New Panda Update - September 27

[BrodyDodes]

Google rolled out its most recent iteration of the Panda algorithm on Wednesday 9/28/11 (or Tuesday according to who you ask). Saw the first recovery of one of my punished sites since the first iteration back in February. Anyone else see recoveries?


[webpronews.com...]

When asked if an iteration of Panda was implemented this week, a Google spokesperson told us, “yes.” She also provided the following statement:

“We’re continuing to iterate on our Panda algorithm as part of our commitment to returning high-quality sites to Google users. This most recent update is one of the roughly 500 changes we make to our ranking algorithms each year.”

If you’ve followed the Google Panda update saga throughout the year, you may recall Dani Horowitz’s story. She runs an IT discussion community called Daniweb, and it was hit hard by the Panda update, but she made a lot of changes, and gradually started to build back some Google cred

[freejung]

about bounce rate, mostly 90-100 on many pages thats be cause they see exactly what they want, look at it or mostly download it at once

I don't know, zeus, I'm just sharing my experience. Maybe it's different in your niche or with your setup. I have a separate HTML page for each image. My category pages have thumbnails, and clicking them leads to the image page with a higher-res version and options for download.

My average pageviews per visitor is around 4-5. People who really like my pics view dozens of pages. In my niche, if you like one of my widget pictures, you'll probably like more of my widget pictures, and you might very well be interested in some doodad pictures as well. Maybe in your niche all they want is one pic and then they leave.

I bet Google is looking at Chrome data which would include file downloads. If your average searcher lands on your site, downloads one file and then wanders off, I expect it would be pretty easy to figure out that means they got what they wanted.

In my setup, if they like what they see they will click on it, which means at least two pageviews per visit, and if they like the first one they are very likely to click "next" and look at another one too.

Edit: incidentally, in case you're wondering, it's not an "adult" site.

[Panthro]

I wrote a while back in some other forum that the less i care about Google the more Google rewards me with free traffic.

I started to nocache than block images and now i blocked about 50% of my site for Google. I tread Google more and more as if it doesnt exist. I take the visitors I get from Google and googlize them, meaning trying to make them stay, leave their email adress or social footprint so they can come again directly without Google.

While my traffic had a steady growth over the last 5 years the share from Google fell from 80% to now down to 40% so i am less dependend on G. Maybe that is one of the new factors from Google, it rewards sites that would work without Google just fine...?

Thank you viggen. This is sage to some degree I think. Quoting so I don't forget it.

[affiliation]

which means at least two pageviews per visit

In the days of Panda how many page views / bounce rate would it be deemed acceptable for a site?

I put a new product search box on one of my sites yesterday, the 24 hour average pageviews per visitor is 3.29 and the bounce rate at the moment seems to be around 36%. What does that sound like compared to other sites?

[zeus]

Hissingsid - I do have a lot of revisits, google even have my domain name in there stupid Google Instant which i hate and can somehow not shut off. About that you dont have that many revisits I would say depends on what you sell

[gehrlekrona]

@asusplay, something certainly happened to me yesterday. I had a couple of days with traffic that went up for a couple of days, but then yesterday it went down approximately 30% and today seems really slow too so something happened but not sure what.

[tedster]

affiliation, I'd never look for hard numbers on metrics like those. They are way too relative to the specific site itself.

I really doubt that "bounce rate" is part of the picture - at least the kind of bounce rate reported by most analytics packages. What any search engine sees as a negative is the "short click" - clicking on the SERP entry and then rapidly going back to the SERP for a different choice.

The healthy sign for entry pages coming from organic search results is a "long click", even if it's only a single page view that your stats package might count as a bounce.

[honestman]

@tedster

Does that mean that those websites (I will not name) and which allow users to quickly cycle through other recommended sites in order to share discoveries are likely to cause a penalty to the site being visited? After all, such social networking sites generally are designed for people with short attention spans...

[Whitey]

After all, such social networking sites generally are designed for people with short attention spans...

Sites i observe seem to demonstrate a lot of "time on site". Different verticals will behave differently, but Google would know the difference between a good user experience and a bad one if all a site is doing is referencing it via a doorway to somewhere else, and/or the information provided is not accurate or compelling, causing a user to signal disatisfaction back to Google ( That's why I've flagged no indexing as a concern on another thread). This is I think is the crux.

This type of thing plays into the navigation heirarchy as well, and I suspect Google pays interest to the funnel, combining user behaviour and "other" signals. We already have some understanding of the semantic relationships within a site and it would be reasonable to assume Google has advanced a long way from those early observations.

So i don't buy the idea that "get folks to what they want, quickly" is a one size fits all approach. It can be great ... but do it badly and it could mess up the quality signals.

[gehrlekrona]

Looks like today is going to be a really BAD day :(
I had a few days with growing numbers but then yesterday it started going down with 30% and today it seems like it going to be another 30%, so not a good day.

[Lenny2]

@gehrlekrona:

Have you been affected by previous iterations of Panda? Specifically have you been slammed down by other versions of Panda only to lose more traffic today/yesterday? Or is this the first time?

[gehrlekrona]

@lenny2, been hit by all of them I think. Wasn't expecting anything now again, but who knows anymore.
What I don't understand is that i usually have the same kewords repeating over and over with little iterations, but when it's bad, then there goes minutes without the usual keywords, actually without any hits at all, whic is REALLY unusual.
If people are like me, then they are lazy, so when I search in Google, i, a lot of time, choose one of the suggestions. Can it be that Google change the suggestionsn all the time and that's what makes a site be hit by Panda?

[tedster]

honestman, I can't imagine that those kinds of sharing sites cause penalties. It's the overall pattern of traffic that people are assuming (it isn't proven) that Google measures, and those sites would not contribute all that much traffic to any one site.

[courier]

The healthy sign for entry pages coming from organic search results is a "long click",

Does this mean we have to slow our search results down, or make it more dificult to find what the user is looking for in order to have users spend more time on our sites? I suppose this is what is being reported that Google has done, making search results not as accurate.

I know speaking about myself, if I am making a purchase or finding information the quicker I can do this the better. Even here, how many people read entire posts? Google is not making money if you are not using their site, so perhaps a short click could benefit.

[Hissingsid]

"Long Click"

Perhaps that is why video sites are boosted.

[zeus]

I must say the traffic I have lost since April is mostly English speaking version of google, like google.com, when I check ranking on .it it looks good.

[suggy]

Courier - no, it means you need to add more value and esnure you only target the most relevant phrases with your SEO.

Bounce rate on its own is not relevant. Time on page/ site on its own is not relevant. Both are only relevant in the context of whether people return to Google to "try again" by clicking another search result or perform a similar search.

So, you absolutely can satisfy visitors quickly, so long as you satisfy them! Otherwise, you need to keep them busy for a bit as a minimum.

[netmeg]

My best performing, best ranked site with the most traffic has a HUGE bounce rate on the home page, because there's a great big box with the most important information that most people want to know. It's been there for over ten years. Hasn't hurt me in the least. YMMV.

[suggy]

@netmeg

Your example is absolutely consistent with my observation; time on site and bounce are only relevant in the context of what happens next.

[Hissingsid]

Bounce rate on its own is not relevant. Time on page/ site on its own is not relevant. Both are only relevant in the context of whether people return to Google to "try again" by clicking another search result or perform a similar search.

Is this a fact or just a hypothesis expressed with great confidence?

[suggy]

Is a fact that I heard from a Google employee? No.

Is it an inescapable fact in the context of searcher behaviour? Absolutely.

We don't need someone to confirm that 2+2=4 and likewise we don't need Google to confirm that bounce rates and time on site analysed out of the clickstream don't say anything about user satisfaction. That's reality.

Now, on the basis that Google hires the smartest PhDs in the world (allegedly), it shouldn't take them long to figure that out.

[courier]

Both are only relevant in the context of whether people return to Google to "try again" by clicking another search result or perform a similar search.

Yes I agree, but if I am shopping for widgets and click on a site only to find out the widget costs $$$, am I not going shop around to find the cheapest? immediately I am going to check 2, 3 or perhaps 5 other sites to see what is the cheapest, which will likely be seen as a high bounce rate. I probably will not read any content, the quicker I make the purchase the better. I am sure I am not the only person who shops like this.

Like netmeg I have a huge bounce rate, on my best earning site the majority of visitors are away within 5 seconds, many may come back later or over the next few days for more prices before either buying from me or another site.

[walkman]

The long click is an old trick but there's something else added here. For almost all sites Google does not have the data to analyze for each keyword along with its click data, and even if they do any SE referrals can easily ruin your profile by sending you better /worse traffic. You may have an amazing site but if Google sent people looking for something you don't have they will go back. If you sell Nike shoes and Google sends you people looking for 'Nike shoes' you will do way better than another site that gets most of its traffic via 'Nike Shoes size 13 black color' type referrals. Who's fault is it?

If you read the threads about Mayday [webmasterworld.com...] you will find a sort of mini-Panda, many sites lost their long tail and 'better sites' (authority, brand) got it. IMO, Panda is the next step, after Vince and Mayday, this time to take away even more or all long tail.

Update on 5/30/10: Matt Cutts from Google has posted a YouTube video about the change. In it, he says “it’s an algorithmic change that changes how we assess which sites are the best match for long tail queries.” He recommends that a site owner who is impacted evaluate the quality of the site and if the site really is the most relevant match for the impacted queries, what “great content” could be added, determine if the the site is considered an “authority”, and ensure that the page does more than simply match the keywords in the query and is relevant and useful for that query. [searchengineland.com...]

It's possible that bounce rate and time on site is not used but correlates with whatever Google uses.

[suggy]

Courier - your example is more weight to my argument. You're just demonstrating further how irrelevant bounce rate is alone and how difficult the context is to untangle! Maybe google has a long memory and notes your subsequent return or not.

Anecdotally, I would say IMDB and Wikipedia also prove my argument. I want specific information, I get it at these sites, I don't need to search for the same info again. I am satisfied and that's why both are doing very nicely in the SERPs thank you, despite what would naively appear a high bounce rate.

[tedster]

Is this a fact or just a hypothesis expressed with great confidence?

It is a fact that several Googlers have reinforced over the years, including Matt Cutts, JohnMu and others. The conventional bounce rate metric is a very noisy signal, too noisy to actually use in ranking.

That hasn't stopped webmasters from thinking they caught a pass and off they go, running down the field. The bounce rate chatter has been extreme over the years, and Panda seems to have brought it back.

It's possible that bounce rate and time on site is not used but correlates with whatever Google uses.

A good point, there. Especially because it might correlate in one site's case, but not in others. As netmeg said above, high bounce rates do not necessarily kill rankings. I also had a client who got worried about an 85% bounce rate - on a highly trafficked page that was #1 for three killer query phrases. So they monkeyed with it, got the bounce rate down and completely lost all three #1 positions.

[londrum]

Yes I agree, but if I am shopping for widgets and click on a site only to find out the widget costs $$$, am I not going shop around to find the cheapest? immediately I am going to check 2, 3 or perhaps 5 other sites to see what is the cheapest, which will likely be seen as a high bounce rate.

i am sure that they take account of this in the algo. if someone returns to the same site multiple times after clicking on the serps, then google are obviously going to want to track that, because it's a sign of satisfaction.
they can do that through chrome, analytics, adsense... loads of things.

so it shouldn't matter how short the first click is.

[Matrix]

walkman it all make sense because i was ranking for long tail keywords and now i have lost allot of traffic because of it..

short key word are harder to rank, need good relevant links and good optimization content.

i rank in first page for allots of long tail keyword but i get just a bit of traffic because not many search for those long tail keywords.

[steerpikegg]

Bounce rate on its own is not relevant. Time on page/ site on its own is not relevant. Both are only relevant in the context of whether people return to Google to "try again" by clicking another search result or perform a similar search.

I can't see how this could be a genuinely useful metric, because I am sure that I am not alone in the way I browse the web.

Often I will have more than one tab open with different Google searches in them as I of search for a number of different or related terms concurrently. Also, from a page of serps, I will usually middle click several results into tabs to look at rather than jump back and forth.

By this rationale, every site I visited would be considered a bounce or dud result.

[Whitey]

Site's I'm observing seem to have had speific keywords hit, and others relatively untouched. Are folks noticing something similar, and when sites return from Panda, is this resolving back to normal?

[suggy]

Yep Whitey, it seems to be keyword/ keyphrase specific

[noumaan204]

One of my pandalized sites recovered on 27th September. I am nowhere closer to the traffic I received from Google before February. But noticed my rankings improve on several keywords.

I lost 90% of traffic. Recovered 20% of lost traffic with this update. I am hoping to see further improvements as I publish new content.

In my case the bad thing that I noticed was that I submitted a few articles on article websites that were copied on countless other sites. I tried my best to remove as many of them as I could. But several websites that copied my articles from article websites did not have contact form or did not respond to my removal requests.

I am hoping that since Google is punishing all article websites these low quality sites that stole my article and are actually linking back to me will soon go out of business and will eventually shutdown.

I also noticed that decreasing the number of ads also helped.

Another thing I noticed is that I worked really hard to improve the page speed. May be this helped improve my rankings.