Bounces (one-page visitors)

[Spider]

My bounce rate - visitors arriving on my site and clicking away without delving deeper - is 66%. Sounds horrible, I know, but every site has a bounce rate - Google's is 26% and Bing's is 48%. Coca Cola has a 61% bounce and eHow has a bounce rate of 74%, so maybe 66% isn't so bad. Nonetheless, I think reducing the bounce rate would be a great way to get more business out of one's site. I mean, the people who searced for your keyword came, they are there, so it might be easier to get them to see what you offer than focussing on new people, all the time. Rather like helping the people who came into your store instead of ignoring them trying to get other people into your store.

So - What is your bounce rate?

Alexa the Web Information Company (http://www.alexa.com/)


What can we do to improve (ie. reduce) the bounce rate?

Dos and Donts of Creating a First Class Homepage (http://blog.ineedhits.com/videos/dos-and-don%e2%80%99ts-of-creating-a-first-class-home-page-14199221.html)


According to the above article, this is what we should have on our homepage...

-- all-encompassing benefit statements
-- general, all-encompassing keywords
-- links to main products/services
-- copy that’s focused around your customer
-- a preview only of what you offer
-- a benefit-oriented title
-- copy that gets to the point


And the things we should not include are...

-- every benefit you have
-- every keyword in the home page copy or in the footer
-- a link to every product or service you offer
-- generalized copy
-- your entire story
-- unclear title filled with keywords
-- anything that makes it hard for visitors to see what you offer.


I thought this was a great article and I'm going through my homepage now to make sure I conform to all the above. I'd be interested to hear what others think of this list and how they think their homepage conforms.

[vangogh]

The home page advice seems fine. I don't know that you need to conform to every point, but it does seem like good general advice. Keep in mind though that the problem could be due a disconnect between traffic and your site. For example your home page might rank well for a phrase that isn't a good match for the page. Imagine your home page ranked well for the phrase "gingerbread cookies." Your home page clearly isn't about gingerbread cookies so you'd expect those people to bounce. That's not an indication that there's anything wrong with your page though.

Bounce rates can be a misleading metric. It's probably not best to make changes based solely on bounce rate. For example a high bounce rate could be an indication that the page visitors landed on was exactly what they were looking for. They got whatever information they needed and left. A seemingly good bounce rate could be an indication that visitors landed on a page and what they expected wasn't there. They click around to another page or two that looks promising, but turns out not to be. They leave without having found what they wanted.

Think about Google. Anyone visiting the home page likely wants something else. If they want search results they type a query and are taken to another page. The process requires 2 clicks so regardless of what you think of Google's home page, the bounce rate has to be low. On the other hand say you search directly through your browser so the page you landed on is the results page. In that case Google's goal is 100% bounce rate as Google wants you to click on one of the links on the page and leave.

I'm not trying to suggest you should ignore bounce rates, but you should look at them in the context of other things. You generally wouldn't want a high bounce rate on your home page, but on a specific article or blog post it might not be a big deal. Bounce rates by themselves don't paint a particularly detailed picture.

[jpohl]

I'm with Vangogh on this. Bounce rate is just another metric. The suggestions are good but unless you are pulling a 90% bounce rate, don't sweat it too much. A lot depends on the industry you are in. In some industries a 50% bounce rate is bad and in other business areas a 80% bounce rate is good. Don't ignore the bounce rate but don't overemphasize it either.

[Harold Mansfield]

One one of my sites my bounce rate says it's 81%. However, it has a consistent visitor-ship (3 years) of around 500 per day, with 45000 page views a month. Will I do anything drastic to try and change that?
Not really. I'm fine with it, especially since I took 6 months off recently and let it go.

I'm more concerned about where the traffic is coming from and how they leave. Since most of my visitors leave through an affiliate or content link, I'm pretty sure that not too many people are getting there looking for something else. Actually , I think very few people get to that site by mistake.

I have more traffic from feeds, and images and to internal pages and posts than I do the home page. If you look at the bounce rate number, it may seem like every one is leaving, but if you look at other metrics, clearly they are not and are clicking links that make me money.

The only reason I actually looked at it was because of this post...otherwise I haven't been concerned about a bounce rate number since the site was new. And if I remember correctly it has always been about the same, but I have seen it lower when the traffic was higher.

[vangogh]

In some industries a 50% bounce rate is bad and in other business areas a 80% bounce rate is good.

Definitely true. I don't think there's any single % that signals good or bad for all sites. It's another reason why you can't look at bounce rate in isolation. You really can't look at any single point of data completely in isolation. Most are meant to be combined with others and taken together they attempt to paint a picture of what's happening on your site.

[CloptonCapital]

I forget what mine is but I remember it was way lower than that. My main website provides a very boring service though so I suppose if someone clicked they were already interested.

[Business Attorney]

The bounce rate can vary considerably by page. Sometimes it means that the user found that the page (and the site) was not useful and bolted for something else. On the other hand, it could mean that the page contained exactly the information the user needed and there was no reason to look further.

For example, on my LLC site I include a separate page for information on each state. The page on my entire site with the worst bounce rate last month was the California page (83%). But it had almost 1,000 page views and the average time spent on the page was over two and a half minutes. Yes, they left my site, but only after spending a lot of time on that page. The fact that enough people found the page useful that they spent an aggregate of 42 hours reading it last month is good enough for me.

Obviously, most of the people who landed on that page were finding it useful. Since that is my main goal with the site, the fact that they didn't read any other articles is fine with me.

To me, a bounce rate combined with a very low average time spent on a page is a red flag. Perhaps the page itself is bad, or maybe it is turning up in searches for terms that are not intended for that page. Either way, it requires some corrective action.

Some sites are created in a manner that virtually guarantees high bounce rates. Many "made for AdSense" sites are complete rubbish and know that anyone landing on their site will leave immediately. They just hope that the user will leave by clicking an ad rather than his "back" button. As Harold pointed out, if someone exits through an affiliate link, and that is the site's revenue source, then a high bounce rate often simply doesn't matter at all.

[vangogh]

David the page on your site is a perfect example of why you can't look at bounce rate in isolation. Sometimes the page in question was exactly what the person was searching for. They may not have clicked to another page, but the page was still successful.

[Spider]

I'm reading that G is now (or has been for some time) taking bounce rate into account in their algorithm.

Argument: If a visitor clicks on a link on a G results page and returns very quickly to G to click on another link, the page did not contain what was wanted. If a visitor clicks on a link on G results page and doesn't come back to G immediately, that is a clear sign that the page delivered what was expected, or led to another page that delivered. ergo, the first page is a poor result for the search term and the second page is a good result for the search term. And G is taking that ito account.

So, I think bouce rate is having more influence on search results, particularly since the Panda update.

[vangogh]

I think what you were reading is mostly speculation. Google is certainly tracking things like bounce rates, but it's unlikely they play any significant part in ranking since it's such an easy thing to spam. All you have to do is visit your competition and bounce. You could pay an army of people from South East Asia a very small fee to do that for you or develop robots to scale it further.

Also a bounce isn't necessarily a bad thing. If someone is looking for something and they find it on the first page of your site they land on, there's a good chance they'll bounce after visiting the page.


If a visitor clicks on a link on G results page and doesn't come back to G immediately

The amount of time it takes for someone to come back has nothing to do with bounce. If I only visit on page of your site and then leave it's a bounce. Whether I spend half a second or 2 hours on the page it's still recorded as a bounce.

Panda was about low quality pages. Bounces aren't necessarily a sign of low quality. Someone visiting a page reading for a few minutes and then leaving is a bounce, but the few minutes spent is an indication the page was quality.

As I said above bounce rates on their own are a mostly useless metric. They really don't tell you anything useful. You need to combine them with other metrics before they tell you much of anything.

[Spider]

"Taking bounce rate into account in their algorithm" = "Combining them with other metrics"

Yes?

[vangogh]

If you can prove that the left side of your equation is happening I'll answer the question.

What I'm saying is even if we assume the left side of your equation is true, it doesn't mean it's anything important to focus on. Google has stated they use more than 200 ranking factors. That was several years ago. More likely they use many more ranking factors. The thing is using one doesn't mean it plays any significant part or even applies to most searches.

There's a danger in focusing on individual ranking factors like this in that you become myopic in your seo. You start chasing after something that might have very limited benefits at best thinking it might be the key to the magic formula. Is Google using bounce rate as a ranking factor? Possibly. I think there are more reasons why they wouldn't use it than why they would use it, but it's certainly possible they are. Again though, even if they are it's unlikely to play any significant role.

There are much more important things to pay attention to than bounce rates. Where search engines are concerned it's unlikely you'd notice a difference in ranking or traffic due to a change in bounce rate. Where real people are concerned bounce rates are a useless metric when talked about in isolation.

[AaronConway]

Want to reduce your bounce rate? Be sure you give them someplace to go. Last fall, when I took over admin of an e-commerce site they didn't have any "related products" listed on the product pages and their bounce rate was around 70-75%. I immediately started working on those related products and the bounce rate immediately started dropping. It's now at 45%.

Aaron

[vangogh]

Definitely one way you can try to improve bounce rates.

Again though I'll say bounce rate alone isn't that meaningful. Someone who lands on a page of your site, spends a few minutes on that page, and finds everything they need before leaving will have a bounce rate of 100%. Someone who lands on a page, doesn't find what they want and clicks around to several other pages and not finding anything of interest before leaving has a bounce rate of 0%. Your site performed rather well for that first person who is more likely to come back than the second person.

You need to consider bounce rate with other metrics before really knowing whether or not there's a problem or how to fix it. A high bounce rate isn't automatically bad and a low bounce rate isn't automatically good.

[tylerhutchinson]

Bounce rate is just another metric to look at. As long as people are coming to your site and learning about your venture that should be good enough. No company has 100% sales on every site hit. I would worry about getting traffic to the site and let the bounce rate just be a mathematical statistic. Think about it a 50% bounce rate on 200 hits per day is 100 people going deeper in your site. But a 50% bounce rate on 10,000 hits per day is still 5,000 potential people going further into your site for business.

To get it lower I think the advice you put is the best way, but I would focus your time and energy on driving traffic to the site instead of fixing bounce rate. Just my opinion.

[vangogh]

I think you can focus on improving bounce rates. To me it's more that they can't be looked at in isolation. No metric really can. You simply can't tell much by looking at bounce rate alone.

[greenoak]

i dont worry about bounce rates.... my front page shows so much that if someone didnt look farther it would be obvious they wouldnt be a likely customer anyway..
like me and probably lots of folks, they know in a second if the site is for them....i dont sell on line tho....

[uwwandrew]

Consider the site/demographic/service when evaluating bounce rate. If you are selling soap a high bounce rate is worse than a law firm who's visitors just want the fax number on the home page.

[vangogh]

Another good example for why bounce rate alone tells you little. It's one metric that needs to be looked at in combination with other metrics before it tells you anything meaningful.

[LLP707]

I agree with Vangogh here bounce rate metrics should be taken with a grain of salt. I think your time would be better spent trying to increase traffic. Howver if you are really worried and want to run some tests you could set up two different layouts on a different area of your site with similar bounce rates and content to compare the change in bounce rate. This will let you determine which layout is more effective. However as mentioned earlier do not get overly worried about the metric.

[Spider]

This is what I learned from the preceding discussion. For all the arguments against concerning oneself with bounce rate, G is still taking bounce rate into account in their algorithms, so I think I will, too. Just how they are taking it into account, I don't know (or care) but I do know that someone who searched for a term that put my page within their reach and they read the title and snippet and then decided to click on and visit my site, that is someone I need to try and satisfy.

I also know that there is nothing they can do or learn or buy on my home page, so if they leave without going further, I know they did not find what they came for. I think it's a fair assumption, given that they searched for a keyword or -phrase that brought my index page forward and they thought the title and snippet was indicative that my page could give them what they wanted, that giving them what they want is within my business concept.

I believe it is easier to satisfy people who come to my website (once I know what will satisfy them) than trying to satisfy people who do not click on my page or who do not search for terms that would bring my page forward. (Just like satisfying a walk-in in your store is easier than trying to satisfy total stangers who do not walk into your store.)

So, I must indicate on my home page that I can provide what they are looking for. The trick is to figure out what the site visitor is really looking for that I am not indicating on the my home page.

That is where I am stuck.

[vangogh]

G is still taking bounce rate into account in their algorithms

Do you have proof? Assuming they are using bounce rates as a signal do you know how much weight they give it? I'd be grateful if you could provide the answers since no one else has ever offer proof bounce rates matter or how much they might contribute.

The problem is Google uses hundreds of different factors to determine where any page ranks. Bounce rate is a poor signal, because it's easy to spam. Assuming Google is using bounce rate it's unlikely it would have any kind of significant impact on where your pages rank. When I suggest over an over not to be myopic about seo this is what I'm talking about. You're going to end up spending time worrying about improving some metric when

1. You have no idea if it's even being used
2. Even if it is being used, you have no idea how much it really contributes to the algorithm

I'll grant you the 1st one or at least say that if Google can track bounce rates then they'll probably try to use the data. However the second point above is why you'll waste time with this. Again Google themselves say they use hundreds of factors to rank web pages. Your goal in optimizing isn't to focus in on all of them. If you do you'll waste time. It's the equivalent of spending $100 on advertising to make back $1 in sales.

As far as real users are concerned you can't look at bounce rates alone. Bounce rate vary wildly across industries and bounce rate alone tells you little to nothing about what's actually happening.

[Spider]

Just read this in an SEO newsletter --
Google's blog mentions three examples of metrics that are important:

Conversion rate: the percentage of visitors to your site who convert (buy something, fill out a form, etc.).
Bounce rate: the percentage of visits to your site who leave your site immediately.
Clickthrough rate: how often people click on your site out of all the times your site gets shown in search results.

[vangogh]

Google's blog says a lot of things that turn out not to be true. Google often posts based on what they want you to do more than they post about what actually works.

However assuming it is true that they look at bounce rates and I have no doubt that they do look at them, the issue is how much does it actually count toward where a page ranks in the algorithm. There are hundreds of different metrics that Google looks at and they aren't equal. Bounce rates are unlikely to be a strong signal on their own. An example:

Someone clicks on a link in the search results, lands on a page, and hits their back button instantly.

Another person clicks on a link in the search results, lands on a page, spends 5 minutes reading an article, and then clicks a link in the article to another site.

Both of the above are recorded as a bounce, but they are very different. The second one is a strong indication that the page contained quality content that was a good match for the query and should probably continue to rank well for that query. The first one isn't. Again those both are recorded as a bounce so if you only consider bounce rates by itself it gives you an incorrect picture of the above two pages.

What I'm saying is that bounce rate by itself is not a useful metric. It's only when you combine it with other metrics that it tells you anything useful. If you focus only on bounce rates without looking at other metrics you're wasting your time and you're very likely going to take some effective pages and make them less effective. You need to look at other metrics in combination with bounce rates before you can know how to respond.