SEO Blog - Internet marketing news and views  

Is Google REALLY using bounce rates as a ranking signal?

Written by David Harry   
Tuesday, 09 December 2008 08:51

Myth Busting 101 - taking media musings as gospel

Search Engine LandOoo look… Search Engine Land is getting into the behavioural metrics game – and creating more headaches.

Now, I love the fine folks at SEL because it’s a steady stream of search goodiness and they’ve been known to promote the Trail now and again. And while I don’t know Eric (Enge), I have a healthy respect for much of his work. But there is a time and a place to even whine at this venerable publication and Eric. (since we recently gave Search engine watch and Bruce Clay a hard time on this). 

You see, we just turned the corner on the whole ‘rankings are dead’ path and things have tempered which is nice to see. This time out more behavioural has been broached with a theory of bounce rate as a (serious) ranking factor. This line of thinking, while logically sensible, can be problematic from the search engines point of view.

Bounce rates as a ranking factor

Do Search Engines Use Bounce Rate As A Ranking Factor? (SEL Article)

The ‘study’ as cited, from SEO Black Hat, has been done to death and there in no way was any form of definitive research... if anything it was contradictory. I have been writing about search patents relating to behavioural metrics for some 18 months and while there is plenty of evidence of interest from Google, that fact remains as one Monsieur Cutts has called them; they are dirty/noisy signals (even an interview Eric did with Matt contains such assertions).

Anyone remember peeps trying to inflate their  Alexa rank with bots? Pretty easy to do so with bounce rates to lower yours and kill the competition. It is a dirty signal at best in the regular index. Add to that, contrary to popular belief, not everyone in the world is running Google Analytics, ToolBar, Desktop etc... Another factor that further dilutes the signal as a serious ranking factor... and it goes on and on... What about peeps that find a quick answer? Bookmark it for later? Or folks that leave a browser open for hours (assuming time on site is in the mix)?

And what of cost? Each time one adds a new signal to the mix there are offsetting processing, spam detection and associated costs. At the rate some folks in the SEO biz are going with predictions, there are definitely some financial considerations that Google would have to look at – these are tuff times you know.

My point being, that it isn’t as simple as, “Hey let’s start giving more weight to behavioural signals” – there is much to consider. We shouldn’t be making such assertions as;

There is a real possibility that bounce rate is a significant ranking factor right now. Even if it isn’t, it is my opinion that it will be made a factor in the near future. Even if it is not a factor, and even if it does not become one, there are plenty of reasons to look closely a bounce rate anyway.” – Eric Enge; Search Engine Land

While I agree with the last part, the one should be working on this metric as a matter of good web mastering, stating it may be a significant factor (or some day, or not), is troubling coming from SEL. Too many people read this stuff like gospel and there will be reams of forum flies (and my clients) talking about said ‘significant’ ranking signal.

Are bounce rates a serious ranking signal?

Let’s get personal


Now, in a personalized setting? Sure, it could certainly be used to some extent there... but personalized search is all about the behavioural metrics.

I think what is the most troubling is how the SEO industry, generally speaking, seems to not only slowly begin to see the evolution of SEO, but makes bold statements instead of tempered, researched hypothesis. There are certainly changes in the search indexing and retrieval landscape, but not of the magnitude some prognosticators want us to believe. Does everything need to be a BIG story these days? Why can’t search engines be slowly integrating? It gives one pause to hope it’s not in the name of ego feeding or traffic whoring.

There are plenty of other entirely valid behavioural metrics that could also be used, some to an ever greater extent, so why not scream those from the mountain top while we’re at it? How about bio-metrics for personalized signals (some day)? The answer is simple, there are plenty of reasons why they are dirty signals and are at best, effective in a personalized search setting.

 

Can we keep it real?

Yes, my friends, search engines continue to evolve. This we do know, but there is little in the way of serious information out there to prove bounce rates are any type of serious ranking signal in the regular index results. My goal is simply to highlight not only some alternate concepts but also impart that we shouldn’t believe everything we read…. This happens too much in the SEO industry..

Eric, SEL… ya know I luvs ya, but how about not making my life more difficult? We simply must be careful what we say, there are those that take it as gospel. Make the posts clearer that this is nothing but hyperbole. It keeps the noobs from running around the jungle talking smack that ain’t jack. Just look at some of the chatter. Yes, behavioural metrics make sense, but more so in a personalized setting… not as a substantial signal in the regular index.

Note: Eric did at least mention Matt’s thoughts here and touches on usability (the real issue IMO) -

 

See this post for more details on bounce rates and behavioral metrics

 

Now… a TON or related reading here on the Trail for those that care about learning about some behavioural metrics and search;

Beware: Google is watching you
Query analysis confirmed as Google ranking signal
Taking off my tin foil hat
Ranking through behavioural metrics
You’re not paranoid, someone IS watching you

What every SEO should know about personalized search
Personalization through application focus
Personalized search and behavioural metrics
the Art of of personalized search optimization
Yahoo personalized PageRank

And since the community is so sensitive these days, understand this is no slight on SEL et al; just trying to keep some perspective….  Peace

 

Comments  

 
0 # TheMadHat 2008-12-09 11:23
You really think they're not using toolbar data to validate the quality of a site? Bounce rates and other behavior metrics are going to be fairly consistent across an industry IMO anyway. Take this data and combine it with feedburner, and the 45 other enourmous data collection points they own and they've got a pretty large amount of data. Not saying they use bounce rate, it is a somewhat dirty signal, but the possibility is certainly there when you look at the big picture.
Reply | Reply with quote | Quote
 
 
0 # Dave 2008-12-09 12:07
Hey, as you know I have been writing/talking about behavioral signals for nearly 1 1/2yrs now... so I am all about the angle.. and the tin foil hats.

I am simply saying that there are many barriers to efficacy that make it, at best, a noisy signal in the regular index. That doesn't mean it isn't a signal nor used in a personalized setting moreso than Reg Index ....

What I think is most disconcerting, is that droves of folks will interpret the SEL post as gospel and scream it from the mountain tops. This mean I end up trying to bring a realistic view to clients that come to me with, "Did you hear google is using bounce rates? What do we do"

The big fish stories tend to grow from the source right?

And what of query spam? There are simply too many problematic areas for this to be a meaningful signal in the RI IMHO

he he... but I have been wrong before - more testing then shall we?
Reply | Reply with quote | Quote
 
 
0 # Brian Combs 2008-12-10 11:47
TMH,

I do not agree that bounce rates are consistent across an industry. They can vary wildly on that factor.

Type of site has a much greater correlation. For instance, with a query for "search engine optimization", a blog or wikipedia entry is going to have a much higher bounce rate than a website for a search agency.

Are you saying that, all else equal, the agency is the more relevant site?
Reply | Reply with quote | Quote
 
 
0 # RedEvo 2008-12-11 14:20
.....they might be using bounce rate but then again they might not be. Also, it could be a good signal but then again it might not be. It's a signal that's easy to measure but at the same time it's dirty.

Thank goodness that's all been cleared up....

d
Reply | Reply with quote | Quote
 
 
0 # Edgar Granados 2008-12-15 08:38
bounce rate does not affect the positioning, if you view the graphic of seoblackhat, you'll see that it is the opposite.

http://edgargranados.es/tasa-rebote-afecta-posicionamiento-serp/13/
Reply | Reply with quote | Quote
 
 
0 # Leo 2008-12-24 17:14
One way to use something like this would be to establish broad benchmarks across niches and then use that as a factor in behavior metrics. In other words, they could simply take an average bounce rate in the niche and measure that way.

At any rate, high bounce rates coupled with time on a site in most cases indicats that something is wrong.
Reply | Reply with quote | Quote
 
 
0 # Marble Host 2009-03-28 02:57
It is amazing how much energy people expend attempting to get out of work, had they just buckled down and done the work it would have required far less effort :unsure:
Reply | Reply with quote | Quote
 
 
0 # Marble Host 2009-03-28 03:00
The big problem with all your suggestions is they require work. Your blog would be much more popular if you could fix that little issue :-)

Seriously, it
Reply | Reply with quote | Quote
 
 
0 # ziv 2009-05-20 14:59
I am sue the br affect the serp. se are smart these days.
Reply | Reply with quote | Quote
 

Add comment


Security code
Refresh

Search the Site

SEO Training

Tools of the Trade

Banner
Banner
Banner

On Twitter

Follow me on Twitter

Site Designed by Verve Developments.