just a thought. I'm curios.

If a large number of people did a search for a specific key phrase and kept clicking on a result that appeared in page 4, for a certain period of time, could it cause the site to move up in the Google results? maybe increase PR?

for all i know this could be the dumbest question in SEO history but i still want to know the answer.

It's certainly not a dumb question and something which I'm planning to run a test on. Google's click tracking on natural search results is there for a reason, and it would a sensible assumption that CTR can alter your ranking. You'll probably need to run the test on a term which isn't too competitive and automate the clicking through at least 100 different IP addresses. It would also be interesting to run a second test where each of those 100 IPs has its own Google account and logs in before the clickthrough.

thanks evilgreen, i've been planning on running a test on this myself..only prob is limited resources, (i.e IP addresses)...i'll think of something..in the meantime i was wondering...

I know everyone talks about linking as THE most important thing in SEO. The way i figure(my logic as warped as it may seem), google places weight on relevance, keyword density, keywords in titles, keyword links etc.... If so a large number of people clicking on a specific result for a specific term HAS to make a big difference.

I'm relatively new to SEO and if I thought of this, the BIG guns must done so ages before me and probably tried it out. This method,(which i believe is a black hat SEO technique) if successful is quite similar to "google bombing", wouldn't google have picked up on it. Its just i can't seem to find any clear info on the net regarding this.

You might want to do a couple of searches on 'usage data' + 'SEO' etc.
Some people believe usage data will eventually become as important as link authority..or at least be a strong 2nd factor. What's gonna happen nobody knows for sure, but I've also read opinions, that it won't become the most important factor, as it's just too easy to manipulate (the way you suggested for example :-)). I think letting the users judge through their actions could be a great way to improve an engine's algorithm, but if it should really be too prone to manipulation, they probably can't depend on it too much.

Usage data is not only the click through rate, but also the bounce rate(exp?), etc. (I dont know if thats already an important factor, though..if at all). If few people click through to your site, chances are it seems irrelevant to their search query. If many people leave the site right after clicking through (bounce rate), that also means it isn't relevant to their search query..or well sucks.

However, I don't believe that CTR analysis in the SERPs is a very important factor, yet (Im just saying, not overly important, now. Im not saying that they done use it):

I know a couple of SERPs, that have absolutely irrelevant results in their top10. Simply because the company name happens to have the keyword in their name, though the company operates in a completely different field (one was for example about 'plastic surgery', but had absolutely nothing whatsoever to do with the keyword). When I ran a search on them..I would have never clicked on any of those results/links I have in mind, because they were not even remotely related and it was very clear in the title tag, that they weren't. And I cant see many (any) users clicking through.

Actually, I believe by using CTR analysis for these SERPs they could definitely make them more relevant by weeding out such results, that aren't even remotely related to the search query (but happen to call their company similar to the query). But apparently, they're not doing it in a very strong way, yet or else those results should be weeded out of the top10 (those are not totally uncompetitive keywords and they seem to have roughly the same link authority as other top10 sites, yet their CTR should pale in comparison).

P.S.: Keep in mind, that Im not trying to say theyre not doing it, at all. I just believe its not nearly as important as link authority (,yet?), but can see usage data becoming more important in the future, too.

I would be interested in hearing other opinions on this, especially yours evilgreenmonkey. You really seemed to be interested in this question, too and are obviously a lot more experienced in this industry than I am. Do you (or anyone else) think, that my theory is valid?

Actually, I'm really wondering why Google (or other SEs) wouldn't sort out such sites, that most likely have extremely low click through rates (most people probably won't click through to a site in the title of which it says plastic surgery, cosmetic surgery, etc. after searching for something totally different)..maybe at the end of all other computations.

Maybe the computation would take up too much time and they don't see enough value in it in order to employ this in their algorithm?

I assume Google should have enough data to see if a site has an extremely low click-through rate for its position in the SERPS. But maybe this assumption is wrong?

I believe Google also tracks the amount of time a visitor stays at a site, not just the click rate. Perhaps you can increase the ranking by more clicks from different IP addresses. But if the visitors don't stay long, that can make Google think the content is not very relevant and end up dropping its rankings.

@gobeyond:

Yep, I agree with you:

Usage data is not only the click through rate, but also the bounce rate(exp?), etc. (I dont know if thats already an important factor, though..if at all). If few people click through to your site, chances are it seems irrelevant to their search query. If many people leave the site right after clicking through (bounce rate), that also means it isn't relevant to their search query..or well sucks.

If users leave the site too quickly, that might not only be an indicator for the site being irrelevant but also for simply being of bad quality/no useful content.

However, the click through rates could be great for weeding results out of the SERPs, that rank (because the company uses the keyword phrase as a part of their company name (Ive seen such top10 rankings quite a few times)), but are absolutely irrelevant to the query.

However, the question is how important are click-through rate & length of stay on a page/site in Google's (or Yahoo's, etc.) algorithm at the moment? And how important will they become. Right now, at least the CTR doesn't seem to be a very important factor, yet (at least imho).

It seems to reason that if engines track click rates for paid that they would also track natural clicks.

After all, nobody visits search engines to click on ads!

It seems to reason that engines would also track time on page and/or back button referers, if for nothing else than to find sites with "sneaky redirects".

So, I think click data is important for both natural and paid!

This is a good topic. I would like to think that CTR, page visits, and web site visit length either plays a part in natural listings or could possible play a part in future natural rankings. I say this because user data - placed against search phrases certainly is could be defined as a factor that revolves around the relevancy debate.

Someone brought up the fact that user data could easily be altered or rigged, this is currently absolutely the case (i.e) click fraud, but I would think the logically answer to that is to have the SE algo pick up on this type of suspect activity.

Just my opinions.

100 different IP addresses. It would also be interesting to run a second test where each of those 100 IPs has its own Google account and logs in before the clickthrough.

Me thinks if you have a google a/c and your logged in then it would be easy to follow your search history to work out what theme of search you do and if your history can be "trusted" to count CTR..

Could be a similar algo to PR .. The more votes the more trust worthy...

Might be worth only visiting pr 9's + and your pr1 .. see what happens then..

True, I think the numbers would have to be way off before any flags went up! I see this as like a second line if you will, used mainly to catch spammers.

I would like to think that CTR, page visits, and web site visit length either plays a part in natural listings or could possible play a part in future natural rankings. I say this because user data - placed against search phrases certainly is could be defined as a factor that revolves around the relevancy debate.

I think it's very likely, that it will play a big(ger) role in the future. I've heard this from more than one source, actually and it does seem very logic.

Someone brought up the fact that user data could easily be altered or rigged, this is currently absolutely the case (i.e) click fraud, but I would think the logically answer to that is to have the SE algo pick up on this type of suspect activity.


I believe, that was me. I think it really all comes down to how much more effective usage data can make their rankings relative to the effort of doing this. For example, I don't doubt, it will help them make their algorithms more effective and that they CAN pick up on that type of suspect activity as you said. But overall will this effort be justified by the increase in relevancy? Or will they notice, that maybe it would be better to spend their money on tweaking their link analysis algorithms to identify quality/spammy links?

I believe Google also tracks the amount of time a visitor stays at a site, not just the click rate. Perhaps you can increase the ranking by more clicks from different IP addresses. But if the visitors don't stay long, that can make Google think the content is not very relevant and end up dropping its rankings.

Now, i'm not super technical when it comes to the back end programming of websites and the internet, but I do know a fair share of programming and development. So exactly how would Google (or any SE for that matter) be able to track how long a visitor stays on any given site?

I guess if they used data collected by G Analytics, but not everyone uses that and personally, that would just give me one more reason to drop Google analytics.

I can't think of a way that an external website would be able to track user movement from another website unless it was given strict permission to do so. And that would most likely require some sort of back end programming or script to do the job.

Last time i checked, Google didn't have any permission from any of my sites.

And as far as click data, I had a client a couple years ago who was hard pressed to believe the same thing as all of you and decided to have his entire web development department (22 people) click on their organic listings on a daily basis. I think he assigned each person to click at least 20 times on each of their top 25 keywords. The results over a year of this insane waste of time.... minimal if anything at all.

I guess if they used data collected by G Analytics, but not everyone uses that and personally, that would just give me one more reason to drop Google analytics.

And as far as click data, I had a client a couple years ago who was hard pressed to believe the same thing as all of you and decided to have his entire web development department (22 people) click on their organic listings on a daily basis. I think he assigned each person to click at least 20 times on each of their top 25 keywords. The results over a year of this insane waste of time.... minimal if anything at all.


Point 1. You would be surprised how many people use or have used Google analytics.
Point 2 That is very scary, that 22 people click natural listings (25 keywords) for a prolonged period, just plain scary!

Thanks for your impute!

And as far as click data, I had a client a couple years ago who was hard pressed to believe the same thing as all of you and decided to have his entire web development department (22 people) click on their organic listings on a daily basis. I think he assigned each person to click at least 20 times on each of their top 25 keywords. The results over a year of this insane waste of time.... minimal if anything at all.

I dont mean to be a wise-guy, but I only said, that I believe there's a chance usage data might become important in the future, but that I didnt think it was very important at the moment and that I couldn't tell, if it would ever really become important. And I think a few others share a similar belief?

I've heard they were using things such as Google Analytics & the Google toolbar(?), etc. in order to try to track usage data. However, I don't know enough about this. I've heard microsoft was at an advantage, if user data should become important, though as they own the Internet Explorer. Could microsoft gather tons of usage data very easily that way?

That being said, if usage data finally becomes an important ranking factor, I would definitely not ask 22 people to click on all of my keywords 25 times a day, but try to influence users in a more indirect way (just as now you have viral/marketing or link baiting in order to get links in a natural looking way).

I guess, that would look a bit suspicuous and be a pattern, that could be easily detected by Google, as they should be smart enough not to make it an important ranking factor unless, they've found a way and taken precautions in order not to get their search results spammed and become irrelevant.

All The Messages I Have Readed Are All Right.

Similar to the prior post about tracking time on a site.

I am unclear as to how google might be gathering click data for organic search results. As far as I can tell google passes the actual link to the site back in the search. Some search engines pass a link back to themselves and then redirect to the site (in which case they can capture the click data)

It doesn't look like google is currently doing this.

I don't think google would base rankings on click popularity, unless they had a good way to measure it without people trying to abuse it.

JP

Google could monitor clicks with their toolbar.

So watch out!!!

Google could monitor clicks with their toolbar.

So watch out!!!


When you "say" clicks, do you clicks from the SE to the site or do you mean visitor clicks on an independent web site? Clairification would be good, thanks.

I was referring to clicks from the search engine query results to the site itself.

Although, as Steven D mentioned the google toolbar could be used to track all kinds of "Unscruplous" things.

Pehaps I'm too trusting but I think they would follow their own privacy policy for the toolbar

http://www.google.com/support/toolbar/bin/static.py?page=privacy.html&hl=en

Sorry, for clarification, I meant the clicks from search engine -> chosen website.

Knowing google, I think they would not readily use that "clicking" information, but experiment with it first. Later they may use that clicking information for both improving their program and for ranking. It is unknown in what extend they use that information thought.

I think they would not only record which website the visitor visited, but also how long the visitor stays on the website, indicating how succesful the click was (this website is a GREAT website, because visitor left only after 30 minutes, please give it a greater PR).

If you see their privacy notice:

The Google Toolbar automatically sends only standard, limited information to Google, which may be retained in Google's server logs. It does not send any information about the web pages you visit (e.g., the URL), unless you use Toolbar's advanced features. You do not need to provide any personal information in order to download and use the Google Toolbar.

In other words: The google toolbar sends standard information to Google, that is retained in their server logs. It sends information about the web pages you visit unless you don't use the advanced features (but what are the advanced features???). But google does not ask for personal information.

In other words: they record and track you anonymously.

It's brim language.

Google is not using time on site as a measure... since they only have the sites using GA to determine it and it can be slanted.

CTR of organic may be another story but little is really known about it. Having your employees do it from home is like Digg - they see all the diggs coming from one city and realise this could be happening.

However, if a site is getting global attention then it is possible to impact.

While I've never heard of this taking place, I would assume that Google uses some form of click tracking to see which links are getting the most play. I'm not sure if your time is best spent simply clicking on your link as I think it would require a fair amount of work to make this technique work.

I think that the best option for a site is to build high quality deep content on the site, and make sure to focus on your title and meta description tags to help influence and drive more clicks to your pages.

Make sure to use the tags on your pages that make the search engines ignore your OpenDirectory and Yahoo Directory listing information as well. This is one more step that will guide search engines towards using your description tag as the content in their results.

I'd love to hear results on this if someone took the time to do the test, just not sure how valuable this activity would be for the time that would need to be put in just to get the results.

Google once told the company I was working for via email that CTR of organic results affected Adwords Quality Score. ie. If a certain ad was clicked on less because the organics results were more popular then this would raise the cost of said ad.

I posted this ages ago:
http://forums.searchenginewatch.com/showthread.php?t=14167

But couldn't actually provide the email they said it in. Or any other online reference. so its pretty anecdotal.

IF they are using that data for Adwords then they are probably using it for the Organic results to some degree??

I'm almost certain that the Google SERP's are influenced by the CTR, the time that is spent on a website,...

For now, Google can obtain information from the data gathered by the Google Toolbar. On the same page that was mentioned before in this thread (Google privacy notice (http://www.google.com/support/toolbar/bin/static.py?page=privacy.html&hl=en)), Google also states the following : "In addition, we use log information about aggregate Toolbar usage to improve the quality of Toolbar and other Google services." Other Google services ? Wouldn't it be strange if they didn't use it for their main service, the Google Search Engine ?

In the future, they might combine the Google toolbar data with data gathered by Google Analytics. A few months ago we talked to one of the European Google Analytics specialists and we asked him if the data from Google Analytics could influence the SERP's. He replied that "for now" that data is not used in the Google algorithm. You might oppose that not all sites are using Google Analytics, so how could Google use the information in a correct way ? If two sites in the SERP's are using Google Analytics, and one is performing a lot better than the other one, maybe the first one should rank above the other one.

Maybe a little far fetched, but I don't think it's impossible...

I think they would not only record which website the visitor visited, but also how long the visitor stays on the website, indicating how succesful the click was (this website is a GREAT website, because visitor left only after 30 minutes, please give it a greater PR).

If you see their privacy notice:

The Google Toolbar automatically sends only standard, limited information to Google, which may be retained in Google's server logs. It does not send any information about the web pages you visit (e.g., the URL), unless you use Toolbar's advanced features. You do not need to provide any personal information in order to download and use the Google Toolbar.

In other words: The google toolbar sends standard information to Google, that is retained in their server logs. It sends information about the web pages you visit unless you don't use the advanced features (but what are the advanced features???). But google does not ask for personal information.




Live Header of a GET request by the Google Tool bar.

http://toolbarqueries.google.com/search?sourceid=navclient-ff&features=Rank&client=navclient-auto-ff&googleip=O;64.233.169.99;160&ch=8459d5eaf&q=info:http%3A%2F%2Fforums.searchenginewatch.com%2 Fshowthread.php%3Ft%3D17879%26page%3D1%26pp%3D20

GET /search?sourceid=navclient-ff&features=Rank&client=navclient-auto-ff&googleip=O;64.233.169.99;160&ch=8459d5eaf&q=info:http%3A%2F%2Fforums.searchenginewatch.com%2 Fshowthread.php%3Ft%3D17879%26page%3D1%26pp%3D20 HTTP/1.1
Host: toolbarqueries.google.com
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.4) Gecko/20070515 Firefox/2.0.0.4
Accept: text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5
Accept-Language: en-us,en;q=0.5
Accept-Encoding: gzip,deflate
Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Keep-Alive: 300
Connection: keep-alive
Cookie: [snip]

HTTP/1.x 200 OK
Cache-Control: private
Content-Type: text/html; charset=ISO-8859-1
Server: GWS/2.1
Content-Length: 0
Date: Fri, 01 Jun 2007 21:38:49 GMT

http://www.seobook.com/

Take a look at the 'data collection: Google's biggest competitive advantage' post.

Today Google bought Feedburner, which (along with AdSense, AdWords, the Toolbar, Analytics, user accounts, Google Feed Reader, Google Checkout, Youtube, etc) is yet another source of data acquisition for them.

Google once told the company I was working for via email that CTR of organic results affected Adwords Quality Score. ie. If a certain ad was clicked on less because the organics results were more popular then this would raise the cost of said ad.

I posted this ages ago:
http://forums.searchenginewatch.com/showthread.php?t=14167

But couldn't actually provide the email they said it in. Or any other online reference. so its pretty anecdotal.

IF they are using that data for Adwords then they are probably using it for the Organic results to some degree??

If they are clicking the organic results your CTR for ads is lower per impressions.... nothing more sinister than that.

If they are clicking the organic results your CTR for ads is lower per impressions.... nothing more sinister than that.

Ooooooo.. click on your own ads as well.. hehhehehehe

</sarcasm>

So exactly how would Google (or any SE for that matter) be able to track how long a visitor stays on any given site?
I wondered when someone would ask this on this thread. Lets have a look at some possibilities

The google toolbar is the easiest way for them to do it. There's enough people with it installed to get a good sample of data.

A redirect page - I've never seen one on G, but occasionally I see it on Y

Session time and click paths - How long did a user spend looking at the SERPs before going to page 2, or try another query?

Javascript - It's very easy, with javascript, to work out the co-ordinates of a click on a page, marry it to a screen size and get implicit click-location data

I'm sure there are other means, but these are just the ones I thought of off the top of my head in few minutes. Remember - no one but google knows for sure and "if it is knowable, assume it's known"

I am waiting for "Impression Fraud" to get a mention....

just a thought. I'm curios.

If a large number of people did a search for a specific key phrase and kept clicking on a result that appeared in page 4, for a certain period of time, could it cause the site to move up in the Google results? maybe increase PR?


Are you all daft? Isn't anyone going to address the morality of this issue?

What you are planning is basically an intentional way to increase your website's ranking on the SE's. There is no more blatent exercise in Black Hat SEO than that!

Do you REALLY think G would make it that easy? And what do you think they'll do to your website when they've traced all those clicks to one organization?

Try it - I dare you. It may not happen right away, but G WILL eventually catch up with you and will ban you from their SERP's altogether. Then where will you be?

Instead, why not try improving your CONTENT!!!!!!!!!!!

Are you all daft? Isn't anyone going to address the morality of this issue?

We Have!!!!

Try it - I dare you. It may not happen right away, but G WILL eventually catch up with you and will ban you from their SERP's altogether. Then where will you be?

I agree Google would catch on and ban you from the SERPS, WITHOUT A DOUBT.

Instead, why not try improving your CONTENT!!!!!!!!!!!

I admit, I have seen some pretty black hat stuff out there and Google seems to be turning the other cheek. I am sure everyone has seen traces of BLACK HAT, when looking for it.

I agree with Page1Rank. If you improve your content, let's say you host a news website. And you write an extensive article about say, Paris Hilton in prison; It will attract people. They will click on your site more often. This is the only good way to improve pagerank. After all, you now increase the gained benefit from the website visitor and you make google happy, because your site really deserves that increased page1rank; because your site has good content. Google wants sites with good content be ranked higher then sites with bad content ofcourse. By doing so, google wants to satisfy their customers (the visitors).