Look at any marketing consultancy, and you’re likely to see something about data. That they eat data for breakfast, are highly data-driven or that they are always looking at the numbers before making decisions. And this is a very good thing! We like to do the same over here at Preppr. The technological advancements in recent years have made it possible to track practically any marketing related metric, equipping marketers with what they need to make better decisions.
There are many different ways to listen to the data. You can base your insights on purchasing habits, customer demographics, or even sending out surveys to learn straight from the people you’re interested in, or some good old fashioned testing. There is value in every single one of these options, and in this blog, we focus on the importance of testing. What better way to learn than to set up experiments?
What is An A/B Test?
You’ve probably heard the words “A/B testing” thrown around from time to time if you engage in marketing. And the premise of it is pretty simple. Create a test with two or more variations and see which one performs better on whatever marketing metric you are researching. You can investigate if a landing page with a blue banner generates more leads than a landing page with the green banner, for instance. Typically, you would split the traffic down the middle. So 50% of visitors will end up seeing variation 1, and the other 50% would see variation 2. The most important thing is to ensure that you control for all other variables besides the one you are testing. Continuing with the example above, the landing pages should be exactly the same, only differing by banner color. This way, you can infer that the differences in behavior are actually because of the variable or element you are experimenting with and not some other extraneous ones.
Setting up A/B tests helps companies to learn how to better provide value to consumers. Once you learn what elements resonate best with your audience, you’ll be able to create much more value-laden experiences for them when they interact with your website and offerings. It’s the best way to figure out what promotions and strategies are going to work towards the success of the business!
What Can You Test?
You can essentially use A/B testing on any element you would like to. Here are some testing variables you may like to consider when setting up your next A/B test:
- Subject lines in emails (Subject A vs. Subject B)
- Images (Image A vs. Image B)
- Personalization in emails (Full Name vs. First Name)
- Call to Action (“Buy Now!” vs. “Start Now!”)
- Color (Yellow vs. Pink)
- The specific offer (Free Shipping vs. Discount)
- Headlines (About Us vs. Who We Are)
- Payment Methods (Single Page Checkout vs. Multi-Page Checkout)
- Captions (Short vs. Long)
- Type of Visual (Product Photo vs. Model Using Product)
The possibilities are endless, but these are just a few examples of everything that can be played with. It’s important to examine the entire service that you offer to consumers. There are tons of different touch points that can either nudge them towards the goal you would like them to perform or derail them. Go through all these touch points and run A/B tests to better understand how friction can be reduced at each one.
Remember to Be Realistic
In the end, the point of all the A/B testing is to learn more about the target audience in order for you to create a better experience for them.
All too often, we’re bombarded with case studies and articles about how changing one small detail on an e-commerce site increased the revenues of some company by 4000%. Sure, this could indeed be the case, but if you go into your A/B test with the same sort of expectation, you may be disappointed. Remember that not all A/B tests get published or talked about, only the successes. The truth is that we don’t know the half of all the other tests that have been run. The results of your A/B tests may not be exactly a home run, and that’s perfectly fine since they don’t have to be! Just because a test doesn’t lead to a growth on a large scale, doesn’t mean it’s a fail.
Instead, focus on achieving continuous improvement through your A/B tests. Each test that you run may help you learn how to best increase the desired behavior. Running tests on all the different elements will supply you with insights that will build on each other to lead to the gains that you want. One test may help you achieve a 10% increase while another may help you to achieve just 3% and another 6%. These incremental gains add up!
Faster Isn’t Always Good
Another note to consider is that velocity isn’t always a good thing. You don’t need to be churning out as many tests as possible at the same time to learn at a faster pace. The trade-off with chasing speed is that you may be spreading your resources too thin. This opens the door for tests to become unstructured and for you to not be able to monitor everything properly. If you are going to do something, make sure to do it right! Secondly, what you learn from one test can help you in designing the following one. Being more strategic with how many tests to run, and when to run them will aid in exploring the full idea to the max potential.
And finally, remember what this is: research. If you find out that one call to action works better than the other, or one color resonates more than the other, don’t just leave the insight there. Ask yourself why this is the case, go into marketing literature and trends. Or, use these insights to design a survey or focus group where you can learn more about these customers’ behavior. There are reasons behind everything, and understanding these reasons will be the key to creating a sustainable business!
Now… What will you be testing next? 😉