How We Decreased Our Bounce Rate By 25 Points

May 24, 2010
By Robbin Steif

Our old website had one of the worst bounce rates I ever see. (And the design was so 1999….)

Prior bounce rate

Nevertheless, I put up with it for a long time, because who has time to write and design a new site?   That is, until I tried usertesting.com, a story I wrote about last summer. (The short version: it was one thing to see the bounce rate in GA and know in your heart of hearts that you have to fix it. It was quite another to hear people tell me that the only site we were neglecting, our own, was significantly less than wonderful.)

The redesign took almost an entire year. Let’s face it, customers came first. But as we did the redesign, we listened very closely to the things that we heard in our first set of user tests. The most important lessons they taught us were the same ones we try to teach our customers:

  • They wanted to understand what we did immediately (i.e. within five seconds)
  • They wanted text they could skim
  • They wanted to see pricing
  • They wanted to know more About Us

New Site bounce rate 1 week

So we worked it (and it was hard — we wanted to create text that customers could scan, but not lose rankings.  We wanted to talk About Us but not be All About Us.)

At this point, we have only a week of data. Not exactly statistically significant, but I can see that our bounce rate for the site (i.e. not including our blog) went from somewhere in the high 60’s to somewhere in the low 40’s. (I highlighted the comparisons in the two screen shots.) We did the identical user tests as before, and while we still got some criticism, we also got a lot of compliments.

More later, when we have more data. I am especially looking forward to see if/why/whether the decrease in bounce rate on blog pages keeps up too.

Robbin