Google Updates and SERP Changes - March 2010

[drall]

< continued from [webmasterworld.com...] >

I have been watching all of my competitors which pretty much live off of large amounts of longtail traffic loose overall traffic the past 4-6 weeks and that includes us.

I dont know whats going on but for big sites in my section of the web site page count has been dropping at a constant decay and new content is coming in very very slowly and not nearly enough to counter the drops.

Overall traffic seems down about 8-10% for all of us and those drops are graphing perfectly to the drops in page counts since mid January.

At the same time the crawl graph in WMT and my local stats package show a flatline since mid Jan with no deep crawls and about a tenth of usual crawl activity.

All of this for us and our competitors started mid January
1. dropoff of the googlebot deepcrawl with a flatline stable crawl at a much lower rate.
2. daily reductions in pagecount across a dozen sites by exact same percentages
3. similar gradual traffic declines matching page count falls
4. uber slow new page caching and ranking

When I see this happening to not only us but many pr6-8 sites in my sector it gives me comfort to see that it isnt only us but also concern because it isnt only us.

Just going to keep chugging along, not much you can do but sit back and watch this stuff unfold.

[edited by: tedster at 5:49 pm (utc) on Mar 1, 2010]
[edit reason] split from earlier thread [/edit]

[kd454]

74.125.79.147 looks like what I order at Starbucks - half-caf

I have been seeing "half-caf" as well on mulitiple datacenters, deff not normal serps but not full caff either very interesting.

[tedster]

I understand the idea that "half-caf" is trying to convey, but we are talking about a new infrastructure here. Being "half-caf" is something like being half pregnant.

These "mixed" search results members are noticing may well be a new Caffeine-based data set, created experimentally to address earlier problems.

[trakkerguy]

74.125.95.132 is now showing the same "half-caff" results as 74.125.79.147.

My guess is they are tweaking caffeine to better match the old serps. Only use the term "half-caff" to make clear that they are not the same (for me) as caffeine was for the last 6 months.

[brinked]

I see no half caff results on all these IP's being mentioned here. I either see caffeine or current serps. When you view SERP's based on IP, they are not going to be the same as when you view them at google.com.

use any datacenter watch tool and you will see all them show these so called "half caff" results

[johnnie]

74.125.95.132 has indeed switched over to this 'half caff'. As has 209.85.225.103. Looks like G might be rolling out a new algo.

[trakkerguy]

I see no half caff results on all these IP's

Just because YOU don't see something different in a serp on a particular datacenter doesn't prove that someone else doesn't see a difference for their search terms.

[edited by: trakkerguy at 9:08 pm (utc) on Mar 12, 2010]

[textex]

Still seeing lots of uk and au in the serps we monitor on half caffeine.

[brinked]

Just because YOU don't see something different in a serp on a particular datacenter doesn't prove that someone else doesn't see a difference for their search terms.

Obviously you have not been following anything that has been going on in this discussion.

Lets not all get bent out of shape about these so called half caff results. There are a lot of changes going on at google and any slight change can mean anything. Things are already complicated as it is with caffeine. There are just too many factors to label something. We all obviously are likely to see different results from each other because we are all unique..just like snowflakes.

I track many google IP's and I only see 2 different results, that is Caffeine and non caffeine. There is no indication what so ever that would imply there is another variation. This is not saying there isnt another variation...I am saying I do not see it. And I am sorry if I need more credible evidence other than a few members I never even heard of say otherwise.

[trakkerguy]

Johnie ASKED about difference in two datacenters and I merely said one seemed to be caffeine, and one was slightly different from caffeine, or caffeine with an update, or whatever.

The two datacenters were definately not the same, and definately not regular serps- Thats why he asked! Now those two show the same for me. Sorry if you missed it, but you don't need to imply Johnie and I didn't see a difference just because you "never heard of us".

[gouri]

There does appear to be a version of the SERP that is not the regular SERP and also not what we are used to seeing in Caffeine.

These new results do show similarities to Caffeine.

Maybe this is the new version of Caffeine. I think that is what we have to try to figure out.

What does this new dataset represent?

[arizonadude]

There is no indication what so ever that would imply there is another variation. This is not saying there isnt another variation...I am saying I do not see it. And I am sorry if I need more credible evidence other than a few members I never even heard of say otherwise.

I too have seen 3 variables today.

Yesterdays Whitenight caffeine data set, then regular data and what appears to me to be a set of both regular and mixed.

[Kelowna]

Still seeing lots of uk and au in the serps we monitor on half caffeine.

On 209.85.225.103 and 74.125.95.132 the page says Google English with a link to Go to Google.com below, I think that is what people are saying is half caffeine, these ips do not appear to be used for regular USA based .com results.

I too see uk results mixed in on those ips but they do not look like any type of coffee I was seeing, just different results served up for non USA visitors.

[SEOPTI]

The site: query results are almost identical now for caff. and non-caff.

[brinked]

IMO there is no such thing as half caff. If you can not see these so called half caff results without searching google.com directly through your computer or through a proxy, then half caff does not exist.

Whenever you search google by an IP address you are going to get slightly different results than if you search google.com directly. It has always been this way.

[whitenight]

hehe,

now that we're finally getting into some deeper group analysis of SERPs (after 4 years of this thread - shout out to reseller wherever you are), we're getting into some "labeling" issues.

Only expected.

As i AM trying to pass off the update reins to someone else, please try to agree upon a "good" label for the 3rd variation.
(Unless Shaddows or Cain want to come along and direct traffic.)

But be careful about labeling the 3rd variation, anything other than the "3rd variation" for now. :)

Not quite sure if it's more related to the "regular" SERPs or the Caff. datasets. And "decaf/half-caff" MAY be misleading to some.

Also please re-read the Oct 08 "Halloween" update thread as mentioned by Shaddows to get a clearer idea of how "various variations" of datasets get weaved into each other during updates until the final datasets "sync up" like a solar eclipse.

_{get ur bets in before the windows close, people ;)}

[edited by: whitenight at 11:57 pm (utc) on Mar 12, 2010]

[BillyS]

There are definitely three sets of results out there right now. My guess:

Old Results - something I was used to seeing in the past. These results are getting harder to find.

Caffeine - a set of results I could easily recognize as the test Caffeine data due to some domain spam sites that were in high ranking positions. I started seeing these results more frequently yesterday.

Filtered Caffeine - what people are calling half-caff, these are a completely new set of data that appear to be the Caffeine set with all filters and factors turned on. I saw this set for the first time late yesterday and they are appearing more frequently. I don’t see the domain spam in these results, but they are not the same as Old Results.

Anyone that states - I don't see it, so it cannot exist is completely wrong. It's like saying the ocean doesn't exist because I've never seen one before and we all know that’s nonsense.

[textex]

'Filtered Caffeine - what people are calling half-caff..... '

At the very least there are DEFINITELY some geo filter knobs that need to be turned.

[walkman]

Something is up, I am getting about 20% more traffic. While very happy, I hope the trend continues as this is spread to other DCs. Can I dream :) ?

[Kelowna]

Filtered Caffeine - what people are calling half-caff, these are a completely new set of data that appear to be the Caffeine set with all filters and factors turned on.

I think I see now what people are calling the half-caff results. I just checked a wide range of keywords and the caff results this morning (on 209.85.225.103) do not seem to have the geo-location filter in place as the results have a mix of sites from everywhere. I see lots of UK results included in the caff stuff even when using a proxy out of Texas. Could this be what people are seeing as the watered down half-caff. I do not think that this will be the end product, just more testing with and without filters I guess.

[Tallon]

I'm not seeing any significant traffic changes, but content that has been MIA in the serps since last fall is now FINALLY popping in and out of the serps, though not much of the new stuff (now 1 to 6 months old) is ranking well.

Google has been crawling the site big time (past day or so).

"Something is up, I am getting about 20% more traffic."

I really hoped that I would be seeing the same thing (when 250-ish pages of content would be finally included in the serps). Things aren't over yet so I'm not going to get bummed out at this point, but it's not looking too good for my site (old stuff-still ranking strong, new stuff--included in the serps finally but mostly craptastic rankings).

[Kelowna]

Now I am seeing the caff results in Florida using a proxy.

There are no 1/2 caff in this set... hope it spreads as for me at least the results look sweet!

[edited by: tedster at 4:34 am (utc) on Mar 14, 2010]

[brinked]

I am in florida now in my winter home since january, still have not yet seen caffeine down here. Maybe later in the day. Kelowna, I think you may be the first person to see caffeine on the east coast, would be a good sign that it is rolling out.

[trakkerguy]

I haven't seen caffeine live on google.com here in Calif for the last 2 days - only on the ip datacenters mentioned above.

[Kelowna]

I still see the caff results using a proxy out of Pansacola Fl, as well my logs show visitors from Ohio, Michigan, Texas, Kentucky... so the caff results seem to be scattered... but increasing for sure.

[edited by: tedster at 6:13 pm (utc) on Mar 14, 2010]

[brinked]

interesting...I am in south florida so maybe thats why. Definitley a positive sign though

[brinked]

here is a small sample of what I am seeing right now on my proxies:

wayne PA: caffeine
dallas TX: 1 caffeine and 1 no caffeine.
providence UT: no caffeine
woodstock IL : caffeine
Las Vegas NV : no caffeine
tucson az: caffeine

will not listed california because there are too many proxies, about 60% im seeing caffeine and 40% have no caffeine.

I am monitoring just over 320 proxies, all either have caffeine or no caffeine. There is no half caffeine because it does not exist. Google is only releasing caffeine. Google is not going to release 2 versions of caffeine, you may see different rankings based on personal search history, geo location along with many other filters. Like I said before, unless you can type google.com somewhere and see these half caffeine results...then they dont exist...searching google by IP means nothing. I thought I saw half caffeine for a minute...then realized that proxy was for canada :)

[whitenight]

good stuff, brinked!

thanks for adding your quality analysis. :)

[Pass the Dutchie]

14 days and site cache still not updated!

[Kelowna]

...about 60% im seeing caffeine and 40% have no caffeine.

Thanks brinked,

I was starting to think that I was the only one seeing the changes coming. Are you saying that 60% of all proxies are now showing the caff results? or just 60% California?

[brinked]

Kelowna, I only have 20 california proxies, a little more than half are seeing caffeine on them. Obviously these findings are based just on a small sampling so you can only get a really rough idea. I will post more detailed caffeine findings tomorrow, when hopefully there will be increased activity.

One thing I do find weird is...caffeine on the cali proxies tend to go on and off a lot more frequently than the IL proxies which have remained the most consistent throughout the caffeine watch.