Panda #2: What's the culprit?

bounce rate definitely plays a factor in SEO, i wont deny that fact. However how much it affects the SERPs is still questionable. Myself believe it's just a minor factor, unless of course if your bounce rate is higher than 80%.

How does google see the bounce rate?!?
 
How does google see the bounce rate?!?

That's a really good question. And in fact, there are actually two (at least) definitions of bounce rate, based on how Google sees it.

One, of course, is if you are using Google Analytics. But it's also possible they could use info from Adsense, or perhaps some other kind of readable footprint, depending on what's on your site.

So in Google Analytics, it will show your bounce rate and this is based on the number of visitors who come to your site and only view one page. Certainly, Google can not measure this for every website. Plus, it's not actually a particularly useful metric.

But in his paper, Panda's definition of bounce rate is different. It's the number of visitors who go to your site and "immediately" return to the search engine. He doesn't specifiy exactly how long immediately is.

But I think this is the type of "bounce" that they are probably looking for in terms of ranking. It just makes sense.

There's all this talk about "quality". But Google can't measure quality directly. They have to use "signals" - measurable variables that can tell you something about the quality of a page. Well, if someone comes to your page and stays for 2 seconds and then goes right back to Google, that's a pretty good indication that your page sucks.

So I suspect that's how Google determines bounce rate -- immediate returns to the search engine.

BTW, if this is the case, that means that the only bounces that matter would be from Google. Depending on your traffic methods, this could be very important, as nearly 100% of traffic from places like Digg and Stumbleupon are "bounces" if you use the 1st method (visiting multiple pages) but are NOT bounces using the second, more likely, method of immeidate returns to Google. In this definition, only search engine traffic can be a bounce.
 
That's a really good question. And in fact, there are actually two (at least) definitions of bounce rate, based on how Google sees it.

One, of course, is if you are using Google Analytics. But it's also possible they could use info from Adsense, or perhaps some other kind of readable footprint, depending on what's on your site.

So in Google Analytics, it will show your bounce rate and this is based on the number of visitors who come to your site and only view one page. Certainly, Google can not measure this for every website. Plus, it's not actually a particularly useful metric.

But in his paper, Panda's definition of bounce rate is different. It's the number of visitors who go to your site and "immediately" return to the search engine. He doesn't specifiy exactly how long immediately is.

But I think this is the type of "bounce" that they are probably looking for in terms of ranking. It just makes sense.

There's all this talk about "quality". But Google can't measure quality directly. They have to use "signals" - measurable variables that can tell you something about the quality of a page. Well, if someone comes to your page and stays for 2 seconds and then goes right back to Google, that's a pretty good indication that your page sucks.

So I suspect that's how Google determines bounce rate -- immediate returns to the search engine.

BTW, if this is the case, that means that the only bounces that matter would be from Google. Depending on your traffic methods, this could be very important, as nearly 100% of traffic from places like Digg and Stumbleupon are "bounces" if you use the 1st method (visiting multiple pages) but are NOT bounces using the second, more likely, method of immeidate returns to Google. In this definition, only search engine traffic can be a bounce.

I fully believe this. The number of sites that uses GA or adsense, compared to the entire universe of sites, is *tiny* - there's no way they could construct their algorithms with data from GA or adsense being a key component; they'd just be guessing, on all the sites they couldn't actually measure. GA and adsense can only provide "extra information."

And that would mean they'd have to define the bounce rate by something they can measure for *all* sites. That would be how long it takes for someone to return to G after clicking a search result.

No reasons. Just common sense.
 
Yesterday,one of my sites went from number 1 to number 420,i had mostly spam links,i will go to build some quality links and will see what will gonna happen
 
A change in link building strategies is in order me thinks... quality in content links seems to be the way forward.
 
This coming from a guy selling scrapebox secrets in his signature (take with a grain of salt)

Funny, my sites actually went up. And all I use for off page seo is SB blasts and AMR blasts.
I say more Panda action please.
 
That's a really good question. And in fact, there are actually two (at least) definitions of bounce rate, based on how Google sees it.

One, of course, is if you are using Google Analytics. But it's also possible they could use info from Adsense, or perhaps some other kind of readable footprint, depending on what's on your site.

So in Google Analytics, it will show your bounce rate and this is based on the number of visitors who come to your site and only view one page. Certainly, Google can not measure this for every website. Plus, it's not actually a particularly useful metric.

But in his paper, Panda's definition of bounce rate is different. It's the number of visitors who go to your site and "immediately" return to the search engine. He doesn't specifiy exactly how long immediately is.

But I think this is the type of "bounce" that they are probably looking for in terms of ranking. It just makes sense.

There's all this talk about "quality". But Google can't measure quality directly. They have to use "signals" - measurable variables that can tell you something about the quality of a page. Well, if someone comes to your page and stays for 2 seconds and then goes right back to Google, that's a pretty good indication that your page sucks.

So I suspect that's how Google determines bounce rate -- immediate returns to the search engine.

BTW, if this is the case, that means that the only bounces that matter would be from Google. Depending on your traffic methods, this could be very important, as nearly 100% of traffic from places like Digg and Stumbleupon are "bounces" if you use the 1st method (visiting multiple pages) but are NOT bounces using the second, more likely, method of immeidate returns to Google. In this definition, only search engine traffic can be a bounce.


So a reasonable response to this is to "bait" the customer into staying for....lets say...30 seconds?

http://www.searchenginejournal.com/website-high-bounce-rate/11223/
 
Methinks its time to hire human outsourcers for ALL linkbuilding.

Amiright?

-
 
do you believe that using gwt or ga will screw you over at the end of the day?
 
Methinks its time to hire human outsourcers for ALL linkbuilding.

Amiright?

-

You probably are. I think we're going to have incorporate more whitehat seo techniques to our link building strategies. Well, thats what I think.
 
Methinks its time to hire human outsourcers for ALL linkbuilding.

Amiright?

-
I've start thinking to do so as well. I still give myself a chance to use different services all over BHW as well as using automaton tool like SB and AMR. If that doesnt work, I would definitely try to hire human outsourcers for manual in-content link buildings
 
I've start thinking to do so as well. I still give myself a chance to use different services all over BHW as well as using automaton tool like SB and AMR. If that doesnt work, I would definitely try to hire human outsourcers for manual in-content link buildings

Aren't most human outsourcers gonna be using software you could normally use? If not they may develop a small bot to fit their purpose. One advantage for going to outsourcers are they are tried and tested and they have learnt from mistake (which can help you) but if you have the time isnt it better to DIY?
 
Back
Top