I'm pleased, we have managed to create a site that Hannah and Matt can quickly get new products up on the site and take old products off.
The new site gives us an opportunity to serve Radiance's customers better as well; over time we can improve. The question is how do we know what to improve and when we have succeeded?
To some extent I can put my expert hat on (it's blue and white stripes with a red propeller.) I have studied and gained engineering experience and this certainly helped with our new design (over 65% increase in visitors.) However Philip Tetlock studied how good expert predictions are and showed they were pretty crummy; Dan Gardner's book Future Babble: Why Expert Predictions Fail and Why We Believe them Anyway is a popular account of Tetlock's research.
So we can't rely on me in my expert hat, what we are going to do is be guided by statistics and a scientific method. This is where the cryptic graphic above comes in.
It is a common occurrence that a visitor to a website arrives and simply leaves again having done nothing more on that site. It is so common it has a name 'bounce rate.' Bounce rate is generally a bad thing, for Radiance we want people to stop and browse and look at lots of nice things.
I have measured bounce rate over the last few months and the results are shown in the graphic. It shows where people came from and how likely they are to stick around.
How might this be useful? Well it is up for debate how useful social networking actually is, it is very fashionable and there are lots of people singing its praises but I and other people have our doubts. So instead of having an expert slanging match I can look for data.
I can see from the bounce rate that when visitors come from Twitter or Facebook they are not likely to stick around when compared with google or better still Hannah's own blog. It isn't the full story because if huge numbers of people arrived via Facebook the bounce rate wouldn't matter so much (actually in this case the absolute number for facebook and twitter visitors are tiny.)
I can look at all the information we have and recommend that Hannah only spends time on social networking that she finds fun; if she were to invest any further time it would be much better spent elsewhere.