| 1:05 am on Jan 6, 2007 (gmt 0)|
tedster, re your list of 6
1. Yes, top ten for main term for 3 years and many #1's
2. #9 to #60, not excatly oblivion but as good as.
3. Yes, minor changes although two new sites seemed to have popped out of nowhere.
4. Yes, only one displayed but when you click to see the similar results it's all threads from my message board
5. What is a GWT account?
6. Pr has maintained at 6, although was showing 5 for afew hours last night.
| 1:11 am on Jan 6, 2007 (gmt 0)|
My site fits into the profile ted has developed. I focus on two primary keyword combos. Got tossed from top ten into the 300s the past few days for one but it made a brief appearance in the first 100 today. Other keyword slapped into the 200s but am holding steady at number 5 today. After reading these posts over the past few days, I have searched the far reaches of my site and can't find any that are included in the Supplemental. Is there are trick to searching the Supplemental I should know?
| 1:44 am on Jan 6, 2007 (gmt 0)|
I seem to notice that as I went up from #14 to #2 (for my main key phrase) over the past couple of months, the sites above me got taken out of the serps one by one, starting from the #10 site to the #2 site.
So I'm sitting pretty at #2, hoping not to go up to #1 when it actually happened one week ago. Then like clockwork, I'm gone from a couple of DCs.
Just my observation, but it's like Google is purposely taking out top 10 sites one by one, putting them through a series of filters and seeing if they pass the tests. If yes, then they come back stronger... if not, then kaput. Right now, on certain DCs, I'm kaput.
| 1:48 am on Jan 6, 2007 (gmt 0)|
hehe nice "out of the box" thinking andrew
@pshea: for supplementals search site:www.mysite.com - www
@steve: WMT= webmastertools
@tedster: 4. no, rest = yes
| 1:59 am on Jan 6, 2007 (gmt 0)|
Not hearing to much about the sites that replaced everyone who fell in the serps. Are they spammy? Competitors, etc..?
| 2:08 am on Jan 6, 2007 (gmt 0)|
" Not hearing to much about the sites that replaced everyone who fell in the serps. Are they spammy? Competitors, etc..?"
well I can put to rest one myth...
one of the sites that took one of the spots after we vanished, is 262k worth of duplicated product links...so much for page size eh? they have probably 300 products, and each one is listed seperately in both sidenav, and bottomnav, no category, manufacturer nothing. smells like canned ham to me
| 2:28 am on Jan 6, 2007 (gmt 0)|
Well, it looks like PR may be shifting around right now, so #6 is possibly not such a good indicator, except that no one is being sent to zero.
The thing that is interesting to me is that the site: supplemental thing (#4) is so widespread, and yet a few here see the big plummet in their rankings without that symptom. For those who do not see the raft of supplementals, do you have strong PR, like at least a 5-6 home page? For those who do see the new supplementals above their domain root in site: -- is your PR also strong, or is it more medium to low?
I feel like there's some common factor we are still missing. Is anyone seeing the deep plunge for a pure informational site, or is it only e-commerce? Other ideas?
| 2:35 am on Jan 6, 2007 (gmt 0)|
Tedster you are really running down several blind alleys. 4 and 6 certainly arent accurate to things being effected, using "sites" is mistaken in itself as it is seldom whole sites effected, and what does e-commerce have to do with anything?
| 3:25 am on Jan 6, 2007 (gmt 0)|
|For those who do not see the raft of supplementals, do you have strong PR, like at least a 5-6 home page? |
Yes, PR 5 home page. Site is informational (not e-commerce) but does have moderate advertising. All unique, internally generated content.
| 3:32 am on Jan 6, 2007 (gmt 0)|
No supplementals on PR 5 and 6. Though the 5 is afftcted traffic wise.
| 3:54 am on Jan 6, 2007 (gmt 0)|
Mine is PR6
| 4:16 am on Jan 6, 2007 (gmt 0)|
Almost had a heart-attack this afternoon when I saw our PR6 (for the last 5 years) home page drop to a PR5. It is now back at PR6 for the time being.
| 4:29 am on Jan 6, 2007 (gmt 0)|
There really are at least two different kinds of information we ca go after. The first would be some clear descriptions that precisely describe what happened to the affected sites. The second would be looking at possible causes, but until we are really clear about the "what", it's not likely we can effectively jump to the "why".
So we seem to have ruled out two descriptors - including #4 (the odd site: results) which was widely reported, but turns out not to be exclusively true. Perhaps it is only true for lower PR urls, or perhaps it's a total red herring -- although it sure does seem to show up frequently.
|it is seldom whole sites effected |
That's a good observation, steveb. No one is reporting zero Google traffic, just a major loss of important SERP positions. So our descriptor list shrinks instead of growing:
1. Sites affected used to rank well for many keywords, and often for a long period.
2. Now those rankings have not just dropped a little bit, but effectively were sent to oblivion
3. The rest of the SERP where the site was removed usually shows only minor changes
4. ruled out
5. For those with a GWT account, there is no message that clarifies what happened
6. ruled out
7. Some SERPs still send good traffic to a url on the domain, but over 50% of total Google traffic to the domain is gone.
So this appears to be some kind of keyword related penalty - is that accurate? If so, are these single words, 2-words, more? Highly competitive, mildly competitive?
I don't think it pays to rule out any particular descriptor as a blind alley, until at least one report comes in that reports an exception. I really hope to organize our reports so that something eventually becomes clear. People have suddenly lost significant traffic, not just wobbled a bit.
| 6:23 am on Jan 6, 2007 (gmt 0)|
The idea I think is leading some people astray is the idea this is a new phenomenon. The only thing "new" is the ever increasing variety of symptoms, and most importantly, instead of a new batch of pages effected by a monthly data refresh, now the data refreshes seem to be every two or so days.
Also new, it seems less things are fixed every refresh, whereas before if 10 things got worse 6 got better. Now it seems for evry thing that gets worse, maybe 2 get better.
Some domains lose maybe 90% of their rankings, while a few things seem totally uneffected. Some domains only have a single page, or handful of pages effected, while the rest of the domain does just fine. In both cases, when looking at things a page level on the same domain, the effect is 100% random in terms of what pages are effected versus which ones aren't. In that way there is nothing whatever for a webmaster to "fix", other than sites with weaker designs seem more prone to be hurt harder, so cleaning up design cruft continues to be the best thing you can do even if does little.
| 6:26 am on Jan 6, 2007 (gmt 0)|
We may not be successful in understanding the factors involved, but I highly doubt that randomity enters into it.
| 6:32 am on Jan 6, 2007 (gmt 0)|
The site I mentioned previously that was dropped for a 1 and 2 keyword search was heavily optimized for those keywords.
Is that common for others?
| 6:59 am on Jan 6, 2007 (gmt 0)|
"So this appears to be some kind of keyword related penalty - is that accurate? If so, are these single words, 2-words, more? Highly competitive, mildly competitive?"
In my case, it's a very competitive keyword (cityname) - a highly searched vacation destination. 20 million plus results returned and I used to be in the top ten.
For the same 2 word term - "cityname vacation" I still hold # 1 and #1 for the plural too.
| 7:16 am on Jan 6, 2007 (gmt 0)|
An other symptom of sites with a penalty. They produce no Google alerts.
I watch certain keywords with Google alerts. The main theme of my sites. But regardless how much new pages with this words, it does not show in Google alerts.
| 7:19 am on Jan 6, 2007 (gmt 0)|
"but I highly doubt that randomity enters into it."
That part is the most obvious part of the phenomenon, since it is so easy to see when comparing pages on the same domain. It's really a fool's errand thinking about this if you ignore the randominity.
Once again, the idea that Google is infallible in the one thing effecting a webmaster negatively while accepting it in everything else they do is allowing self-centeredness to delude you.
(Of course it is not ONLY random, most obviously since when example.com/dir1/ is penalized everything under it in the directory structure is penalized too).
| 7:47 am on Jan 6, 2007 (gmt 0)|
My website was penalized on 14th-15th of Dec. It was PR7, 50K-70K daily visitors and had few subdomains. 1 subdomain had lost about 50% of traffic. Other subdomains have lost 99.9% of Google traffic. My income totally crashed.
This website was number 1-10 on hundreds of keywords for few years. Now I can't even find a cache of almost all pages.
I found the other website, that have exactly copy of my website with redirect to my pages. It looks like my competitor tries to ban my website.
By the way, I have non-affected website with similar content with one of my affected subdomains content (not 100% copy). It was not affected.
| 8:32 am on Jan 6, 2007 (gmt 0)|
Random is an impossible thing to achieve with software.
This is not random, I see a very distinct pattern for why my site is being effected. I accept that other factors can kick in and have new effects, but for us we are slowly clawing back our previous rankings and the overall pattern is consistent and changes are having predictable effects.
The supplemental problem is currently only on a www search as opposed to a 'UK only' search where all is healthy for site:search. However, our rankings were and are effected for any search, be it worldwide or just in the UK.
If your site is showing supplementals on top in a worldwide site:search, then your rankings seem to be subject to a dramatic increase in the algo's demands for the pages from your site to qualify a good ranking in the serps. As Matt said, no new algo but imho the current algo needs to be completely satisfied or else you are toast. This is why you generally either see top 10 rankings or 950+. Having said that, you may not trip any part of the algo and achieve position 50 as normal. In our case, the key appears to be a pages relationship with other pages that also qualify for that search. It is related to the way we link to another site. Other sites/pages may trip the algo for other reasons. In all cases, if you are
'supplemental top heavy' then you are subject to an increase in the demands of the current algos.
I think there are several possible reasons why a site firstly becomes 'supplemental top heavy'. These range from clumsy and duplicate metatags, keyword density suspiciously high to doubtful linking strategies suggesting the pages are affiliate landing pages. Googles intention was probably to flag sites that have a suspect profile and subject these to a more critical application of the current algo. In practice, some innocent sites may have been effected..... but if seo webmasters are noticing it.... maybe they got it spot on! In our case, it could have been any one of these plus an unhealthy relationship with another site. If any of these cause you to be 'supplemental top heavy' then you are deemed less worthy and achieving top rankings becomes more difficult.... but not impossible. The dramatic loss in traffic is caused because the site becomes more critically assessed by the same algo as everyone else, but if anyone of the algo's requirements are not met, you suffer more than others.
| 10:40 am on Jan 6, 2007 (gmt 0)|
"7. Some SERPs still send good traffic to a url on the domain, but over 50% of total Google traffic to the domain is gone.
So this appears to be some kind of keyword related penalty - is that accurate? If so, are these single words, 2-words, more? Highly competitive, mildly competitive? "
again, yes spot on Tedster
I still rank for various 2 word searches (or 3 if you include the word "in")
I also own 4 other sites (none as big as the tanked one) and these are all unaffected
as yet anyway
| 10:52 am on Jan 6, 2007 (gmt 0)|
Yes, it is part of the 'local rank' algo which is turned up for supplemental top heavy sites... thus it is keyword/phrase sensitive. I've noticed it happenning on 2 and 3 word phrases.
| 11:01 am on Jan 6, 2007 (gmt 0)|
Yes i too see change in ranking and that too very unbalance.
One of my keywords was ranked 4th yesterday but after reviewing all thought i checked my website ranking and found, it the same page which Google use to show me on mid of december.
For confermation i check my keyword ranking in google datacenter and there its showing the position...
| 12:15 pm on Jan 6, 2007 (gmt 0)|
Of course random is possible with "software". There are even spam sites that have the exact same words on 50 pages, with the exact same linking from the exact same blogs, and some get penalized and others don't. The notion google is infallible is nonsense. They make errors. They do it every single day, and all you have to do is look at the serps to see a result that they obviously don't wish to rank appear at #1 for something.
More obviously, if you have fifty pages that are essentailly structured the same, and three are penalized during data refreshA, then in data refreshB, one is unpenalized while two more are, then in data refreshC, two get unpenalized and one other ones gets penalized, etc etc, it's plain that errors are taking place... EITHER they intend to penalize all consistently, or the pages being penalized are collaterall damage to something else they are trying to do.
And of course the also obvious evidence is the other sites pinned to the bottom of the results around #950... some are similar mistakes on other domains, but that whole group down there is usually extremely spammy with huge numbers of bad links or other blatantly spammy tactics. These aren't the sites that naturally score at around 940, they are put there. In trying to put some pages at #940, they screw up sometimes and put many pages down there they don't intend to.
It's just unproductive to look at "my site" and assume Google woke up one morning and decided to deliberately give you rankings 900 spots different than they have for half a decade... and then two days later put some of the pages back up 900 spots but not some of the others.
"I found the other website, that have exactly copy of my website with redirect to my pages. It looks like my competitor tries to ban my website."
Yeah, sadly this works all too well, one of the clear ways that shows how off base one of Adam's previous statements was. Google is in pure denial about this.
| 12:31 pm on Jan 6, 2007 (gmt 0)|
I have been following this basic topic for months now. My site was originally 'hit' In April 2006.
For most of the posters here, this is a 'new' phenomenon. But for a great many others this is an all too familiar plight. As I read on, I keep thinking to myself that the only real change I see is that it is a new year/month/week/day.
Even the Googlers Matt and Adam claim that there is nothing 'new' going on (no new changes). IF this is indeed true - what on earth could be causing this continuing devistation.
Tedster, thank you for trying to pin-point the exact symptoms folks are currently experiencing. I think it is extremely valuable info. But, AGAIN, I am of the "been there - done that" ilk. ALL of the things you have on your list - I have experienced - but I experienced them MONTHS AGO! (and unfortunately am still trying to recover)
As a 'battle worn' veteran, here are my thoughts...
If this is 'nothing new' then why was I effected last April (June and Sept), while others are just experiencing this now?
I think we need to really go back and examine the significant changes to the algo that Big Daddy ushered in.
One major change that I have noticed is a sudden Major emphasis on PR. For years, site PR seemed to be of little value, as far as ranking and indexing goes. The sudden rise in PRs importance when it comes to INDEXING is very problematic.
No site is an Island! We ALL depend on good quality links. Some (depending on competitiveness of keywords) more-so, than others. My site depended on good quality links from other sites and in-turn, other sites depended on good quality links from my site.
When my site was sent to 'Supplimental Hell', how do you suppose it effected those sites that were dependent on links from my pages? My, once potent out-going links, were now rendered impotent! Sites depending on them were bound to suffer 'some' residual effect.
(I will interject here that my site started out with an original PR5. I was first hit in April. In the next PR update, following my demise, I saw my PR DROP to PR4)
Now take my example above and multiply it exponetially! Apply the same to the sites that are currently being 'hit'. Ask yourself - "How will the out-bound links on the currently effected sites impact those depending on those links, down the line? ...and so-on and so-forth.
WHY do I keep coming back to 'the strength of linking'? Because IF Big Daddy just penalized duplicate content (example) independently - then EVERYONE would have been effected at once!
BUT it seems that this phenomenon has 'rolled' out more like a tsunami following a major quake. Those on lower ground (lower PR) being effected first. But when the 'lower ground' has been washed away, the higher ground will become much more vulnerable.
I think that what we continue to witness is due, at least partially, from a "Big Daddy Tsunami Effect" - when our linking supports are compromised, our sites are then more vulnerable to (example) indexing and duplicate content issues - yada-yada.
IMHO - (Sorry for being so long winded)
| 12:46 pm on Jan 6, 2007 (gmt 0)|
I would like to add here one additional comment/thought...
I Think that the 'new' Big Daddy "LazyBot" (what I like to call it) has compounded the problems we experience.
Google, in their attempt to be 'frugal' with spidering, has made recovering from penalties a difficult and often lengthy process.
I mourn the passing of "freshbot" and "deepbot".
| 1:04 pm on Jan 6, 2007 (gmt 0)|
One more domain recovered on folowing DC's for site: search.
Home page list first and supplementals at the end.
[edited by: shogun_ro at 1:16 pm (utc) on Jan. 6, 2007]
| 1:14 pm on Jan 6, 2007 (gmt 0)|
I have been following this basic topic for months now. My site was originally 'hit' In April 2006.
That is the same time I started seeing this. It only happens to 1 site of mine. The site is 6 years old and had 1000's of top 5 rankings. It still does but just not as many or in the same way as before.
| 2:17 pm on Jan 6, 2007 (gmt 0)|
yes, am out of supplemental on those dc's
site:mysite.co.uk shows 269,000 deliveries
this was the same number in supplemental yesterday
www.mysite.co.uk shows 717,000 indexed on those dc's
it was 1,200,000 yesterday
still supp on my local dc though (on g.com - not on g.co.uk)
and local one shows 1,100,000 indexed on g.com
5th headache of the week coming on.......
| 2:23 pm on Jan 6, 2007 (gmt 0)|
SteveB - >Of course random is possible with "software"
"Anyone who considers arithmetical methods of producing random digits is, of course, in a state of sin." - John von Neumann
Yes google makes mistakes, that is not in doubt. However your suggestion that because some sites get hit with data refreshA and then not in data refreshB is evidence that errors are taking place is not a correct assumption. You are missing many other possibilities including deliberate confusion for webmasters by constantly changing the algo. Only this would be random.... not the software driving it. The concept that errors create random effects is nonsense. There will be a pattern, even if you are the victim of collateral damage and some bizarre site becomes number 1.
The important point is that it is extremely unlikely that a human is introducing a bizarre random element. There is possibly a random tweek done by engineers within established rules but I don't think this issue is anything to do with 'random'. There will be a pattern even if it is deliberate or a mistake.
| This 205 message thread spans 7 pages: < < 205 ( 1  3 4 5 6 7 ) > > |