined the rest of us bobbing on the high seas.
There are about 20 really good companies in my industry and about 6 have really good websites. These guys have been my website competition for the past 5 years ( our company is 30 years old ). I know their sites inside and out and not many of them make changes that often and have not made any that I have picked up. I made changes thru the summer trying to combat what I preceived as a simple algo change and managed to stay on page 1 as I watched some of my competition drop to page 2. Now we are ALL bobbing up and down with no reason. Yesterday, I was in 12 different positions ranging from 2 to 46.
Google - I'm getting seasick. Please let this end.
[edited by: tedster at 11:18 pm (utc) on Sep. 16, 2007]
Some of our sites still have some pages no indexed for whatever reason despites a perfect internal navigation. I would say because some proxy hijacking and copyright infrigement where google was again manipulated too easily and still hasn't fixed.
It's almost like Google is unable to crawl certain pages anymore. Or did they just loose a few databases? It really doesn't make sense.
Some of our websites are never updated and never promoted, they have a few very few cheap backlinks and they rank just great. And sites we really work on just drop.
For some reason I don't think that this is voluntary, looks like somebody at the plex screwed up or maybe they have some type of infrastructure problem.
I just can't compute the fact that even some scrappers rank for our own brand names or eventually a competitor places some baclinks on adult sites and we drop almost entirely for the Google index with our squeaky clean websites.
I also found that some sites that were penalized, in my opinion for very very agressive link building are now back and stable. So what makes them cleaner now that sites that have always been clean?
Is there any info coming out of the plex recently?
Even if there is no fresh cache date with a URL, crawling/indexing still takes place with this URL.
I did a experiment and removed a URL which has not been cached since July for a few days. After adding it back it got a fresh cache date. This is proof enough for me.
[edited by: SEOPTI at 5:34 pm (utc) on Sep. 16, 2007]
I experienced a massive surge of traffic on September 6th (around 3k hits per day), then on September 9th and ever since, traffic from Google has almost completely disappeared. When I type in my domain without quotes, it comes up #11 of 13. With quotes, it is #9 of 66. I received a few (3) random hits for irrelevant search terms, but now when I try to search for those terms, my site doesn't show up at all. Sadly, I believe my site is officially dead to Google. All those months of hard work down the tubes....
All of my income was dependent on Google, so to go from 3k hits per day to zero is quite disheartening to say the least. What I find most frustrating is not knowing what I did to cause Google to knock my site into oblivion.
I thought perhaps it was penalized for duplicate content, as I had recently changed my permalinks and the old permalinks did not redirect to the new permalinks (thus creating duplicate posts). I have since disallowed Googlebot from accessing the old permalinks, hoping this might solve the duplicate content issue and any penalty which may have been incurred as a result.
The same time my site plummeted, I started to see some very strange results in the index, so it was immediately obvious that something major was going on. I'll be keeping close tabs on this thread, with the hope of learning more about the recent changes and what may have caused my site to fall out of Google's good graces.
I experienced a massive surge of traffic on September 6th (around 3k hits per day), then on September 9th and ever since, traffic from Google has almost completely disappeared.
All of my income was dependent on Google, so to go from 3k hits per day to zero is quite disheartening to say the least.
You had 3k hits for three days and all of you income was dependent on Google? You might want to consider yourself lucky for the three days you had traffic. Google can knock you out for years.
My site is an old authority site (when searched via the name of the site it shows up on Google with sub-topics) and has been online for about seven years. It is probably one of five of the oldest sites on the topic in question around the same scale.
The SERPs are dominated by a few of the other old sites but mostly by new ones that have cross linked to one another and have almost identical content.
For all the time of its existence, the site has not faced severe unpredicted drops in traffic.
On Friday night it got demolished and traffic levels dropped to about 1/3 of normal levels. Before that point, traffic had been steadily increasing.
After much analysis and deduction, it was concluded that although various changes took place throughout the site, because of the vast number of different variables and sub-pages that were affected, it was eventually found out that the only consistent change that affected ALL the sites that were almost spontaneously dropped from google had their titles changed in a very minor way.
Some of the pages that had their titles changed did NOT drop. Therefore, it was not solely the title change, but the title change and some other variable that was not initiated by me.
After some more analysis, I have concluded what is the most likely cause:
The titles were changed in one pertinent way: They all had the site's name added to them. So they originally looked like "Widgets" but now they had "Widgets by Decius". This was actually done for visitors, and not for SEO reasons.
Regardless, it was discovered that the most popular pages, with the most inbound links that were changed were the ones to be penalized - and a majority of those inbound links did not just have "Widgets" in the title, but also "Decius", the site's name.
Therefore, the title change IS the culprit to the problem, and the problem was the addition of the site's name to these titles tripped some filter that seems to have concluded that I am spamming because my inbound links now match the title of the target too closely.
I have reverted to the old titles without the site's name in them in the hope that they will escape the wrath of this filter. In addition, I have contacted Google in the hope that this can be expediated, since it obviously is not my intention to make myself higher in the rankings for the site name "Decius" (since it is already #1).
Are there any channels or ways that anyone can suggest I quicken the speed at which this penalty is reversed?
>> Searching for "search engines" does return neither Google, MSN nor Yahoo in the first 30 results
This is all happening for a few days. I know something happened to the singular/plurals too a few days ago and it was bad but I didn't know it was that bad.
But the singular/plural screwups are nothing compared to some other results though.
Now Google can fix the examples above manually, but what about the millions of other queries?
It's just wrong.
(to moderators: I think that the example above is ok with the rules)
Edit: removed some other examples that are mostly commercial.
In one area it is just kind of wild - many of the big sites are out and ezinearticles, tags from Amazon, an obscure article directory, etc. are all ranking - at least for today.
I had recently changed my permalinks and the old permalinks did not redirect to the new permalinks (...) I have since disallowed Googlebot from accessing the old permalinks
Not (only) dupe content but:
Old URLs disappeared overnight, new ones emerged by the bulk. No redirects, old URLs disallowed. You basically started a new site. New navigation analysis, new phrase filtering, new URL structure, new everything with inbounds pointing to a.: homepage, b.: now disallowed old URLs.
I kinda see all your new URLs at -950 once they get indexed. If they get indexed at all. If your site was attempting suicide, this is the way to go.
If not, then pull down that disallow, and put up 301 redirects for ALL of the old URLs to the new ones ( and wait out the 1-6 months until your ranking becomes the same ) or if you like extreme sports, pull down the *new* URLs, and restore the old ones.
The latter isn't something I'd advise unless someone needs to play Russian roulette in order to buy food.
Which isn't a healthy way to do sites.
You'll need to diversify your income, even if it comes from AdSense, it could come from more than a single site.
[edited by: Miamacs at 12:17 pm (utc) on Sep. 17, 2007]
During the winnowing process period, results are helter skelter. After the winnowing process has ratcheted into the "off" position results revert to normal.
Here's why I say that: 3 months ago my kw "widget" had 75 million results. Now it is down to 50 million. What happened to the 25 million pages?
My site shifts from between position 5 to position 10 when things are going well. During helter skelter periods it nukes down into outer space ( I cant even find it myself)
During helter skelter periods, the results are mush -- databases with no data, cloakers, joke sites, etc.
BUT after helter skelter, the results are solid - good authoritative content with very llittle spam.
What happened to the 25 million $hit pages?
Let us think outside the box. They have discovered a smart anti-spam tool. During SES in NYC, I asked the SE panel (who included techies from MSN & Google) about why in the heck can not multibillion companies w/ the best and brightest engineers get rid of a teen age black hat cloaker. They said it's an arms race. Maybe they are using human eyeballs for the pages with the highest cost per click pages because those advertisers are angry about fraud clicks.
To those of you who have good content sites, hold fast - there is a good chance that your pages will reappear as fast as they disappeared like mine do every 3 or 4 days,
What are these guys at G doing?
Why should a site move from literally nowhere to 1st page of results and then back?
Not once, not twice but 14 times over an 18-month period.
Guys at Google, you should make your mind up for once and for all: is our site a good one or a bad one?
Another went from mid 20's to #8 and now has disappeared entirley (one that ranked top 10 for the past 5 years).
Meanwhile I am seeing sites that have not ranked since Florida. Oh, and a number of wholly irrelevant blogs, marginally relevant inner pages, and some very small brochure type sites, not to mention the same site appearing under multiple URL's...
Think I will sit back and wait for things to settle down before taking ANY action.
Perhaps to see which pages the Google users like the best (if any)?
[edited by: Jane_Doe at 5:29 am (utc) on Sep. 18, 2007]
- Copy pasted one site's intro text without quotes for the query.
- It was #1, and a few closely related sites followed, let's say... 6 entries.
But then I saw there were over 450 results. And this wasn't an estimate, neither hidden below the "turn off the filter" ... uh I mean "click here to see the omitted results" link. These were prime time, fully indexed, non supplemental pages on approx 450 domains.
Yeah you guessed it.
Chinese spam sites.
This isn't even news but I noticed something I didn't see before.
A good majority of these domains were listed as follows:
fake title whatever blahblah
... description with some of our keyphrases plus a lot of
other scraped stuff all mixed up yet showing for this query...
There's some kind of a character between the domain and cn.
It looks like a dot in Western encoding. But it's not really a dot.
I don't quite get it. Not all domains are listed like that.
[edited by: Miamacs at 10:42 am (utc) on Sep. 18, 2007]
Some keywords show remarkably stable results while others are just let's call...cheap, very cheap.
Chinese spam sites, scrappers, empty blogs or blog pasts (yep empty pages or almost empty for some), quickly setup overseas websites cheaply promoted solely from cheap directories rule the game on some SERPS I watch.
They've replaced some industry leaders and feature along with authority sites.
As someone noticed there is also a lot of url's for some sites not counting those who've setup both sites+blogs who rank both on fairly competitive terms, out of any logic or true popularity or even mid-class type of content if you ask me, most of the time.
It's really annoying to see scrappers, spammers of all kind ranking well for the content they've hijacked from your site (barely scrapped or rewritten) while your own unique humanly written 2-5 years old pages are not cached and apparently not indexed anymore for whatever reason...
There's some kind of a character between the domain and cn.
It looks like a dot in Western encoding. But it's not really a dot.
I don't quite get it. Not all domains are listed like that.
I came across this on Sunday when I entered an involved search string having to do with FreeBSD - EVERY result after the first four was this dot then cn thing. Clicking on one redirected to a parking page, and something tried to install itself on my computer. There are hundreds if not thousands if not tens of thousands of these results in the SERPS right now. I'm surprised there hasn't been a bigger outcry about it. I reported it in GWT with the spam reporting, but who knows how long or even if they actually look at it - I would think this is pretty urgent, but there's really no way to get an urgent request through to Google. So I hope they find it eventually, and not too much malware gets spread in the meantime.
" there’s at least one infrastructure change we wanted to make in order to handle these better. Infrastructure changes take more time than just killing specific spam sites.
I’m happy to hear examples of sites that host malware/viruses. We detect quite a few such sites, but I wouldn’t claim that the detection has 100% recall."
Or a combination of a clever person and... their character encode bug of the decade.
My guess would be that these sites don't exist, and have never existed anywhere else than in the imagination of Google. The data is real, I mean all the fragments shown for these 'domains' are taken from the net. As if it was some kinda information graveyard.
I repeat, these sites don't exist.
This must be a very well coordinated attact directly on their database. Unless someone figured out how to actually feed all this stuff to Google with their own algo translating garbage into "URLs" to store.
No matter how obscure of a search I enter, if it doesn't require TrustRank, but is full of words that could have been found on the net at any given time... these ". cn" pages are in there. And by the hundreds, sometimes thousands. Even if it's a highly unlikely combination... there are several matches.
Then add but a single word that is a typo, and NO results are shown.
All the 169.000 . nc pages vanish from the SERPs.
The domains don't show any information no matter where I check. Just another proof of these URLs not being 'found' by Googlebot, but being either faked with a virus, or fed through some kind of a trick. Perhaps this non-dot dot would give away a clue, but I'm no expert. Go and try this for example [cnnic.net.cn]. The domains don't seem to exist.
And as previously said, while we're calling them "Chinese spam sites", it's it's just that, a name for the phenomenon. These are not Chinese sites. Aren't real domains. The data looks... uh... entirely fictive.
Seems someone hacked Google pretty bad.
This is not spam, this is more sophisticated and serious than that.
Apparently there is always a western lookalike character in these "domain names". If it's not the 'dot with a space after it in a single character', then it's somewhere else, like a number or the letter X or something. Another suspect of these sites not really existing, is that over at Yahoo, their titles aren't there. Many of the same domains show up, but... there are no titles. Perhaps someone found out how to send in a sitemap without a site? ( all pages are half a MB in size. Some other non-fake-.cn-domains are mixed in there, same size, no cahce, perhaps that's another clue ) ... whoa. These still have whois info. Registrar country code: NL... *grin* ...not sure if it's the same gang, but... very similar. Hmm. Oh I see, privacy huh? Privacy for the pirates.
[edited by: Miamacs at 9:36 pm (utc) on Sep. 18, 2007]
I'm becomming ever more convinced that what I'm seeing is an Allinanchor issue. A new site has gone in at #1.
It uses an ASP.NET dynamic menu system. The top level links include one which contains the target 2 word term in anchor and then waiting to be expanded are 3 other links that contain in anchor those two words plus one other.
Googlebot sees these on every single page of the site, the user only sees the sub ones if they expand the top level. In fact if you disable CSS then you see them yourself on every page.
Am I the only one noticing this kind of cr4p?
Now I better understand why MC says "there’s at least one infrastructure change we wanted to make in order to handle these better"
Someone found a way to get all these sites into Google and other SE's and Google doesn't know yet to handle them without a MAJOR infrastructure change.
The pages are all over the place and I am wondering if the hosting even knows they are there, if they even exists on the server.
Someone asked why not more people ar complaining about it and I think the only reason is that they targeted a few big areas, whereof mine is one. Not only are they in there but they are highly ranked also.
If I was Google, I'd remove ALL .cn and . cn domains for now. If anybody wants to read chinese sites or go to a chinese site, go to Google.cn instead..... Maybe this is what it means when Google are talking about "Universal Search"?
It is starting to look like this is a new Google with constant changing of the index for popular 1 word keywords.
I made a post here about it [webmasterworld.com...] but obviously it was too dangersous or something and deleted.
There is a phenomenal amount of junk sites in a search for a specific TV tuner, and the sites seemed to pull the words off the search and generate a page on the fly.
Again as I posted earlier, I am amazed how Google is unfortunately easily manipulated these days. Not good for anyone, hope they fix all these things soon.
Apparently it may take time, when I think of how long it took them with the www, proxy hijacks and 302 issues which in my opinion are still not resolved entirely I worry!