You can read Google’s blog post about the new offering. Google Analytics Content Experiments lets you quickly and easily offer your visitors different variations of your website, and check to see which ones lead to more conversions. For example, if you want to try out different versions of a product page to see which one leads to more conversions, you can test up to six variations of a specific page.
Assuming you already use Google Analytics, you won’t need to do much (aside from creating the page variations) to use this new feature. Google Analytics provides a wizard to guide you through creating the experiment. A video in the blog post explains all the details. You can control what percentage of visitors see the new page(s), and the goal on which you wish to focus. You can monitor the data and the results from the experiment while it’s in progress, and even view predictive data to see which pages you can expect to convert better.
What this means, as Mike Fleming notes, is that if you’re not testing your website to see which versions of your product pages perform better, you’re telling everybody that you think you’re making enough profit. “Your site exists to persuade visitors to take actions, right? Well, you don’t know how good it is, or how good it could be, until you fully embrace experimentation and testing with it,” he pointed out.
Even if you’ve built your website with all of the “best practices” to be as good as it can be, and you’ve learned all you think you can from examining and dissecting other websites (especially those in your own field), you may be missing one vital point: your business is unique. “The truth is that what works for one business doesn’t necessarily work for another, even if they sell the same thing,” Fleming explained. You need to build pages that are informed by your own understanding of your individual audience.
For example, assume that you and I both sell electronic devices. If we’re going after a different demographic, different kinds of pages will perform better. Even if we’re both selling iPads, if you’re selling them to college students, you’re going to use a different approach than I would if I’m selling them to people like my neighbors in the “active adult” community in which I live!
Or let’s look at this idea just a bit differently. You and I sell the same electronic devices to different audiences. We both make assumptions based on what we know about those audiences as to what kinds of pages will perform better. But we could BOTH be wrong. And we’ll never know unless we replace those assumptions with data. Google Analytics Content Experiments offers what could be the cheapest and easiest way of getting that data. What are you waiting for?