Do you want to grow your small business and increase your sales with split testing? If so, you’re not alone. Research shows that 44 percent of companies have tried to split test campaigns in the past. One of the most notorious success stories comes from former U.S. President Barack Obama, who generated an additional 60 million dollars by gradually changing tiny details on his campaign landing page.
Essentially, split testing is when you take an existing marketing campaign and make small changes in an attempt to improve consumer engagement and your conversion rate. We are going to take a look at several things you should keep in mind if you plan on using A/B testing on your website, email, or social media advertisements.
Let’s dive in!
Understand Your Hypothesis and Variable
The first thing you’ll have to do is come up with a hypothesis and a variable you want to test. For example, if no one is signing up for your email list, you may decide to change the color of the text block to red. In color psychology, red is known to illicit excitement and encourages consumers to act impulsively.
Similarly, someone with a low email marketing response rate may want to split test their email subject lines for more engagement. If one headline sees a click-through rate of 50 percent, while the other is at less than 10 percent, you may have found a variable worth changing. Studies show that split testing subject lines in emails and headlines on blog posts can double your click-through rate.
The takeaway here is you should have a hypothesis based on observable data before you start thinking about the variables you want to change. A lapse in judgment on your assumption can result in a split test that harms your business instead of helping.
A/B Test Your Most Popular Pages
If you’re new to A/B testing, it’s a good idea to start with your most popular pages first. Optimizing your poorly performing pages could result in no change, but that lack of growth could have an underlying cause, such as slow loading. There may be a point where you think a split test has failed, but it may have something to do with your design, UX, or customer flow.
When you split test pages that are already performing well, you can judge whether this change had a significant impact on your already popular pages. Your popular pages likely generate the most traffic, is
- Home page
- Contact form
- About us
- Relevant landing pages
These pages see way more traffic than an individual blog post or product page, making them the perfect starting point for business owners interested in performing split tests.
Know Your Sample Size
Your sample size can skew your results in favor of or against your hypothesis. Think about how wildly inconsistent your statistics would look if you tested 100,000 people the first time, and only 10,000 the second time. There’s no way of telling whether your additional hypothesis was correct based on this data because the sample size is inconsistent.
When you first start running A/B tests, your goal should be to try and show this change to about half of your overall audiences. You’ll have a nice 50/50 split, and using this method consistently will help you obtain relevant results for your marketing research.
As your business grows, you may have to adjust your sample size to match your new audience. This is a positive thing, as it means your audience is growing to the point where your sample size represents less than 50 percent of your overall audience.
Make One Change at a Time
Many marketers make the mistake of trying to change too many variables at once. You have to remain patient if you want to make the most of your A/B testing. Let your initial test run before you make an additional change to your campaign.
For instance, if you’re running an online contest and want to spit test the color of your background, you wouldn’t also change the entry methods. As with faulty sample sizes, this will result in skewed data that will have you guessing about the success of your change. You never want to feel like you’re guessing. Success in business stems from making actionable changes based on observable data. Some people have found success by just guessing, but statistically, the odds are not in your favor.
Similarly, if you’re testing social media adds, you wouldn’t split test the image and copy at the same time. You would have a hard time telling what change prompted in more sales, leaving you confused about the direction you should take with future campaigns. Both of these changes may have a positive impact on your business, but you’ll never be able to tell what percentage is responsible for the majority of your new sales.
Review Your Results
Once you’ve completed the various phases of testing, take some time to review your results. You can check your Google Analytics, social media statistics, and email campaign percentages to determine whether the split test resulted in an overall net positive for your business. You can test your headlines, images, copy, sales, and more across all of your marketing campaigns.
Split testing requires patience and big data for the best results. You’re not going to reach a major breakthrough in a day. Extensive testing helps you come to thoughtful, fact-based marketing decisions about your business. Take the time to test individual components based on your hypothesis, know your sample size, and optimize your most popular pages to get the most value from this excellent growth strategy.
