The bulk of our knowledge about the world comes from seeing what brings about the best results. We test-drive different cars before we drive one off the lot; we try on different clothes before settling on an outfit; we ask our real estate agent to give us several options and go to many viewings before forking over a downpayment.
The same principle works in business—the simplest way to find out what works and what doesn’t work for your company is through A/B testing, a sort of test-drive for your new assets. Conducting an A/B test will provide you with insights about your website’s visitors, as well as changes that best help boost traffic or conversions. A/B tests will get you much closer to achieving marketing goals; after all, if you don’t try out new ways of doing something, you will never have a chance to find out whether that may be the solution you’ve been looking for. In this post we’ll explain what A/B testing is, the A/B testing process, and 4 ways social media can help with your next A/B test.
What is A/B testing?
A/B tests are a way to methodically assess whether different ideas on how to grow your business will improve your marketing efforts.
During an A/B test, you have a “control,” which is the original version that is currently live on your site. Your “variant” is your new idea, which will be tested against your control. If you’re new to testing, stick to a simple two-version test that has a control and a variant. For more advanced testers, you can execute a multivariate test, which tests more than one variation at the same time.
For example, if you want to roll out a new homepage design and you have two versions—two variants—of a redesign in mind, you can perform the following A/B test to choose the best version: launch two versions side-by-side, each to an audience of roughly equal size, and see which version performs better against the control, which is the current homepage design. After your test, you’ll feel more comfortable moving forward with the version that performed best.
The A/B testing process
Businesses who are fully immersed in testing may have already created their own process of conducting an A/B test. But the recommended way to execute any test is to use the scientific method of testing.
- Ask a Question – e.g. “How can I get more people to click on my button?”
- Do Background Research – Use web analytics tools like Google Analytics to track the behaviour of your visitors.
- Construct a Hypothesis – e.g. “Changing the button color to blue will increase clicks on the button.”
- Test Your Hypothesis by Doing an Experiment – You create an A/B test where your variation has a blue button, and see how the variation performs against the control, which is the current colour of the button.
- Analyze Your Data and Draw a Conclusion – If the button that was blue received more click than your control, you can conclude that a blue button is more likely to attract a greater number of clicks than your control.
- Communicate Your Results – Let others in your company know about your findings. This will allow you to all be on the same page, as well as avoid others in your company to test something that has already been tested.
What can you A/B test that social media can help with?
Almost anything on your site can be A/B tested, because anything on your website that affects visitor behaviour can be subject to a test. But, what are some A/B tests you can do that social media can help with? Here are 4 examples of basic A/B tests in which you can use social media to help you create your variant. Keep in mind, there are numerous elements of your social media presence you can test—these are just a couple to get you familiar with the idea.
- Calls to action
Testing headlines on Twitter
Whether you’re trying to optimize headlines for a landing page or for a blog post, you can use Twitter to test out variations of a headline. Before venturing out into testing headlines on Twitter, be sure to make a list of the headlines you want to test. It’s also important to stay consistent, so remember to use the same Twitter handle if you’re testing multiple variations of the headline.
After you test your headlines, analyze the results to see which one received the most engagement. This can be based on of which one received the most favourites or retweets. Use the variant that received the most engagement on your blog or landing page headline.
Use social media searches for call to action inspiration
Are you testing the call to action on your landing page, but not sure of what language to use? Well, look no further than your own social channels—your social media followers’ messages are a great source of language that resonates with them. A simple search on Facebook, Twitter, or Instagram can lead you to a gold mine of customers who are talking about your company or product. The more relevant you can get to your audience, the higher the likelihood your call to action will improve conversions. For example, in the search below, someone described Hootsuite as a “one stop shop”— this is the type of language you want to take note of when doing your search.
Test images on social
Stumped on what image to use on your landing page? You can use Facebook ads to test what images works best with a value proposition. During the test, make sure that the only thing you change is the image, in order to get the best results. The ad that received the higher click-through rate will identify which image worked best to attract customers. You can use that information to influence which image you’re going to place on your landing page.
Find your reviews on social media
You’ve been tasked to find reviews about your product, so you can add them on your next landing page A/B test, but have no clue where to start? Don’t worry, that’s where social media comes in. Similar to how you can do a search for call to action inspiration on social networks, you can do the same to find your reviews. Remember, before you take a review from social media, be sure you get permission from the reviewer.
Do you know of more ways social media can help with your A/B tests? Share them with us in the comments below.