Challenging popular intelligence in Web Analytics 1 Bounce Rate


this series of articles should be specially invited by NetConcepts’s Allen. It was also the first article to be done on a commuter train.


must the wisdom of the masses be the correct wisdom? Perhaps you have also heard the word "·" for most people. No matter where you are, because of the "popular wisdom", higher IQ does not always ensure that we make the right decisions, as 36kr article said, smart people may sometimes be easy. So this series of articles is looking at what popular intelligence in the field of web analytics might lead you astray". In the first article, we will focus on bounce rate.


Bounce Rate, site analysis of one of the basic metrics, I believe everyone is not strange. Bounce rate seems to represent some of the magical challenges that Web Analytics has given us, and if we have uniforms or the power to dominate it, we seem to have achieved something remarkable.

people believe bounce rate as low as possible, but as the saying goes "too far", lower bounce rate and perhaps not as we expected so beautiful.

super low bounce rate implies monitoring technical problems

anyway, if the site (or some of the major landing page) bounce rate is less than 5%, or even less than 10%, there may be monitoring technology problems. There are at least two conditions that cause bounce rate to be distorted.


page framework (i.e., pages through iframe or frame nested) the parent page and if sub pages are added to the monitoring code, so in most cases, the page in the browser load, monitoring code in a short time by running two times. For web site analysis tools, such as Google Analytics, will record two different pages to be opened (because the parent page and sub page URI is certainly different), so this situation will obviously is not considered to be a bounce, even if the users do not click on any links to this page to browse other pages. In the case I’ve seen, bounce rate is less than 5% in this case.

second, if you accidentally add two identical monitoring code on the same page, or for some reason, make the same monitoring code run two times. So when the page is loaded, for Google Analytics, as this page is refreshed once, so GA doesn’t count this as >