Bounce rate is an inherently subjective measure of performance. That is, you shouldn't really compare your bounce rate to that of other websites other than for entertainment purposes. The reason being there are simply too many variables affecting the bounce rate, and what applies to one website, even if it's a direct competitor, may not apply to your website.

For example, my SEO firm's website is located at mwi.com and has a bounce rate of 65.75% over the last 30 days. Having a three letter domain is great when I'm telling people to go to my website, or when it's being used in print marketing, but when it comes to SEO and being listed in Google it's a dubious advantage. The issue is that there are a lot of other companies and organizations out there with the name MWI, and as a result I get a lot of traffic from people looking for MWI Veterinary Supplies, MWI Partners, Mid-wives International, etc. Then there's the traffic I get because MWI is an acronym for terms like many worlds interpretation, and something to do with Cisco IP phones. Obviously MWI has nothing to do with these entities or terms, but that doesn't keep people from typing "mwi" into Google and clicking through to our site. Once on our site, people immediately realize we're an SEO company, not a law firm or a theory of how the universe works, and they bounce away. So this makes our bounce rate considerably higher than it would be otherwise. Is this a bad thing? Surprisingly, it might be. According to this bounce factor experiment there is evidence to suggest that Google uses data from Google Analytics to rank sites, and sites with lower bounce rates can see their rankings affected positively.

But leaving that for the moment, what if I were to compare my bounce rate to the bounce rate of one of my competitors? Chances are my competitor's don't have their websites commonly confused as that of another company's, and so all other things being equal their bounce rate would be lower than mine. But if I didn't take that into consideration I'd be thinking "Man, what's wrong with my site?" and that might lead me to make changes that aren't necessary and that wouldn't produce any benefit.

The data becomes much more relevant if I measure it against itself over time or against other factors. If you have launched a new website then you will probably see your traffic increase over time. If, as your traffic increases, your bounce rate also increases, it might do to investigate why this is happening. It could be that your website is ranking in Google for keywords that you think are relevant to what you do, but perhaps when other people search for that keyword they are actually looking for something different than what you do, and so perhaps you should revise your keyword strategy.

If you experience high bounce rates (over 70%) then it might be that your website stinks, to put it bluntly, and you need to redesign it or do a better job communicating through your entry pages. Then again, it could be something else entirely. In looking through data for the 30+ sites I have access to I see bounce rates all the way from 16% to 90%, and in each case there are various reasons why those sites have those bounce rates, and a low bounce rate isn't necessarily a good thing, since it can mean a lack of traffic.

If you see a sudden increase in your bounce rate, chances are some sort of event has taken place that is driving new traffic to your site which, for some reason, is not sticking on your site. If there were a sudden spike in my site's bounce rate the first things I would do would be to look to see if there is a corresponding spike in overall traffic, and then I would look at my traffic sources to see if there is a spike there that explains where the traffic is coming from.

But even with this data, it's often hard to say what you should do with it. Obviously a high bounce rate means your site has low relevance to visitors, and so to capture more of those visitors you can work on making your site more relevant. But this only works if your site really is more relevant than people think. In my case, there's no reason for me to try to make my firm's site more relevant for the people who are searching for veterinary supplies. That is traffic I have to live with and can't do much about. I'd have a bigger problem if that traffic wasn't bouncing, since that would mean my site seems relevant to those people and if it seems relevant to them then it's probably not relevant to those I'm truly targeting.

To sum up, the bounce rate is affected by many things and there is no blanket answer that can be applied to all website to say "When your bounce rate is high you should..." Each website is different, each situation is different, and analyzing bounce rate data requires a hands-on approach--just one more reason why SEO can never be fully automated.

digg it!   Add to del.icio.us