The A to Z of A/B Testing
Just because you have a website, doesn’t mean it is benefiting you. Rather than staring at your screen and hoping that cyberspace has taken a liking to your pages, you can find out how well your site is working for you. Maybe it is the product you lead with. Maybe the color scheme isn’t building confidence. Perhaps your page titles are confusing. It might even be that image on your homepage takes up too much screen space. Testing your theories is the only way to solve the problem, and even the smallest budgets have the ability to perform analytical studies that won’t necessarily wreck their bank accounts.
What is A/B testing? A/B testing, or split testing is the splitting up of a group of people onto two different test pages. This will allow you to test if one page has a higher conversion rate or bounce rate than another. Or one product image or headline inspires more clicks than another.
If you use the A/B testing feature through an advertising service like Google Adwords, you can direct visitors to your A/B testing pages, and the analytical software will keep track of the behavior of each set of visitors. Obtaining paid advertising can get very costly, but with a little care, your A/B testing will reap high rewards without the cost of the ads that would otherwise direct people to your pages. You can find out valuable info like: are people clicking on your testimonials more in one version than the other? Is one page getting more comments from readers than its counterpart? Make a list of all of the things that you want to make sure are working at their finest, then you can put them to the test. If you’re not sure about your results, mix and match them, then whittle down the products until you have one that looks like it will give you the highest return on your investment.
What should I test? You can test almost anything online, but it is smart to focus on the big picture, then narrow it down to the action-information. It is best to use the Scientific Method to see what you truly need to know:
- Ask a Question. What are you ultimately trying to optimize?
- Do Background Research. Learn more about your competitors, your customers and the analytical software you are using to ensure you get the most knowledge out of your results.
- Construct a Hypothesis. Will a certain color or specific layout draw more eyes (and clicks) than another?
- Test Your Hypothesis by Doing an Experiment. Here’s your A/B. The first site has the color and/or layout that you’re testing against. The second site has the color and/or layout (or whatever is referenced in your hypothesis) that you think will get the highest results. Just make sure you test one thing at a time, to ensure you know where your positive/negative results came from.
- Analyze Your Data and Draw a Conclusion. After you’ve had a bunch of people poking around on test pages, step back and see which one performed better, or if they were equal. If the latter, re-evaluate any other hypothesis that you wish to test, and re-create the experiment.
- Communicate Your Results. If you are a typical small business, you are the accountant, manager, marketing expert, salesperson, and everything-else guru. If you’re lucky enough to have a marketing team or web folks, or are partnered with a company like Optimum Systems Online, let them know what you’ve discovered, and work from there to make your website be the very best that it can be!
- Make Changes. Don’t just sit on your results; update your website to solve the problem! If that image is too big, make it smaller. If the headline was too confusing, update it to something more simple. Your viewers have spoken, so make it count!
We A/B test with our clients constantly. For a grocery product company, we tested how customers reacted to a promotional homepage vs. beautiful food images, and then retested the winner against a homepage that had a large amount of text instead of images. For a doctor, we tested if longer or shorter emails yielded higher click-throughs. We’ve tested if we had higher participation when visitors were presented with a quiz or free download. We’ve given visitors different categories of navigation to see which they found easier to use. We’ve even tested logo options to see if a revamp of the design yielded positive results.
A/B testing is a constant task, as website technology evolves, new opportunities are presented, your audience changes, and visitor preferences change. It is always important to keep your finger on the pulse of these changes. Are you a business owner that is experienced in A/B testing? Great! We’d love to hear your tips, tricks, and advice! Post your thoughts in the ‘Comments’ section below. If you’re looking for a little help of your own, post about that, too! Someone may be able to answer your questions!