I heard somewhere that “You can’t manage what you can’t measure”. Web Analytics are understanding how a website is being used in order to tweak it performance, or is it that simple? It has been said about web analytics that the tool tells you what is happening on a web site but not why and the why might be more important than you think.
Case – the Bank:
I did a project way back with a bank that offered anybody that opened up an account through the Internet that they would match what people inserted to these accounts a pound for a pound, in my case a kronu for a kronu (with some limit though).
They bought advertisements in the national papers and the largest web portals and started promoting this unbelievable offer, but very little happened, they did no know if people came in to the site or if they did why they did not convert.
So the stopped and asked the key question “what is wrong?” and called on my company to help, the first thing we did was to insert a piece of code on their site and start measuring, created banner campaigns and target every banner with a unique code and to make sure that we did get as little out-side influence as possible we asked them to stop all none-online ads.
Okay, so what happened? Firstly we saw some good of traffic, the CTR was great, but only 5 or 6% went pass the first stage of the form, why? At that stage we called in some 5 or 10 people and we gave them a task, open an account. The first thing we noticed was that when programming a form you normally have the submit button on the right and the clear all on the left (if you have one) and in this case it was vise verse and our test subjects actually cleared their form in 90% of first tries and 35% of seccond tries. We could have guessed but the Web analytic tools gave us the proof.
We asked that the “Reset Form” button would be removed, but it was stopped by the designers, “it did not look”. Then we asked them if it was okay to switch so that the submit button was on the right and we got that through. So guess what! The conversion rate when up nearly 30% and now we had our proof and the upper management decided to scratch the designers and get the HTML programmers to delete the Reset button and do you know what happened? The conversion went up even further making this one of the most successful banner campaigns I have ever worked at.
And what is the moral of this story? Do not use Reset buttons….