I’ve always wondered why alexa reports such a massively different bounce rate for my site, compared to that given by google analytics. Google says about 50% whereas alexas states 17%.
I’ve always been sceptical about alexa stats, though. Do they still gather their information from alexa toolbar users? In which case a limited (but you could argue statistically significant) subset.
Google, on the other hand, get data for every single page visit on my site. So surely that’s more believable?
Presumably, they have a different definition of “bounce”. Or a different way of detecting it.
Mike
Alexa uses a broader sample than their toolbar but it’s just not reliable at giving you a real read on a site. People cite it because they need a metric but that’s the wrong way to go about analytics… if it’s not reasonable, don’t use it.
Google is going to be accurate to your site within their parameters and definitions. Any margin around javascript use, cookies, etc is minute and far less likely to be off than a third party trying to read your site.