I’m always looking for effective ways to communicate the need for online testing.  Online testing is simply trying out different variations of an idea on your website to improve things like your purchase conversion rate or newsletter signup rate.  This may be testing different layouts, button sizes, colors or even what text or images work best.

“A Fun Game: Guess The Outcome Of 15 Tests” by Lars Johansson is an effective display of how little we know about how our customers behave and what makes them tick.  My opinion is that the closer you are to your business, the less you know about how your customer reacts to your marketing.  When working with the Sun Developer Network eCampaigns team, I had a colleague present a similar case study depicting which banner ads performed better.  It was an enlightening experience for all.

You owe it to your business to ensure that marketing dollars are being used on the most effective campaigns.  You also owe it to your business to optimize those that aren’t.  Don’t just throw them out the window.  Test different ideas before assuming it’s just a bad idea.  For example, you may have tried utilizing an email list to generate new sales leads.  It may have generated a less than stellar conversion rate.  Before you toss it, have you considered trying different subject lines or different content?  In fact, when you start a campaign, you should ask, “How can I test a few ideas instead of going with my instincts?”  With this method, you would have taken a small portion of your email list and tested a few subject lines.  Then you would have used the best performing one on the remainder of the list.

So how did you do on the Outcomes Test by Lars?  I am not embarrassed to tell you that I was only correct 53.3% of the time.  I should not be surprised.  After all, my point is that you should be testing, testing, testing.  I’m not an exceptional web analyst because I make the best guesses, it’s because I test.