Ever since Panda, most good SEOs have been drawing people’s attention to a few factors that according to the new algorithm changes, have greater importance as ranking factors now. Two of those are bounce rates, and time on site.
If you missed all that talk, we’ve basically said that now, Google is looking more and more at how people interact with your website in order to determine it’s quality or lack thereof. So Google wants to see how long visitors are spending on your site (longer indicating greater quality than shorter) and at what rate people “bounce” from your site, leaving it for an outbound link or hitting *back* on their browser to return to the search results.
Google cares more about that stuff now. But how much do they really care? An accidental case study we did recently gave us an insight…
Really what I’m about describe is remarkable. Even if you’re an experienced SEO, this should surprise you. Here goes:
Our Accidental Experiment
We set up an example website for our Unstoppable Affiliate training course. It was one we set up just for people in the course, just to show them exactly how we created our sites. It wasn’t supposed to be a site we had that was already making money, just an example of exactly what we do in a real scenario.
We set it up targeting a real affiliate offer and with our real aff links in there, real (quality) content & monetization… the works.
But we didn’t start SEO on the site. It wasn’t meant as an example of our SEO campaign, just of the site layout & structure. I’m not aware of any backlink we built to it at all.
This was a NEW domain name too.
Fast forward about a month…
We noticed the site had made a couple of sales.
It didn’t surprise us much. We knew the market it was in well (we’d been in it before) so we knew the offer converted. We figured we’d somehow picked up rankings for a long tail keyword and had pulled a couple of high converting visitors just with our quality content and by having the keyword in the domain name.
But then things started to get weird…
Sales kept coming.
In one month the site made something like $300. We still hadn’t built a single backlink.
In November so far, the site’s up to $583 earnings, it’s bringing about a sale a day, and there’s still a week to go in the month.
We’re ranking on page 1 for the main keyword, plus a host of others, and bringing in a great volume of traffic. We’re outranking sites with hundreds of backlinks that have been around for much longer than ours.
Oh, and we still haven’t built a single backlink.
Question: How the HECK is this possible?
Now, I can’t confirm this, but here’s our hypothesis… (keep in mind it’s easier to analyze what’s had an effect here because we’ve done so little on this site… there’s not much to choose from!)
We know that Google takes data from web surfers. Not just data about your Google searches but about your browsing habits in general if you have Google toolbar or you use Chrome. Probably in other ways too.
And we know that data about how people interact with your site is an increasingly large part of the judgment of quality of your site.
When we listed the site in UA, it got a lot of traffic. People from the course went to our site and checked it out. Google saw that.
What’s more, they didn’t bounce, and they stayed on the site for an average of 2 minutes 30. A pretty long time for a site like this (and in this market given it’s nature). People were checking out the site to see what we did with it. Of course they stayed a long time!
That’s the only thing that’s happened to the site (including it’s solid content, on page optimization and a keyword in the domain) that could warrant a ranking. A site with JUST solid content, on page optimization and a keyword in the domain is not ranking in this market after a couple of months. We’ve tried it.
So you make up your own mind…
NOW…
What Does This Tell Us?
Before you go out trying to recreate this scenario: don’t. I’m going to assume that this is an overcompensation towards these new ranking factors that’s a part of the algorithm now, and as more people game it, it’ll get leveled out. If in fact it is a real thing, (our humble one case study doesn’t prove much) it won’t last long.
But this does tell us something very important. Google is VERY serious about this bounce rate/time on site/user experience stuff.
There aren’t many factors that can compete with backlinks when it comes to determining your rankings, but this case study tells me that – at least right now – those experience factors can and are approaching that level of “weight” in the algorithm. Remember, our 3 month old site with no backlinks outranked sites that had all other factors equal (content, keyword in domain, on page optimization) AND were older and more established. That is a BIG deal.
What do you take from it?
Where you can improve your content, lower bounce rate, and keep people on your site for longer while still having them click your affiliate links… do it.
There’s a fine line there for affiliates. The longer people stay on your sites the longer they’re NOT clicking your affiliate links. So it’s a very fine balance to strike. But it’s possible.
We’re going to do more and more testing with regard to this and we’ll be sharing the results with you in the coming months.
In the meantime, I hope this provides you some food for thought 
Post to Facebook
Share on Twitter
Digg This Article
Save on del.icio.us
Interesting theory, It’s well known google counts user experience on your site as important but nobody knows exactly how much that ‘weight’ is compared to backlinks. I do think they are monitoring closely at users clicking the back button more than clicking out via an outbound link.
It would be make sense as there are a ton of websites with adsense that are rank no1.
Cheers Andrew,
Thanks for the insight… when it comes to this type of information I have to split my brain (what is left) I own 6 ecommerce sites where the mindset is to get the customer off the home page and on the money page asap.
I also have 5 affiliate sites where the home page has the “money” links … so if you get your customer excited and they click off your home page headed for the shopping cart (clickbank or ecommerce cart) is that action going to reflect a bad bounce rate? Or does the search engine take into account they made a purchase from you because they liked what they saw (good user experince) rather that just exit the site.
This is a classic example of “the more I know the more I know I don’t know”….
Thanks…
Sounds pretty cool… There is a piece of software called SE Sniper from Casey Costello that searches your chosen keywords and than finds your site in Google or a whole list of search engines and clicks on it. But, with what you have just shared (your always sharing) the best part about this software is… You can set it to STAY on your site for as long as you want before it leaves and starts the whole process over again. I’ll give it a test of 30-40 minutes and let you know. By the way, this is not my piece of software, nor do I know the guy that created it. Just thought it might be useful to you and your readers.
Cheers,
Mack
My website has been jumping in and out of rank two for its super competitive keyword and maybe that is the reason for it. My visitors have only been staying on the average of one and a half minutes. Two minutes tops.
Should I put my affiliate links only at the end of the blog posts?
Hey Rothman,
You can try making changes like that and see how they affect your bounce rates. Just monitor it in Google Analytics to make sure you can definitely see the changes if there are any.
Andrew
Thank you. I will do that and see if my visitors stay longer.
Steven B brings up a good point that needs to be tested. Is it time on your whole site? Or just time on the homepage or any initial page they may land on? (few of my visitors on my bigger sites come in through the homepage!)
If you think it through, Google would want people to spend time on the webpage they landed on BECAUSE it gave them what they wanted, and that would factor into “quality”. But if that page simply linked to other areas of the site that had specific thngs they want, and time on it is short but time on subsequent site pages are longer? Hmm. I gotta wonder how fine-grained they’ve written this algorithm and how many factors it takes into account.
I think the bounce rate is determined from when you leave Google and land on a page to the time you return to Google. This will mean you don’t like the website or the website does not have relevant info for you and you hit the ‘Back’ button to go back to the search engine.
Correct me if I’m wrong.
If it’s bounce time Google is after now then your example isn’t fair as it’s automatically pulling in people who will spend time on it. That’s not the normal case. So, how about adding an appropriate and high valee youtube video to your page? People seem to watch these now and if they do it will guarantee them reducing your bounce rate. Add this to the standard SEO as well of course.
Barry & Stu – Definitely guys. That’s part of what we’re going to start testing from here on out! Will let you know.
Video. Will try that! Thanks!
totally cool Andrew,, thanks for sharing… I actually remember you talking about that a little while ago.. but the fact that it’s making you that many sales is awesome.
it makes me wonder, if this finding is true, then we could easily engineer our sites so they stay longer and click around more i.e .videos, better interlinking etc. something to factor in to the whole formula.
Stu
You’re right spot on Andrew… This isn’t in any way new; in fact there have been a few tools around the forums to fake visitors to our site, but I think is becoming increasingly important…
And not just for rankings; I’m seeing that Google is also increasing PR to sites with good traffic even if the have little or no backlinks… Very interesting stuff…
Thanks for the great info
Agreed. That’s why I put “New” in quotes like that
Oh Barry, video is a GREAT idea! ** goes off to create some vids . . . **
That’s not exactly what I meant. I am finding relevant youtube videos and embedding them in my posts. In the top half so it’s seen as soon as a visitor gets to the page. The videos I embed are relevant to the post and I add the youtube link at the bottom of the post as well. I’m certain Google can work out a youtube embed code so throwing in a link to it as well should help. Apart from the fact it’s unfair to the video maker not to.
Hi Andrew,
Sounds like you got a very pleasant surprise with the new site you set up. I have been tweaking my site to make it more search engine friendly. I just haven’t been posting often enough. I am going to do more starting now.
I downloaded your 6 part video series on seo and it gave me more information on tweaking my site.
Thanks, Bill
Thanks, Andrew, for sharing this info! The bounce rate thing is, like Steven B suggested, kind of a tricky thing. It does seem to make sense that if google knows how long someone is staying on a page, they would also know if that person clicks a link to a product site or clicks the “back” button or just exits the page. So they should be giving “good user” service kudos to the site where someone clicks a link rather than just exits, but……….? Again, I agree with Steven – “the more I know, the more I know I don’t know”…. But the journey is interesting! Thanks again for all the info you share, Andrew!
That’s true Karleen. Google is going to look differently at a person who clicks on one of your links and leaves as opposed to clicks back on the search engine and leaves. Those aren’t the same kind of “bounce” and they would be factored into the algorithm.
Thanks for sharing that. Big G is surely making everyone very busy with all these changes. I do agree that the time spent on a site should play a strong part in the overall SEO quality. I am happy too because for those fly by the night sites – they are in trouble
Don’t you just love it when you stumble across something like this when not even testing for it ??
To me it makes perferct sense and the best indicator possible for Google to measure what they should be putting in front of searchers eyes. If I was them, I would use this soley to rank sites…….
Having said that, new sites with no traffic would never get ranked if that was the case, so I guess they are never going to be able to rely on just this method of ‘social proof’ after all (certainly not for new sites anyway….. Top 10 players maybe ???).
Excellent stuff as always Andrew – now I’m off to that site you’ve mentioned to click on it for 1 second…. before hitting the back button
(just kidding)
Haha, I hope your data stands out like an outlier and they punish your Google account for behaving weirdly
Well Andrew, it certainly proves that by helping others we can actually help ourselves, huh. Thanks for sharing this. I will definitely be sure to pass on your news to my clients in hope of encouraging them to share quality info on their websites too, willingly!
Best wishes,
Trish
I think this is really interesting. Time will tell exactly how it works and google will probably have changed it again by then! but either way, just make the sites the best possible how we all know how and hopefully we’ll all do well!
thanks for an excellent course with UA too Andrew, will drop you an email soon!
Andrew, I quite agree with your assessment. I have been paying close attention to Google and other rankings. I won’t change the on page seo that I’m already doing. But I have noticed the direct correlation between my rankings getting a boost the more time my blog visitors spend on multiple pages. The new Google? Like you, I doubt it will last. So it is better to make this a part of your overall seo strategy, but not the sole focus.
Agreed Tim – and well said.
Clare – very cool. Look forward to hearing from you!
Trish – Can’t hurt!
Great anecdote, I wish more people would speak up when they see things like that.
If the keywords you were ranking for initially all had to do with phrases in your domain name, that’s not so surprising – see:
http://www.coconutheadphones.com/googles-secret-ranking-algorithm-exposed/
However, it *is* surprising that you got spidered with no links, since Google “prioritizes crawling roughly in order of PageRank” (I think Matt Cutts said that).
Maybe it prioritizes the home page of a domain regardless and that’s the only page that was indexed?
I’ve seen other sites with no apparent links get indexed in the past, presumably Google finds them through Chrome and Toolbar data.
- Ted
I agree with you Ted. I think it’s possible Google can find websites with more than just spidering links now. Like if a couple hundred people load up a URL using Chrome and give it a good browse… (I’ve no proof of this, just speculation) and it appeared in the index just because of that, I wouldn’t be surprised.
Hey Andrew, your speculation is right….read what a couple Google Geeks recently told Dori Friend….
Got a link mate?
Ted and Andrew….
I got this from Dori Friend who recently spent some time with a couple of Google engineer-brainiacs in Vegas across a Black Jack table….
oops got cut off. Here’s Dori’s story…
“So in-between the Google engineer-brainiacs teaching me how to play black jack (and I won over $500 that night!) l squeezed them on indexing techniques.
The G’ engineers said, “Actually, no need to do anything, we do it automatically for .com and some other extensions!” I said, “what are you talking about, how do you know where a site exists without a link?”
And as he look at me like duh and said, “we track .com and some other extensions registration, so once they are registered, we know it.”
Well my jaw hit the floor while I double down on a pair of Aces and thought, ok, I will just have to test that when i get home, (as I never trust what Google or Google people say! lol)
So, I tested it and yep, sure as sh*t it’s true.
Within 3 days of a .com or .org being registered (and a WP blog put on it) they are indexed! No other action needed!!
So, forget about all those other fancy ways or services to get your site in the index! That is soooo 2011! “
Andrew this makes me want to jump on to UA to see the site your talking about…
I haven’t seen how you presented this site but I would bet that along with staying on the site an avg. of 2.5 min. they may have also book marked the site for future reference as well (maybe even while signed into their Google account). I do this a lot and gotta believe Google takes note of it.
Besides getting visitors to stay longer incorporating some type of interactive form or questionaire is also great for conversions.
Great post, Andrew.
I’m a new learner in UA course, I was pretty surprise too when I searched the keyword of the sample website, and saw it ranked #1 in Google all the time. I was assuming that you guys were pretty serious in building websites even with one sample site like that, but when looking at the links and PR, I did have a lot of questions. And here it goes now the post to answer my questions
Awesome job!
I tend to think we’re pretty serious with building websites and doing SEO, haha
Hi Andrew, when I read your post, I feel like how easy for you to make money. Not like us, we still struggling.
Anyway, I am agree with you. I think google already doing this for quite a long time but maybe after Panda update they start to do it very seriously. Maybe they increase the percentage in the algorithm that bounce rate will contribute into relevancy of a website.
If this is true, then Google favors those who have Google Analytics installed on their site or use Chrome or the Google toolbar …
Hey Andrew why don’t you let us help you further improve your rankings and tell us ALL what the site is….(I promise to stay on for at least 3 min.! Ha!
This kind of confirms what I have been feeling for a while. Google has painted itself into a corner with backlinks as a site vote.
How can a site owner throwing backlinks at a site really be considered a vote. It sounds like what the political machines say in Chicago- “Vote early and vote often”
So Google must find other ways to evaluate site quality because backlinks are too easy to game the system with.
Google is smart and it will find other ways to work into its algorithm to determine quality and it will pick up on automated linking strategies and devalue those links too.
Another type of interaction that might be relevant here is blog commenting. I am often tempted to shut down comments completely because I’m sick of deleting huge quantities of spam every day, but I never do so because all those spammers are visitors to the site. They probably don’t spend a lot of time there – especially when they’re using automated software – but some of them do.
Any thoughts on this, Andrew? I know you’ve discussed switching off comments in order to limit the visitor’s options for exiting the page (i.e. to encourage them to click on our offer), and then of course there’s the on-page SEO issue of keyword dilution from irrelevant comments. But, in light of your findings, how valuable is the traffic?
As usual, you’re delivering the goods Andrew!
There is some really amazing content in your post and in the comments. Here’s what I’m walking away with:
==========
Decrease Bounce Rates:
- Add video
- Test lowering aff links on the page
Increase Site Load Speed
Install a site on new domains quickly (for indexing)
===========
Got any recommendations for increasing WordPress load speeds?
Perfect takeaways Dustin.
There are a handful of plugins for improving load speeds. Search the plugin database and you won’t miss them. Highest rated one should be best.
Andrew
Great post Andrew.
Google are definitely putting more weight on user experience.
I’ve noticed that when I click into a site from the SE results, if I immediately click back out, Google ask if you want to “block this site from the search results” therby giving a vote of no confidence in that particular site.
That’s good news for decent affiliate sites that are being outranked by forums etc, but how long will it be before shady affiliates start to game this feature, by blocking all their competitors?
Great bit of “new” information, Andrew, that makes total sense. Thank god for good content still having an important place. Easy to understand that is what will keep people “glued” to your website!
When you get to the bottom of it, of course quality content HAS TO be one of if not the most important parts of a website.
After all the internet is / should be about seeking and finding knowledge. Whether it be about something as basic as information about a product you are wanting to purchase or something as important as finding some different trusted information sources about a serious health problem. Selling or advertising “stuff” should be secondary to that, although obviously absolutely necessary to monetize the whole internet thing and thereby “grow it”.
What a”pain” it is when you are trying to do some quick searching and you are led “up the garden path” by crappy business directory sites that have nothing to do with your search except that they have a good SEO manager that can get them on the first page for almost any keyword you can think of; or you find an “aged” site full of articles / information written by the equivalent of a 10 year old. (But they know how to stuff them with keywords).
I may be a bit naive, but I can’t understand how a company of super geeks like Google actually take as much heed of back-links as they do.
Surely some of them hang out around in our “internet marketing world” and realize that internet marketers just prostitute the whole back-linking “thing”.
On the other hand you can’t prostitute good, quality content.
I love the quote from another commenter : “the more I know, the more I know I don’t know”…similar to my quote on my namesake website which reads “you don’t know what you don’t know”! ….. and thats what the internet should be all about …. finding and learning about some of that stuff you don’t know.
Your theory makes a lot of sense and I agree with you that it has a short shelf-life.
But…if you add, longer good quality content, good on page SEO, helpful video(s) or interactive tool(s) and ample back-linking, then I think you’ve got the secret sauce that Google is looking for.
That’s the test that is really asking to be run…
best – jessie
My 1st site, which was really set up as a bit of an experiment has only around 15 backlinks. I play around testing plugins on it and it fell out of site for 10 days a few months back, then when it returned held spots 1-4. It still ranks #1 worldwide for my main long tail. Other sites below it have 1000s of backlinks. I haven’t added new content since its inception. It doesn’t make me a fortune or anything, $50-$100 a month via an affiliate product and adsense, and only gets 3000+ visitors a month. But it certainly has held it’s position for over 6 months. I think that has more to do with good content, a well set up site and the SEO has been done right. I checked the average stay time, and it was only 46 seconds.
PS. I don’t use the Google Toolbar, or have analytics installed or use Google Chrome.
That’s great!
By the way, how do you manage to track your bounce rates if you don’t have Google Analytics installed?
I use StatCounter before when I was starting out but ever since I started using Analytics, I was hooked by the detailed reports of views and visits. But some marketers recommend NOT using Analytics, so I would love to know of other tracking/website analysis services (hopefully Free, that is)
Hey Andrew, good observation! I have been telling people for a while now that traffic effects rankings.
Hmmmm…food for thought that is for sure. I guess I am off to contact “the most interesting man in the world” to make a 4 minute video for me that will be niche universal and get it on all my sites for super stickiness power! If it works, I will sell it via WSO complete with up sells, downsells, sidesells starting with a dime sale on roller coster pricing. Riviera…here I come!
Being able to share the new site with your list (in your case students) was probably the dominant factor since that achieved higher visitor count than normal but the results simply confirm that with Google and SEO nothing is really set in stone. I have found posts on several of my niche sites ranking page #1 for primary keyword within a few days without any backlinking to speak of – certainly nothing I did myself. I drew the conclusion that focusing on ‘on-page’ SEO (internal linking and outbound links to authority sites – was the reason.
A further thought. I used to make my affiliate links open in the same tab because I didn’t want the visitor distracted from the buying urge. Now I assume it’s better to open all links in a separate tab, that way they will keep your site open while they investigate the new tab. Does this make sense?
Hi Andrew,
Just curious how this world affect article marketing as a source for backlinks ?
This being one of the pillars of effective marketing of our site for traffic ( or at
least it was ) . What’s your take on that ?
I spent a great deal of time on that site
That is pretty wild. No offense, but I know the site and its nothing to write home about.
So maybe the next way to game the system will be to pay for site visitors instead of paying for backlinks?
Dan – great story from your friend in Vegas there – Duhhhh – of course! – Google is a domain registrar so that makes perfect sense, should have been obvious now that you point it out..Thanks!
Thanks for this. Waiting to see results of your testing.
Hi Andrew,
Can you tell me if UA is being updated for 2012 or is there going to be something to replace it? Just curious as I am cosiderng the product but from recent comments and Google SEO changes for smaller sites I was wondering if you had different plans for it in the near future.
Thanks
Steve