Returning Visitor Rate: Could That Be A Factor?

[Planet13]

There has been a lot of speculation on these boards about about bounce rate possibly affecting SERPs. In the past, google has said that was a "noisy" metric.

Is it possible that google is instead placing an increased emphasis on returning visitors? (I apologize if this is a naive question.)

At first, I thought that would require google to have inside knowledge of server stats (such as through google analytics data).

But, if google is able to do "personalized search results" then surely they must be able to track how often a visitor returns to a particular domain when using google search, right?

So despite having thin content, maybe ehow still ranks highly because searchers who have found and visited ehow once will often visit subsequent times (because they are familiar with it or like the format that ehow presents the information or whatever...) when ehows pages show up in the search results?

We all agree that google will use the clickthrough rate in the SERPs to influence position, right?

Perhaps there is added emphasis on RETURNING VISITOR clickthrough rates? Maybe the clickthough rate of returning visitors counts more than that of new visitors to a site?

Again, I apologize if this is a naive question, but it seems more plausible than many of the other Post Panda theories being floated around.

[chrisains]

Hi Planet13,

Interesting question. I wont claim to know the answer but I think it is a highly speculative question. Personally I don't believe Google or any other search engine could EVER use 'returning visitor' stats as any kind of ranking indication. For starters if they did then they would have to make various exceptions for different types of website which algorithmically would be extremely difficult!

For example: i visit bbc.co.uk far more than ebuyer.com but they're two completely different websites so it's perfectly feasible. How would a search engine gauge the difference between the two separate websites effectively enough to influence search ranking for the different industries? Just because I visit one less frequently doesn't mean its not equally important to me as a user, it merely serves a different purpose - I cant see how that can be determined algorithmically but I guess anything is possible!

As a more relevant example: I'm trying to use this WebMasterWorld forum whenever possible, but again its completely irregular. I'll post a few responses tonight then may not post again for a week or a month. I am therefore a 'returning visitor' but its on a irregular basis so how would a search engine apply any level of relevancy to this?

Finally, I'm sure there are millions of people like me who have visited a website, purchased a product, and never visited the website again. The fact that I am not a returning visitor should not bear any impact on the visibility of the given website within search because it has served its purpose. If I had entered the website and left almost immediately because it does not serve its purpose, thus increasing the bounce rate, then it is feasible that bounce rate could/should be a factor.

Also what about the people with personalised search "disabled"? At least bounce rate is a measurable factor as 'theoretically' it can be calculated from each individual visit; whereas if personalised is disabled then it is unmeasurable.

Personally I do not see how a search engine would ever correlate returning visitors with search rankings, but as I said I don't claim to know with absolutely certainty.

I'll be interested to hear other opinions on this...

Chris

[Planet13]

Thanks for your opinions, Chris. Would love to hear what other people think.

[TheMadScientist]

How would a search engine gauge the difference between the two separate websites effectively enough to influence search ranking for the different industries?

Relatively ... Meaning relative to other results shown for the query, the same as they do for many other things.

Returning Visitor Rate: Could That Be A Factor?

I would say it's possible, but not something I would overly concern myself with, because it's one of those things that 'happens naturally' when your site is the correct answer for visitors ... It may be too noisy, but I think it is theoretically possible (to the point of 'likely used in some way' even if not a direct ranking impact) but the only way to know is to have access to their data.

[tedster]

One metric Google could use would be navigational searches. If people are entering your domain name in the Google search box, that would be a very positive sign that the site is seen positively - and it's also related to return visitors.

[Planet13]

One metric Google could use would be navigational searches. If people are entering your domain name in the Google search box, that would be a very positive sign that the site is seen positively - and it's also related to return visitors.

Brilliant!

That also helps to explain a bit about the "boost" that brands seem to get.

[TheMadScientist]

Definitely, and the best way I can personally think of to 'manipulate' either is to build a great website and 'be the right answer', because then imo you get both the return visitors and the navigational searches. IOW: I don't really focus on either specifically, but try to build a site they both happen with.

[crobb305]

One metric Google could use would be navigational searches. If people are entering your domain name in the Google search box, that would be a very positive sign that the site is seen positively - and it's also related to return visitors.

That could certainly be a way to identify household names (i.e., "brands"). But for that matter, some brands have been successfully identified and seem to be forbidden terms for use in Adwords. I can search for a major bank, for example, and see no Ads on Google except those of that bank. Alternatively, I can search affiliate domains and see dozens of advertisers bidding for those. What is being used to determine brand/Trademark? Are they comparing against a Trademark database? If so, that same list could be used for branding in search. And, does a DMCA to Google requesting that your domain be prohibited for use in Adwords bidding get your brand/domain added to a list?

[johnhh]

DMCA to Google requesting that your domain be prohibited for use in Adwords bidding get your brand/domain added to a list?

thats interesting as I think so

we once had adwords based on, for example, super widgets - get your super widgets here

These were banned as the owner of superwidget.com objected - even though individually the words appear in any dictionary

[crobb305]

I guess technically it wouldn't be a DMCA, it would be a trademark infringement (even a common-law trademark infringement). I wonder if it would get Adwords to recognize the name, then feed into search as a brand. Not to spend too much time on this, but certainly might be one way to get it established. I have a 10 year old domain/site and people are still bidding for its domain name in Adwords. That shouldn't be happening if they are successfully identifying brands.

[tedster]

That could certainly be a way to identify household names (i.e., "brands").

After any website has been online and pleasing visitors for a period of time, in my experience it is common to see some search traffic coming in using the domain name as a search term. Many people do not even know there is a difference between a navigation bar and a Google search box. In some browsers today the navigational text box IS also a search box.

I don't think it's essential to have a registered trademark for Google to see your domain as a "brand". All that I think needs to happen is for your domain name to be identified as a "semantic entity" in its own right. Navigational searches, unlinked citations, social sharing - all kinds of effects that Google can measure would establish this.

[Shatner]

@tedster So do you agree that Google measuring the searches for a brand name might be used as a metric?

That makes a lot of sense. I'm thinking of getting on AdWords and applying that, see if I can find a pattern wherin a large percentage of the sites Pandalized have a lower number of volume searches for their name than similar sites in the same business who weren't Pandalized do.

[tedster]

I think that Google, among other things, is looking for the signs of successful and healthy engagement in the marketplace, and for social awareness of the site on the part of others. Pure MFA sites rarely show those signs, so I do think this might be one signal that Panda picked up on.

[Planet13]

All that I think needs to happen is for your domain name to be identified as a "semantic entity" in its own right.

That is very interesting. Lets look at amazon for a moment. Probably in the early days people would type amazon.com or www.amazon.com or ww.amazon.com or something like that into the search boxes.

Then would google LEARN that the site amazon.com was somehow associated with the word amazon? (Despite the fact that amazon.com had nothing to do with the river of the same name - or with muscular women?)

I'm not saying that was the only way that google would associate the brand amazon with a semantic entity, but could it have been one of the clues that hastened it?

[brinked]

My site that was hit had a 18-22% bounce rate. Since the official panda announcement, my site has a 30% bounce rate.

Also, searches for my domain name and brand name have increased over 400% which means these people are trying to find my website (google are you listening?)