In a recent conversation with a designer the question “What’s wrong with the website now?” was asked in response to the idea of setting up an A/B test on the homepage. Because he designed the homepage and would likely design the new tests I had to tread lightly. In a hurry I explained that “nothing” was “wrong” with the homepage but a small variation might have improved results. Changing the hero-shot, the title, or the color of the call-to-action button all might perform better than the current test. Conversely it could also perform worst. I tried to explain that we just don’t understand exactly what makes customers click. The only way to know what they really like is to ask them, through an A/B test.
I’m not sure I answered his question or aleviated any of his concern but his sentament is common. The idea that any one thing could be the best version the first time out is a pipe dream. The idea that a professional baseball player could step to the plate and hit a grand slam on the first swing is a pipe dream as well. Let’s face it, the first version isn’t always the best (which is normal). Why not try hundreds of versions and pick the best one?
So in short, there’s nothing wrong with the website, just your process.