Is Panda all about Exit Rate?

[claaarky]

Commercially, what I am about to do may not be the most sensible thing, but I feel it’s right and I want to share what I have discovered about Panda. It may help you understand more about quality and how to escape Panda.

Please note, I have not escaped from Panda yet – I came to these conclusions on June 22nd 2012 and began addressing my issues based on a new understanding of how Panda works. This theory could develop and I could end up with egg on my face massively, but it makes more sense than anything I’ve ever read anywhere before. Here goes............(apologies for the long post in advance).

Since Panda hit my ecommerce site in April 2011 I’ve been trying to improve the quality of my site using Amit Singhal’s guidelines as a basis but completely without success.

I always imagined that Panda was a magical formula Google concocted using their human guinea pigs when they sat them down and asked all those questions relating to quality, and that ‘Panda’ is Google crawling your site looking for the signs of low quality. It’s not.

Last week my attention was drawn to a statistic in Google Analytics that for some reason I’d never noticed before – Exit Rate. That’s when it dawned on me – Panda is all about user metrics, they can’t ‘see’ your site, they don’t crawl it using a magical formula, they collect signals given off by humans as they use your site to tell them where the bad quality is. If you have too much of it, they demote the rankings of the pages with bad content and any pages that link closely to those pages to protect Googlers from hitting your bad content (what we know as Panda).

I compared pages on my site with very high exit rates to those with very low exit rates and immediately it struck me how much better the pages with low exit rates were. It also struck me how many different reasons there were for the high exit rate pages being worse (in many cases it was just a bad product that rarely or never sold, or the price was too high, the description was poor, the image as poor, etc.). The low exit rate pages were our top sellers, good products, good descriptions, nothing bad to say about the product or the content or presentation of the page.

Then I realised this is where Google started. They wondered about Exit Rate, sat people down, asked them to compare web pages, asked them why they liked or didn’t like a page, and found that Exit Rate correlated with human feedback. It’s obvious really – people leave your site because they’ve either done what they came there to do or something put them off. This is the ultimate test of quality.

Google doesn’t need to ‘see’ your pages, it just looks at where people leave your site, maybe what they did before leaving your site (how long they were there, how many pages visited, etc.) and if your site has a high proportion of pages with a high exit rate, your users probably don’t like the quality of those pages.

Of course, people have to leave your site at some point, and that may be because they’ve found what they want, so there has to be an allowance for that. And there may be a different model for different types of sites. But I found, when I looked at my high exit rate pages, in most cases it was obvious why people didn’t like them. In the case of our product pages, the high exit rate pages were generally non-sellers, cluttering up the site and, as I now realise, turning off customers.
To try to disprove my theory I started reading back through Amit Singhals guidelines and it all made sense (as I knew it would one day!). I also looked back at various discussions about Panda, things that people did to recover from Panda, and it explained everything.

It’s beautifully simple and it deals with a huge range of Google’s problems in one hit. Users naturally react differently to webspam, duplicate content, scraped content (or original content if it’s been scraped), brands. Sites with a high proportion of high exit rate pages tell Google all they need to know about the quality of your site, from a web user’s perspective (which takes into account an unfathomable range of considerations that even Google have struggled to document - what Google needed to say is what I’m saying now, look at your exit rates!).
This theory explains so many of the things we’ve all noticed about Panda, how it works, its’ effects on Google’s results and our sites. Here’s a few......

Why can’t they run Panda more regularly?
They need a month’s worth of user metrics to be able to make a judgement about your site.

How did I recover from Panda without changing anything?
Scrapers can hurt your exit rate. So can competitors. If you have content people have seen elsewhere it affects their perception of your site. If Google got rid of your scrapers, your user metrics would improve without doing anything. The user metrics of your site are affected by what’s happening on other sites so even a new competitor doing something similar to you can affect your user metrics.

Why are brands dominating the results?
It’s not brands that are dominating the results, it’s websites people like and trust that are dominating. Not every brand will always be loved and trusted, and their user metrics will reflect those changes. But generally people trust what they know so sites people like (let’s create a new term to replace brands – SPL, sites people like) can have bad pages but people won’t leave their site just because of it, so their user metrics are better. You could set up an identical site with your name at the top, the ‘quality’ of the content would be identical, but the user metrics would be much worse.

Why did Google suggest merging pages?
I’m guessing user metrics show that users don’t like seeing several similar pages on your site, in the same way they don’t like seeing similar content on numerous sites.

Do images or photos help improve quality?
Not necessarily. Every page on the web can produce a different response from users. The only way to know is to experiment, check your exit rates, repeat until exit rate is low.

Should I add more content to my site?
See “Do images or photos help improve quality”.

Can I escape Panda by improving my brand signals?
If you mean getting backlinks with your website name in them, no. That does not make you a brand. Google doesn’t actually care if you’re a brand or not, it just knows that people react better to SPL’s (sites people like). User metrics prove it.

Can I escape Panda by getting better quality links?
No. Users can’t see your links, links make no difference to how users perceive your site alongside the rest of the web. In my experience, while demoted by Panda, links won’t get you anywhere. Once you get out of Panda though......well, hold onto your hat.

Why does moving content to subdomains work for some people?
If you correctly identify, fix or remove bad content from your site (using Exit Rate as a guide) you will be left with only good content. It doesn’t matter how you do it, what matters is that you get your exit rates down.

Will no-indexing or blocking robots from pages of bad content help?
No. If a user can see it, Google has user metrics on it. This is not about googlebot, it’s about your users.

I could go on, but you get the point. User metrics tell Google everything they need to know about how real human beings perceive your site in context with other sites they may have come across. It really is as simple and as complex as that.

What I don’t know is where the threshold is and I’m guessing it may be different for different types of sites (information versus ecommerce for example) and there will be other factors combined with it but, put simply, I think Exit Rate is the place to start looking if you want to find and fix your bad content issues. It really opened my eyes. (Note: I think bounce rate impacts exit rate, I might be wrong on that – removing bounce rate from the Exit Rate calculation may give you a truer reflection on how people react with your content as they move around it).

I suspect that some sites may not able to get below it if they are basically set up with spammy intent (you see how effective this simple method is) but for many of us, understanding that exit rate tells you where you bad content may be, could be the answer to your prayers.

I hope I’m right, or at least on the right tracks. If I am, it’s time to end the Panda woe and improve our website KNOWING what quality content really is. Maybe a discussion here will help test the theory and perhaps help us gain an even greater understanding.

I hope this helps you and me escape Panda, I truly do.

[claaarky]

Oh my God, another realisation......

Links are now completely useless to you unless they drive traffic to your site and the visitor then interacts with your site in a way that looks natural or perhaps reacted well to your site. Any anchor text attribution will only have impact if the link is there for genuine reasons AND drives people to your site who like it.

Blimey, they've nailed this!

Traditional business rules apply. Advertise, drive traffic to your site, make your site a great user experience.

This is like a massive weight off my shoulders. I'm sick of all the SEO rubbish I've done over the years - eye of toad and wing of bat, that'll get you to top of Google. Yeh right.

[Leosghost]

Imagine they've done a deal with Microsoft and Apple for user metric stats via their browsers (internal exit rate, time on page, url).

Your imagination overunneth :)
Nothing in doing that for MS or Apple..in fact the precise opposite..it would harm them ..they would be fools to do so ..they are not fools..:)

Oh my God, another realisation......

Links are now completely useless to you unless they drive traffic to your site and the visitor then interacts with your site in a way that looks natural or perhaps reacted well to your site. Any anchor text attribution will only have impact if the link is there for genuine reasons AND drives people to your site who like it.

Blimey, they've nailed this!

Traditional business rules apply. Advertise, drive traffic to your site, make your site a great user experience.

That is how it has always supposed to have worked..:) since before Google..it sometimes didn't, and was manipulatable..

What changed the way that webmasters looked at user experience etc was adsense ..where the goal of many webmasters became how to get the maximum number of visitors, any "quality", by whatever means, and how to get them in and off the the page as quickly as possible via a click on an adsense ad..

Google have been dealing with the fallout from their creation of this cesspool ever since..

Those of us who did not build our sites with this "churn" model ..have always known what you have realised..

This is why many of us have not suffered either during or after these purges..even though frequently the scum does seem to float up to surround us..

And ehow gets a pass because Google will not apply the ressources needed to ID original from human or machine scraped..and as that method "churn" appeared to be working..many did not realise that to base an Ecommerce site on anything resembling it, made it too vulnerable..

[claaarky]

Just had a look at ehow for the first time ever......first impressions, looks nice.

People must like it, otherwise it wouldn't be sailing at the top of Google's results.

[Leosghost]

Shame it gets there by being written by people who know little about the subject they write on..but who go away and "source" their articles by rewriting the websites of others in very sketchy and usually inaccurate form..

On the rare occasions that their "writers" ( at $15.oo per "article" ) acknowledge where they lifted their "articles" from..they "nofollow" the links ..or don't link at all, but just say the name of the site(s) with no link..most times they don't even do that ..they just steal content change a couple of words ( frequently losing the accuracy in the process ) and slap adsense on ..makes ehow and Google a fortune..so they get a "pass"..

Panda only knocked them down for a few days..they are now right back up there, monetising other peoples content without permission..

Visitors may like it..visitors don't know ( or don't care maybe ) that is made entirely with other websites content re-spun by humans working as freelancers for ehow..

[diberry]

The fact that I haven't recovered from Panda yet (having only just started fixing my site on this basis a few days ago) is obviously getting in the way.

The other part of it is your tendency to just dismiss or not really address any points that show holes in your theory. Just looking at the last page:

Martin Ice Web is right: if his bounce and exit rates went up in response to Panda, that implies something about Panda.

Shaddows' point is also good, and it's wrong to advise people to only think about user experience when dealing with affiliate links. What has the "nofollow" tag Cutts advises us to slap on those got to do with user experience? It's very clear we can't just focus on user experience if we're affiliates.

Leosghost points out that Ehow doesn't belong at the top, going by your theory, and you're not answering how that's happening.

And I shared with you the fact that I improved my exit rate situation, and got Penguinized. All my user metrics were pointing to ever-improving visitor experiences, and Google's response was that I must be a spammer to be improving those metrics that much.

These are points that need more discussion.

[claaarky]

Leosghost
That's it, people like it. If people saw that content in lots of different places before they'd leave ehow in their droves, user metrics would reveal that and they'd fall down the rankings.

It's all about what people like.

Not sure where you are but in the UK we have a TV soap called Eastenders......it's not my cup of tea by any stretch of the imagination but it gets massive viewings 5 nights a week. Most people like it - Google is giving users what people say they like.

If people got bored of Eastenders they'd stop watching and the BBC would drop it in a second. Same rules apply in Google, in my view.

[claaarky]

diberry
Apologies if you felt I'd been dismissive to anyone (and apologies to anyone who felt that).

I feel like I'm saying the world is round and people keep challenging that by saying well why don't we all fall off then.

I see things so clearly now that I've got my brain around the idea Google is collecting user metrics via the browser. It just changes everything and it's something that, as I think about it more, explains everything that's being discussed on this forum about Google.

I have no experience with Penguin but what I have come to realise about Panda is it's here to tell us to stop wasting time on traditional SEO stuff and focus on creating a great user experience. I think Penguin may be here to tell us to stop building links, not because Google is annoyed but because people are wasting their time and money on these things. Google wants to save you time and money - it's another difficult idea to digest considering all the pain we've all been through, but that's how I see it now. It can be fixed, people need to stop thinking of Google as the enemy.

I'm not the enemy either. I'm trying to help and I'll be very happy to try to help people if they want to explore the idea with an open mind and in a respectful manner.

[diberry]

I feel like I'm saying the world is round and people keep challenging that by saying well why don't we all fall off then.

It's a lot more like you're telling us the world is triangle shaped, and we're showing you the evidence that it's roughly spherical, and you're saying, "Well, maybe there's a little bit of rounding at the corners or something, but it's still pretty much a triangle."

It's been explained to you that a lot of us were looking at Exit Rate stats long before you suddenly realized their importance. Does that not make you think that maybe some people in this thread know more about this than you do?

And then here is where it gets really strange, and rather evangelical in tone:

Google wants to save you time and money - it's another difficult idea to digest considering all the pain we've all been through, but that's how I see it now. It can be fixed, people need to stop thinking of Google as the enemy.

You don't know precisely what Google is trying to accomplish, and neither does anyone posting here. And most of us don't think of Google as the enemy - we think of them as a marketing channel that has its own agenda, just like we do.

Here's a question for you.

They said Panda was targeting content farms - but clearly, ehow found a way to beat that without changing its essential nature as a content farm, and one that is full of not only scraped but often incorrect information. I guarantee you ehow has terribly high exit rates on the very pages Google keeps serving up to us. Explain how that fits with your theory.

[claaarky]

People like it, that's all I can say.

If it's at the top of Google people like it.

[Leosghost]

I remember eastenders, the crap for the masses was running even back in the eighties when I left the UK, with grange hill for the under 15s, brookside and the rest..part of the reason I left, was that if that was becoming popular, and what the majority wanted ..I didn't want too live where that majority did..

I agree with the broad view of what you see ( re quality and interest,"engagement" etc, and do visitors "like" pages )..always did think that way, and built accordingly..but as diberry says exit data isn't all the story..each site is different ..and different exit metrics may have the same result , and vice versa..

From me personally they get no info via Ganalytics because I never used it, and never would , especially on an ecommerce site..I would strongly advise against using Ganalytics , it gives them all the data they need to take over any vertical with their own products..ask the people in travel etc whether they think giving G all their data was a good idea ;-))

I also don't use WMT and never will ( some here say it doesn't harm and that "it lets you know what G knows/ thinks about your site" )..The only thing that interests me about what G thinks / knows about my sites , I can see in my serps positions..and ( for now ) I don't have to be signed in to search..and I know ways to nullify their geo targeting aspects, and their personalisation of serps when I want to ..and I don't trust them when they say they don't not use data from Ganalytics or what you do when inside WMT to fold into what they do with your site or your "niche"..

They were caught lying over the "wi-fi slurp" and trying to cover up their lies..and then they tried to take the 5th..so I only trust them to send any adsense money due..what their "cut" is , I don't care..it is "extra"..

I'm the grandson of an Irish horse trader..you don't look in the mouths of "gifts" ..but you do check that they have solid legs and backs..and slap their ribs to check for hollowness and trojans ..;-) and you know when you ride them that there is always a risk of being throw for a reason that you probably won't see coming..

Google isn't interested in saving you money re "links"..the only links that G truly approves of ("approved" is decided at a level above Matt's pay grade ) are the links that they sell you via adwords..

They figure if you do have the money and the time to buy links "to be seen" ( as distinct from links to bring actual "organic" traffic ) then you can damn well buy them via adwords from them..

Google are not your enemy..neither are they your friend..as netmeg said..( cant' find the exact post to quote , but it made me laugh out loud and spill my wine;-) ."they'll throw you under the bus in a heartbeat if it suits them, their bottom line, their advertisers and their users" ( and in that order of importance to them when they make that decision ) they 'll do that if your site is good or bad, sticky or not, good metrics , bad metrics, makes no difference to them..

And if they want your vertical for themselves, what they cannot buy they'll crush into oblivion with "free"..anyone who thinks or says differently is IMO and IME deluded, or is trying to sell you "how to rank" potions, books, software and courses..or has not been paying attention, or is shilling..and all four of those things happen in these threads, and on this site and elsewhere..

You can use Google to help you make money, but don't count on them for the mortgage, and don't make your "niche" look so attractive that they'll move on it , and don't make them look dumb..long term it is actually easier and pays off better being unique and with a good site or sites and building for real live users, rather than farting about trying to second guess the next animal..sites that look too much like they depend on Google to exist, they seem to slap hardest, they just don't get all of them each time, and ehow gets a pass..when all the drama of each animals passage has died down..
edited for "speeling"

[edited by: Leosghost at 5:15 pm (utc) on Jun 28, 2012]

[diberry]

People like it, that's all I can say.

If it's at the top of Google people like it.

So, you're saying Panda has a metric that measures whether people "like" your site, and if people like it, it doesn't matter how bad your Exit Rate is. In other words, you're saying that Panda is all about Exit Rate, except not.

You just invalidated your own theory, claarky.

[Leosghost]

if it is "at the top of Google" ..all that says is that it is "at the top of Google" ..the reasons can be , and frequently are multiple..people "just liking it" is merely one of the possible ones..and exit metrics are just one of the possible ways to measure those "likes"..

But any thread which gets those who are stuck thinking in alternative paths to the ones that they have not found fruitful ..is of merit ..

[tedster]

I'm really happy that claaarky shared his insight here. He noticed a very strong correlation between high Exit Rate pages and pages that were dinged by Panda. Then he took a hard-nosed look at those pages and they looked very weak to him and not just to his visitors - so weak that he started a project fixing them. That sounds really sane to me.

We can't know for sure if Exit Rate is specifically part of the Panda formula or not. But there's no doubt in my mind that Google and Bing both use LOTS of user metrics in their ranking algos. If you've been following along over recent years, it's really clear that they do. And as they learn more and more and more, their dependence on links and other easily manipulated signals starts to lower.

Obviously there are lots of flaws in the current Google rankings. I'd say Google is in the middle of a long term project (more than a year long already) to move exactly that way - use machine learning to measure user engagement and eventually place their algo on a whole new foundation. While Google is on this particular learning curve, things are going to be a mess in various ways.

I think claaarky may have spotted signs of the approaching new age of SEO, and it won't look much like the old age did.

[freejung]

I just had a look at my analytics from this perspective, and I would say that average time on page correlates much more closely with quality. Actually, my lowest quality pages (in my opinion) typically have a very low exit rate, a high bounce rate, and a very low average time on page.

What's happening is that people who land on the page decide they don't like the site and leave, thus the high bounce rate. People who navigate in from other pages have already decided they like the site, but they don't like that particular page, so they navigate quickly to another page.

In any case, I think it's a lot more complicated than just exit rate and depends a lot on the nature of your site. But Google has a lot of data and can calculate correlations between all of these factors and perceived quality. However, I definitely think you're on to something in terms of just looking through your analytics to try to find metrics that indicate low quality pages.

Anyway, thanks for this insight, it has definitely led me to look at my pages in a new light and to realize that there are still pages on my site that people clearly don't like.

I'll work on improving or getting rid of those.

[Copeland]

Focusing on exit rate and bounce rate without considering avg time on page is a big mistake. If you have a page with a an exit and/or bounce rate of 95% but the avg time on page is 5 minutes, the high exit/bounce rate is actually a positive to Google. Why? Because, combined with avg time on page, it's sending the message that the content is so good that nearly everyone consumes it and then leaves without looking elsewhere on your site.

[claaarky]

Believe me, my world has turned on its' head in the last week!

Everything I thought about SEO that hasn't made any sense in a normal business environment for years, has gone out of the window this week for me. We don't have to waste our time and money on those things any more. We can work on making our sites better and operating like normal businesses do.

They're somehow getting the stats via the browser, not Google Analytics. Get that and lots of things make sense.

If something is at the top of Google it's 99% (I believe) down to that site providing the best user experience, as determined by that site's user metrics. Whenever you think of a good example of why I'm wrong, look at the site from a normal person's perspective, ignoring your knowledge of how they operate.

If you find an empty page ranking from a popular site think of it this way. Panda evaluates and ranks every single page based on how people interact with it. If a popular site has one bad page ranking but people click on it and, because they know and trust the site, they click to another page, Google could misinterpret that as a sign people liked that page.

HOWEVER, I think they'd like to turn up the dial to catch that stuff, but if they did that right now your Panda hit site would be wiped out of existence.

Google is giving us a chance to sort this out but not enough people have understood what they're after regarding user experience, so they are waiting, looking at the results thinking they're not perfect but if we tighten things up we could have suicides on our hands.

When you get what I'm saying (and you may need to read my posts on this thread numerous times) it will blow your mind.

[jinxed]

The most important metric, in my opinion of course, is the one that we cannot ourselves fully measure.

User A enters a key phrase: "Keywords 1 + Keyword 2", and is presented with 10 results. User A clicks on result 1. They view the page, decide it’s not what they are looking for, and so click the back button. User A then decides that result number 4 might be helpful - so they then click on that. Viola! They get just what they want, so exit that page.

User B also enters key phrase “Keyword 1 + Keyword 2”. They look at all the results. Result number 4 looks good, let’s investigate! Guess what, it just what they were looking for! Time to go on their merry way – straight off to Facebook (or whatever mind numbing site they want to visit next. User A and B aren’t the brightest chaps – but result 4 gave them just what they wanted).

What were both their exit rates? Or am I missing something? It’s not as simple as 1 metric – but checking this metric has got to be a productive thing. Only an opinion of course.

[Leosghost]

They're somehow getting the stats via the browser, not Google Analytics. Get that and lots of things make sense.

Except where they are not ..because we don't all have chrome users as a high or even a relatively significant percentage of our visitors ..hint ..you don't need Ganalytics to know what browsers your visitors are using :) ..

HOWEVER, I think they'd like to turn up the dial to catch that stuff, but if they did that right now your Panda hit site would be wiped out of existence.

Google is giving us a chance to sort this out but not enough people have understood what they're after regarding user experience, so they are waiting, looking at the results thinking they're not perfect but if we tighten things up we could have suicides on our hands.

You are still believing ( inspite of all the evidence to the contrary and even the legal requirements of a US "corp " with shareholders ) that the owners of a corporation Google care about webmasters individually or collectively..they don't..

I would recommend you ( and anyone else who thinks this way ) to read about monkeysphere..you could begin here [cracked.com...] it will also help explain why many other things happen between people or people and corps ..or even visitors and websites..:)

[diberry]

I'm really happy that claaarky shared his insight here. He noticed a very strong correlation between high Exit Rate pages and pages that were dinged by Panda. Then he took a hard-nosed look at those pages and they looked very weak to him and not just to his visitors - so weak that he started a project fixing them. That sounds really sane to me.

I feel what Claarky is really onto is a user metric that, in many cases, tells a webmaster a lot about how much visitors are liking the site. In fact, that was my only interest in this discussion. I ignore Google and build my sites for users, but I wanted to make sure the Exit Rate wasn't a valuable metric I had missed - not for beating Panda, but for knowing if I'm pleasing visitors.

But I reject the notion that it's the key to Panda. We've offered numerous examples of cases where it would actually be a really poor indicator of visitor satisfaction - quick answer pages, for example. Recipes. Sites where people check in to a community to see if there's any news, and upon seeing there isn't, back out and go somewhere else. Google is plenty smart enough to realize that exit rate, useful as it is, just cannot be the center of their calculations.

I think what Panda may be doing is parsing pages into niches, and then comparing various metrics within each niche so that your page is getting compared to similar pages. If that's true, engineers would be able to create metrics profiles for an imaginary ideal site in each niche, and compare your page to it. I don't know if that's what Panda is trying to do, but that's the feel I get from it. If so, there is no single metric - not even a single collection of metrics, or ranking of those metrics in importance - that applies to every site across the web.

[claaarky]

Wonder what Webmasterworld's user metrics look like for this page at the moment.

Average Time on Page: 3 hours
Exit Rate: 0%
Unique page views: 1,000,000

Pretty engaging content isn't it. The sort of thing that deserves to rank top in Google.

[Leosghost]

So you didn't click on my monkeysphere link then :)
if you think "Exit Rate: 0%"
and I would hope that the members here read a lot faster than 45 minutes per page.. 10 is excruciatingly slow IMO..I would hope no more than 2 to 3 min, say 5 max per page..

[jinxed]

Average Time on Page: 3 hours
Exit Rate: 0%
Unique page views: 1,000,000

Huh?

I posted a reply earlier, and then proceeded to exit this website via this page. How does that say anything about the quality of this thread?

Surely the fact that I am a "returning user" to this page says alot more about the engagement of this page?

[tedster]

They're somehow getting the stats via the browser...

Some ISPs are also willing to sell their user data, or so I've heard. And then there's the free wi-fi that Google sponsors here and there. Plenty user data for them via that avanue, I'd think.

-----

One problem I see in this discussion we're having is that some people are actually doing the experiment and looking at their Exit Rates, comparing that to Panda-smacked pages. Other people are trying to discuss this in an abstract, theoretical way.

The opening post makes it clear that claaarky was NOT being abstract. This was a personal journey done in concrete terms with real data on a real website - and one that matters for his livelihood. Theoretical discussion is pretty much beside the point, then, isn't it?

I just looked into one site that Panda hurt, and I do see I strong correlation between Exit Rate and harshly demoted pages. It's not 100% - more like 75% - but that makes the data worth taking quite seriously, I think - whether it fits into my theories of how Google works or not.

[edited by: tedster at 10:10 pm (utc) on Jun 28, 2012]

[getcooking]

It's not 100% - more like 75%

This is almost exactly what I was seeing on my Panda-beaten site.

[Leosghost]

This was a personal journey done in concrete terms with real data on a real website - and one that matters for his levlihood

And thus his premise was valid for him..

Theoretical discussion is pretty much beside the point, then, isn't it?

I think you'll find though...the "theoretical discussion" only grew from other members reactions to claaarky's extrapolating his case and findings into being the solution to the ranking since panda of all others, whether they were affected adversely or not..

Done no doubt with the best of wishes..but there is no "one size fits all" "exit metrics solution"..his posts are nevertheless useful directly to some and indirectly, via the stimulation of ideas to others..as I said earlier, the thread has merit..many post panda and post penguin threads here, IMO do not ..

[lizardx]

claarky I think you're right to look more at external factors, though I share the view that exit rate is not the best to look at, in fact, there is no finer way to make google like your site I think than having a search term be entered, they pick your site, and do not come back to the search. And using outbound external links to help the user on their way, maybe helps, but if they are tracking each user via toolbar and chrome usage, might not.

However, it's also clear that google is using their toolbar and google chrome browser to collect highly accurate information that is 100% accurate about what real users do when surfing. And google loves their data.

google analytics depends on people placing it on the page, it's not a very good or reliable tool for true behavior detections, but chrome and the toolbar are. And both re now being used in statistically meaningful numbers.

While I have serious doubts about google being able to actually implement any type of AI process that can understand a page (doubts reconfirmed every time I see spam and black hat pages ranking, such as two of our competitors currently...), I do believe they can very easily integrate user surfing patterns and the actual search actions, ie, return, no return, modify search term, not modify. Google has such massive numbers that all human actions become totally predictable and consistent, I see such things with traffic as low as 1000 a week, imagine many orders of magnitude larger data sets.

What I would guess is that, the exit rate PLUS the user searching again for the item on another site, ie, clicking another link, or modifying their search, is the actual trigger. An exit page where no further search occurs means the user found precisely what they were looking for, which is the theoretical goal.

I need to go read some more blackhat seo forums, it's actually a useful process, more agressive views there of what goes on, tedious, sort of soiling if you have any ethics, but still educational.

[diberry]

The opening post makes it clear that claaarky was NOT being abstract. This was a personal journey done in concrete terms with real data on a real website - and one that matters for his levlihood. Theoretical discussion is pretty much beside the point, then, isn't it?

This is why I thought my story of already having done what claarky is just starting to do - and the ensuing reduction of my livelihood from my real website - might have some relevance as a cautionary tale. But that was brushed aside and reinterpreted to support claarky's theory.

If we're not here to talk theory, but we're also not welcome to share reports of concrete experience/data that conflict with claarky's theory, then what is this thread for?

[Leosghost]

Google has such massive numbers that all human actions become totally predictable and consistent,

Psychohistory..Foundation and Empire..Asimov..I've been saying here and elsewhere for years, that it is obvious that Page or Brin or both have read the series..and are trying in their own ways to put the imagined ( but IMO self evidentially sound ) theories into practice..Making shedloads of money is almost a sideline as against the heady drug of "influence" over such huge numbers of people and society in general at the moment and well into the future..

Watching the manifestations of the various currents of thinking at Google influenced by this and other philosophies, and their real world effects, directed sometimes, ( and at others given an impetus in a direction and then left to their own devices to make their own structures ) by the "tool" of Google, is the most fascinating thing about Google..

[lizardx]

claarky, my apologies, I posted above before noticing the pages 2,3,4, now have read the rest.

"Links are now completely useless to you unless they drive traffic to your site and the visitor then interacts with your site in a way that looks natural or perhaps reacted well to your site. Any anchor text attribution will only have impact if the link is there for genuine reasons AND drives people to your site who like it."

I want to add in my support for your rough view, and also to agree with what leosghost, and I, by the way, have always known and acted on with my own sites. My sites never see any issues with google updates because they are designed to give the users what they were looking for in all cases. The sites I do that do not do well in google do not do that, consistently, but I don't do those sites for google, I do them for me, and a handful of friends, plus anyone else who might stumble on them during some surfing.

To be concrete, I've referred to client's site, we'll call it x. This site is probably guilty of all 9 categories of bad user experience, bad usability, dupe content, and so many other things it's not worth mentioning. But the thing that strikes me very specifically is the notion of exit page re the seo generated spam links, which are themselves something I assume you are totally not considering because you simply did not do them, but let me assure you, if you did, all the issues you experienced would be amplified in terms of traffic drop by about 4x or so.

To the point, our bad outsourced seo (man, let me warn you, do NOT price shop seo, it will kill your site) linked to about 6 to 10 target pages, all category index pages. All those target search terms, plus one they had somehow missed but caught in the last two panda updates, and anchor text repetitions, ALL pointed to index pages for sections, almost none pointed to the actual page, because, frankly, it takes work for the seo to actually write an article that is topically related to the target page, and to make anchor text that actually refers to the content of the target page. That is human, not machine behavior, and I am virtually certain that the black hats are now correcting this obvious and sloppy oversight on their part.

Leosghost, as usual, I agree with your more meta analysis, particularly this: Making shedloads of money is almost a sideline as against the heady drug of "influence" over such huge numbers of people and society in general at the moment and well into the future..

It's worth noting that George Soros, a man who also requires no seo to get his name or products out there, also engaged in his initial finance experiments not to get rich, but to verify a specific theory he had about human behaviors, a theory he explains to some degree in 'The Alchemy of Finance'. Needless to note, his experiments were largely successful, and validated his philosophical theories.

Now as to the raw idiocy and incredibly ignorant treatment of the world's books in google books scanning project, well, don't ever try to read or even look at one of those scans if you love books is all I have to say, the world is better off without them. Clearly the only part of the world google guys can grasp and work with is the purely virtual one.

I want to note a few other things of interest re our site, x: our user demographic is about the least likely to use google chrome out there. And of users of google chrome out there, they are about the most likely to leave because our site was not actually what they were looking for. And I believe that could actually be a part of the problem, it's possible that google's treatment of the toolbar and chrome data does not actually take into account the demographics of who uses MSIE, Firefox, or Chrome. That's my guess anyway.

I have been after client for years and years to improve user experience and put users first.

Also of note, the sites we run that were not impacted much, no section/category index pages, no link farm bulk links to section index pages, both have unique and valuable components of the site that do not exist on any other website.

In a sense this is, as claarky notes and leosghost reinforces, just a confirmation of what I have always known and practiced on my own websites, and what an seo can NEVER do for you at any price, ie, actually makes something worthwhile for your visitors. It's not hard to figure out why, if they could do it, they'd do it, and wouldnt' be seos. There is one exception, when seos write good stuff on seo, but that is not something that you can buy, Danny Sullivan is his own product, and he sells it to you. Just like any other good site.

I don't read routinely any website because of seo bringing me there, I read it because I was following other links and finally realized I was ending on the same site all the time, and that is a quality signal that is obviously something an algo can work with.

In now way however do I want to discount the abilities of black hat seos to find ways to emulate the behavior of real sites (can someone say: automated chrome surfing sessions...) and real users, they will solve these issues in my opinion, at least the smart ones will, and they will laugh, as they usually do, but that's not related to anyone's real sites, and you can't worry about that part that you cannot change.

The negative seo however is an issue I do worry about.

I should apologize by the way for just dropping in then vanishing, but I do not get paid for this type of work, I already know how to do it right, and have always known, since at least 2003, how to do it, so the only reason I have been forced to dip back into the world of seo sleaze etc was because we lost and continue to lose a major site, but that's our fault.

claarky, while incomplete and ignoring penguin, I think your thinking on this is quite good, about as good as anythng I've seen. I can't contribute exit/bounce page stats because frankly I do not get paid for my time on this garbage, but I do need to keep the client from falling into more seo traps in the process of figuring this out (and let met tell you, AVOID seos at this point, like the plague, the white hats have almost no clue what they are doing from what I can see, and the black hats are working on their new software and link networks, and I expect to see higher quality and more highly focused articles being generated by them soon.

[edited by: lizardx at 10:00 pm (utc) on Jun 28, 2012]

[grippo]

They're somehow getting the stats via the browser, not Google Analytics. Get that and lots of things make sense.

Just to point out that there is no "Exit Rate" from the side Google sees the pages. Google Analytics data can't be used, because it can be manipulated. Analytis is an API, you can do whatever you want with it. For example, inflate or deflate Exit Rate.

Having said that, you can discover a correlation between Panda and or Penguin and your Analytics Exit Rate, but there is no point trying to figure it out how Google measures or uses Exit Rate. They are apples and oranges.

About the browser and other Google products, anybody knows how the provice policy allow / disallow getting signals for search? I have no idea.