| 6:46 pm on Apr 15, 2010 (gmt 0)|
I see the same too. How can you tell that it's Caffeine in IE though?
| 7:00 pm on Apr 15, 2010 (gmt 0)|
I am looking at the number of competitive sites that are showing up for a keyword search.
| 8:03 pm on Apr 15, 2010 (gmt 0)|
Gouri I was doing a little testing last Monday on google.com.au and found that my results for one randomly chosen term were different when I searched with FF in our home office and IE at our business office.
However several other random terms returned the same search results.
| 8:05 pm on Apr 15, 2010 (gmt 0)|
Sorry Gouri - that last comment was meant to be a reply for visualscope
| 8:56 pm on Apr 15, 2010 (gmt 0)|
Caffeine finally hits the UK. I'll take that as a sign that this is a serious, final roll out! ;) We're always last ...
| 9:31 pm on Apr 15, 2010 (gmt 0)|
Yesterday I saw an alien on youtube, does it mean they will roll out caffeine tomorrow?
But long tail traffic still suffers big time.
| 9:38 pm on Apr 15, 2010 (gmt 0)|
Just (5.40pm) EST tested and no caff on the main google.com in 5 major areas. I even got the google local results based on the proxy location.
Apparently many DCs have it but maybe it's not getting as much rotation ?
Either way, Google revised the schedule to 'weeks' in mid March so maybe, just maybe
| 12:22 pm on Apr 16, 2010 (gmt 0)|
Saw caffeine in the UK last night... today... GONE :-(
I liked it :)
| 12:33 pm on Apr 16, 2010 (gmt 0)|
How do you know it is caffeine that you saw ? is it because you saw a change in your rankings ?
This is what I call a backlinks update but so far no new PR and no backlinks update anywhere that I can see so I am really wondering what everyone sees...
| 6:19 pm on Apr 16, 2010 (gmt 0)|
Wow, things are reeeeally bouncing around here in Toronto the last couple days. Not seen so many major shifts in rankings for ages.
The dials are getting tweaks on Google.ca finally. Get ready fellow Canucks! :-)
| 6:29 pm on Apr 16, 2010 (gmt 0)|
Ready when you are, eh! ; )
| 8:37 pm on Apr 16, 2010 (gmt 0)|
I dont see any bouncing locally in Kelowna... But the USA stuff is all over the place, I am seeing three separate sets of serps using proxies from the USA, could be a fun weekend eh!
| 9:16 pm on Apr 16, 2010 (gmt 0)|
Wow! Caffeine is really awful in the UK!
Someone here has really figured it out ... they own nearly every position from #7 through to #86 all with different domains (www.widgetgreen1.com, www.widgetgreenxyz.co.uk etc.) but practically mirrored content and design! Even all the title tags start with their company name.
Decent terms too in this particular industry ... so, where did I put the phone number for that SEO content writing company that kept bugging me a few years back? ;)
| 8:07 am on Apr 17, 2010 (gmt 0)|
Big changes, including at least some redirects finally being picked up after three months.
| 10:31 am on Apr 17, 2010 (gmt 0)|
Would this be caffine? For my key search term:
My site was ranked for google.com in the US around 25. Now it is ranked 290.
For google.com non-US it's ranked 7 and was previously ranked about 11.
For google.co.uk I am now ranked 700ish and was previously top 10.
For google.com.au I am ranked first and have been for a long time.
Pretty inconsistent. Sound like penalties? Or just algo changes?
| 10:42 am on Apr 17, 2010 (gmt 0)|
From the experience I have everytime google does an update if your site is meant to rank better than it used to be...google moves your site further in the ranking for some time ( it is because it doing its ranking calculations ) and soon you will be back where you use to be if not better.
| 3:17 pm on Apr 17, 2010 (gmt 0)|
I am seeing an unusually large number of sites ranking where they were not before by simply having their keyword exactly in the domain/url. For one really competitive 3 word phrase all top 10 positions had this, very unusual. I am also seeing 80% of the top 10 on another 1 word phrase being wordpress blogs, a major shift for that keyword.
Some really unhealthy results with scraper/clone sites getting 1st page SERP.
Not too thrilled with some of what I am seeing.
| 3:25 pm on Apr 17, 2010 (gmt 0)|
oh here is something I see now (multiple occasions diff sites) that I *never* saw a few weeks back. Ranked 4th for a big 2 word phrase.
This is just one example of "invalid data". Below is the title and description listed in google and sure enough if you go to the site it is down.
Now you may say but google is being nice to not kill their rank, but it has been this way for over a week. Don't you think they can handle this better?
NOTE: From my previous post, it DOES have the 2 word phrase in it's domain and thus ranks nicely for it!
To access IIS Help
Under Construction. The site you are trying to view does not currently have a default page. It may be in the process of being upgraded and configured. ...
| 3:38 pm on Apr 17, 2010 (gmt 0)|
Some people are going to love what they see, and others are going to hate it. That's the way it's always been and will continue to be.
I'm thrilled with what I'm seeing and hope they don't change a thing.
But then again, I was also OK with what I was seeing before, it's just that caf results rank a lot of my sites much higher.
| 6:50 pm on Apr 17, 2010 (gmt 0)|
I agree with AZ, but keep in mind that as far as I can tell, this is not caffeine.
While caff is on most DCs it is not on 'Google.com' in 5 major areas accross US I just tested.
| 7:23 pm on Apr 17, 2010 (gmt 0)|
I'm tracking keywords and phrases that have absolutely nothing to do with my niche to see what's going on elsewhere, so I don't have an emotional interest in those results (love or hate).
More often than not, the results are nuts. On the first page there's sites with only a few pages, low traffic, no links, irrelevant to the keyword, or are in other ways not what I expect for first-page results.
If I do a search for Acme Tractors, there's a site that sells toy tractors. If I do a search for a particular musician, there's a link to some FM radio site that has a sliver of information about the musician.
These sites are coming and going every few hours or even minutes, while the relevant, solid sites that were there consistently a few weeks ago come and go as well.
| 8:16 pm on Apr 17, 2010 (gmt 0)|
I can only confirm a big, big chaos. Sites are vanishing and coming back after 1-2 weeks without any reason - even old sites. Positions are flickering like a bunch of jumping crackers for weeks now. Come on Google, you can do better!
| 11:17 pm on Apr 17, 2010 (gmt 0)|
I think we are seeing an internal war between Google Freshrank and Google Trustrank.
They'll have to pick one or work the dials better, as right now spam gets in to the top if it's fresh spam and irrelevant results get in to the top if they are from a Trusted site ... but fresh.
| 2:51 am on Apr 18, 2010 (gmt 0)|
Looks like they did what they did on the 15th of March again on 16th April last night, many sites going down to 10% of the usual traffic for few days, then recover with better traffic. I am starting to somehow understand the pattern, it takes 3-7 days to roll chunks of data from the old structure to the new one, meanwhile the sites concerned have their SERPs CUT DOWN TO NEXT TO 10%, stand to reason as the old data is not available because of the new structure switch from old to new, and that gap (3-7 days for most) is basically a time out (very low traffic) while the massive database dumps of the new structures are turned on once all is imported.
At least they started it on a weekend this time instead of Monday last month. Well, at least that's what I think is happening, wonder if iSCRIPt has any major shifts, he is one I recall affected by the Ash Cloud of the 15th of March, so was I with many sites, most got better after few days of grounding, however, all are still on the runway, never took off yet, BUT one actually flew to the sky with 40% increase in traffic and that's a massive increase of few thousands uniques daily as it is an old authority and large site. Today, the latter site and most others dipped to 90% of traffic, nowhere to be found on 99% of their keywords, just like last month.
| 12:52 pm on Apr 18, 2010 (gmt 0)|
The Caff data-set I observed before 5 days [4th April to 13th April] for site:domain.com was the most preferable data set.
I guess, half of the (Caff) searcher, G might have turn-off for sometimes, because now some data are clearly missing (I know crawl date and observed cached copy 6 days ago.. now Gone)
Hoping for data stability on caff, after G turn on all the searchers.
| 8:51 pm on Apr 18, 2010 (gmt 0)|
Had a customer email me in a panic as he'd dropped from position 4 on page 1 for his keyword to page 4. I told him to hold tight and he's back to position 4 page 1.
Interesting though is that during the blip the top site on page 1 had virtually no text content on the homepage and no title or description meta tags at all. What it did have going for it was it was a 10 year old domain!
| 9:53 pm on Apr 18, 2010 (gmt 0)|
I'm getting sick of this :(
10 years domain, 1 200 000+ UV/day, stable traffic for 2 years.
Since 15 of March, 20% of traffic has gone, since last week-end We've lost another 10%.
I have no explanation for this, but is getting ugly every day.
I see my # of pages reported by the "site:" command shrinking every day...
| 11:01 pm on Apr 18, 2010 (gmt 0)|
If you've seen a rebound, even a temporary one, consider yourself lucky. None of my sites have recovered to any degree. In fact, one may have been axed again on 04/16, down to approx. half of the remaining 10% of G* traffic. There may be more casualties yet - I'm looking at more sites' logs as we speak.
I see what you are saying about the data dumps timeout but the exact 90% drop figure is too suspicious. Why would you keep elsewhere only 10% of the data belonging to one site. Also, the drop affects a site (as in all or most pages of the same site) yet I think it makes sense to think that they arrange the data per page (individually).
I think it may be the ranking factors' data that's missing, not the actual page data. The remaining ranking pages are probably still ranking based on the on-page factors that have survived with the page's content. Anything that's off-page or too CPU intensive to be processed in real time (hence stored separately) is gone.
Looking at my own sites, I think it can be argued that it's possible to lose chunks of data permanently. Question is, will they ever restore that data or was it too time consuming to collect/create/process in the first place that they'd rather give up on it.
| 12:09 am on Apr 19, 2010 (gmt 0)|
if data is lost googlebot can restore it as it re-crawls.
IMO we still have not seen caff in major google datacenters. I just tested and it's not on google.com in at least 5 major market across US, so hold tight.
| 3:37 am on Apr 19, 2010 (gmt 0)|
I agree with 1script. Something seems to have gone wrong since April 16 (infact it started on 15, another thursday).
This is really frustrating...sometimes I feel that Google has blundered a lot recently and it is really struggling to keep things stable...
| 3:57 am on Apr 19, 2010 (gmt 0)|
I'm seeing it on Google.com in the Phoenix area and have seen it all day today and half the day yesterday.