So where did this positive feedback come from and how large was the sample?

My assumption is that this feedback is user DATA - increases in "long clicks" and such - rather than subjective, verbal feedback which is notorious for being inaccurate. Google runs on data, not surveys.

(Long clicks are clicks on a result where the user doesn't come right back for a click on a different result)

I equate to this as like working on a construction job. This construction job is to build a new skyscraper. Understand that this construction is being done in the middle of an earthquake. Not a 5 minute earthquake, but one that is rather periodic. Sure it occasionally knocks 5 stories down, but the ground will settle for a week. Then the aftershock comes and knocks down 2 stories. Just keep on building. Welcome to Google organic traffic.

BTW, if I could post a link I would. It was showing July traffic numbers and well, seems that a few more people are voting with their mice. As in Yahoo and Bing gaining again. Maybe by the time all 3 players even up Panda will start bringing those promised high quality sites. Easy to discount a griper like me, however I'm also a user of Google who is slowly but surely not getting the results I expect. That has me switching a lot more to Bing.

I'm sure there are a few more thousand webmaster out there who had some decent traffic in those other countries who are wondering what the hell just happened. For those reading this, welcome. Grab a chair and join in. It's all about solutions remember! What are those solutions? I won't say but it doesn't involve chasing a pot of gold at the end of a rainbow which is how I liken solving your Panda situation.

Cheers, hugs and laughs everyone!

Long clicks are clicks on a result where the user doesn't come right back for a click on a different result

Sorta like when they get distracted by a phone call, bathroom break, kids arguing, gone to make a cup of tea, gone to make dinner, opening each result in separate tabs at the same time and maybe finding what you liked in the last one you opened 30 seconds ago? That kind of solid feedback?

The number of long clicks such as you describe would be statistically stable and wash out of the data from one time to another. No, I'm assuming that this is about a lower level of "bounce backs" of the kind that occur in maybe 10 seconds or so - instant "that's not it" judgments. At the scale of data Google collects, statistically significant patterns become easy to extract.

OK tedster I'm seriously trying to work with you on this one but my original question was legitimate but it really just seems like more of a PR statement on their part. I'm sorry but that method of analysis doesn't even get past my common sense scrutiny (not your way of describing it but that they would actually use that).

Tedster's reasoning would be what they would say as no one knows what their user data is and how it was interpreted by them.

Google+ would have the shares from its users and Google is shameless in showing them in their SERPS. It is against all what they say Panda has been introduced for - thin, redundant, useless etc. etc.

Wait for a few days and your original content reposted on google+ will rank and google will safely claim that theirs is better.

But then that is the way it is now and we will have to choose what we want...

Not only has Google cared about long clicks for many years, but so has Bing. It requires data at a mammoth scale to make such information valuable, but the big search engines absolutely have that.

If an algo change creates SERPs that show a higher percentage of long clicks overall compared to the previous algorithm's SERPs, that's one sign than users are having a better experience than they used to.

One of the keys to understanding Google is apprecation for the SCALE involved. This means getting a point of view that goes beyond the niches and keywords that are important just for our websites.

just read this on user satisfaction - [pcmag.com...]

Is anyone seeing anything unusual in Entrance/Landings stats in GA? I am seeing large upticks in entrances starting on 8/11, in specific directories of a large, US oriented site that was hit hard by Panda 1. If it's real, it's the most positive movement we've seen since Panda hit... which makes me wonder about the 'minor' changes for English language sites mentioned at [searchengineland.com...]

But, we're also seeing some quirky increases in visits overall, and corresponding drops in engagement metrics that appear to be associated with the GA session tracking methodology change discussed earlier in the thread. So, perhaps that's behind it...

...and the new entrances appear to be driving up bounce rate in the directories in question.

indyank, that's an interesting article:

Hitwise found that Yahoo had the highest "success rate," with 81.36 percent of its searches resulting in a visit to a Web site. Microsoft's Bing was second with 80.04 percent, while Google came in third with 67.56 percent.

[pcmag.com...]
However, the idea that any visit at all to a website indicates more "satisfaction" is certainly peculiar in my view. Still, that's quite an interesting data spread.

Hitwise found that Yahoo had the highest "success rate," with 81.36 percent of its searches resulting in a visit to a Web site. Microsoft's Bing was second with 80.04 percent, while Google came in third with 67.56 percent.

Well thats US only. You have to admit that with roling Panda out internationally for better or worse: Google at least has someting to role out internationally. If you look at Bing international search results in languages other then english are still of inferior quality and the market shares are accordingly. In some european countries Google has a market share of 97%. If Bing doesn't get it's act together with non english search they won't be able to take some additional market share from Google for some time to come. Europe alone has a population of 850 Million compared to the US with 300 million. But in dozens of languages.

I have two sites in Spanish thay yesterday got hit by Panda. Both are "news blogs". Original content, optimized, not much ads, nice PR, etc... I really don't know how to fix this. Those sites have more than 5000, how can I know what is wrong and never was since about 5 years?

I already posted another thread about this "case" but no response. Any tip that could help a news site (some things to check) would be much appreciated :)

Cheers.

PS: others sites in my networks, in other niches, got a increase of 30% traffic.

@oxydata: are user metrics ( bounce rate, average pages/visit, time on site ) worse for the affected sites than for the unaffected ones?

Oxydada,

Nobody really knows a "trick" for getting out of Panda, alas.

One thought does spring out at me, though it's pretty much random and just sparked by hearing that your site is entertainment-based: Are your images original, or used with permission, duplicated on the web, used without permission, etc? I have this vague idea I read somewhere on Webmasterworld that Google's gotten better at noticing copies of images.

@tranquilito the sites that got better, did it with a 30% increase in traffic.

The sites that got it worst, did it by 30% decrease in traffic. I have 15 sites, and noticed the same stats +30/-30 in every one of them. Weird!

@lapizuli I know there is no trick (yet?), but I know what to look for in other niche sites. News are different.
About the images, yes, they are used with permission - but as everything in news, most are "duplicated" on other websites (same photo of celebrity doing something, same picture with politician doing something else, and on).

The tricky thing in news site, is that sometimes news sound alike week to week, but every article is written originally.

I'm really lost :(

I think "Absolute Unique Visitors" is actually more accurate than "Visits" to see the real impact of panda update

Hope this helps for analyzing ranking and traffic changes. Only the changes right around these dates are likely to be caused by Panda.

Panda 1.0 - Feb 24, 2011
Panda 2.0 - April 11, 2011
Panda 2.1 - May 10, 2011
Panda 2.2 - June 16, 2011
Panda 2.3 - July 23, 2011
Panda 2.4 - August 12, 2011
[webmasterworld.com...]

Something not mentioned is that this comes "mid term" on a 3 week cycle for the first time. I did notice shift in EN site results occuring at the same time.

I think "Absolute Unique Visitors" is actually more accurate than "Visits" to see the real impact of panda update

Heartbreaking. Here I was all excited about a 12% increase in visits this weekend...only to be brought down to earth with an actual 4% drop.

Also note that:

Visits: +11.81%
New Visitors: -14.45%
Bounce Rate: +9.34%
Pageviews: -6.89%
Pages per Visit: -16.72%
Time on Site: -33.90% (!)

Blast you, Google Analytics tweakers!

a reputable international opinion research company to conduct an independent, valid, scientific survey of Google search users

By merely getting someone to pay for it negates the independent aspect. Why upset the customer ? He might not come back and pay again .... best tell him what he wants to hear.

So our website, which initially got hit when Panda rolled out in the UK, had a strange oddity about it. We got Pandalized sitewide, with the exception of our forum. However, when Panda 2.4 rolled out, it didn't affect any pages apart from our forum. I'm now so confused!

What does everyone think about the 5 Panda updates we've seen so far? Everyone been playing catch up just like me? Sick of it?

One of my pandalized sites has seen its traffic reduced ( from 35K+ visits each day to 10K+) in each update. Strangely, we recovered partially in some of the important keywords but not the traffic.

This is definitely a slow Google Doom for me unless thing starts to turn around in Fall season (In US). I'm fortunate that I'm still single, have huge saving from the past golden days and have not bought any overpriced properties (to become debt slave to bank).

The big bad Panda has taken another chunk out of my leg :o(

Well, according to webmaster tools my pandalized site shows a noticeable increase in traffic from the US, other English speaking countries are about the same, and non-English speaking countries show a dramatic drop in traffic.

So the English version of Panda is different? How?

So the English version of Panda is different? How?

Different set of sites
More experienced machine learning

I was afraid that the Panda-Different-Languages release will negatively impact my site, because it was pronounced "bad" by Google on the first Panda update and hasn't recovered since then. But it didn't happened, the traffic didn't drop.

If Panda is now live on Google.de, Google.es, etc, why I don't have 90% drop on these, like I had on Google.com in February? (The site is in English language, but the majority of my visitors come via product-related keywords which are called the same in all languages.)

This is the part 2 of my previous post (with a little bit of speculation):

My thought is that before the "Different Languages" release they rated a newer, cleaned-up version of the site, and concluded that it is "good", so it hasn't been Pandalized in other languages. However, on Google.com I'm almost dead since Feb.

This difference between English and non-Engish Panda impact on my site makes me think that Google hasn't re-run the thing they did on February 24 for English language, so they still don't know that the site is now "good".

makes me think that Google hasn't re-run the thing they did on February 24 for English language

Bad news on that - it has definitely been re-run - on April 11, May 10, June 16 and July 23. In fact, the last two in particular saw previously devalued sites finally rebound.

I have just checked my analytic, big crash in traffic from non-English speaking countries such as Indonesia (-50%), Thailand ( -60%), Romania (-60%) and etc. UK/US/Canada remain the same in term of traffic.