In today’s data rich world, it is not a good idea to make blind business decisions. This is especially true for marketers. When it comes to improving digital materials we often make decisions about various materials such as targeting, copy, hero images, landing pages, email copy, and the like. A/B testing is a very valuable method for helping digital marketers make smart decisions. A good test can spell the difference between materials that make an impact and lead to high conversions and those that do not. This is why marketers cannot simply rely on their preferences or intuition to determine marketing decisions, especially concerning elements that are crucial in encouraging people to click and convert. Rather than relying on guesswork, successful marketers listen to the results of an A/B test or split test to determine which material would work best and lead to more conversions. If you’re looking for an in-depth guide on how to conduct A/B testing for your advertisements and other materials, you’re in the right place.

A/B testing or “split testing” is a testing strategy that compares two different versions of a material to determine which one is most effective at reaching a specific goal. Some examples of materials that may be tested through this method are email title copy, CTA’s, website banners, advertisements, email blasts, website popups, hero images, and landing pages. Testing two different versions of the same material (such as two different images for a Facebook ads) allows you to see which drives more traffic, conversions, or sign-ups. This way, you’ll be able to gauge what works on your target audiences, and you’ll be able to use the more effective material and expand its reach during the next run. Eventually, a series of A/B tests on different materials will provide you with valuable insights concerning your target market. You can discover the preferred color of a button for a CTA, what style of imagery they like, what cut of a video etc. These tests can even be used to see what product description resonates with your customers in a way that drives them to purchase! You’ll be able to see where and how it is most prudent to invest your marketing time and budget. A/B testing also allows you to test out unique or potentially risky campaign ideas on a smaller population before going all-out with a larger audience.

A/B testing is largely used on customer-facing content, so there are a lot of materials that you can evaluate with A/B testing. Some examples include:
Paid ads (Google Ads, Facebook Ads, Instagram Ads, etc.)
Newsletters
Web design
Individual emails (Follow up emails, personalized emails)
Email campaigns
Multimedia marketing strategies
All sorts of website and ad copy
CTA buttons
You can conduct A/B tests on several variables in each of the categories above. You can essentially test any element of the material’s style or content in a customer-facing material. Examples of different options that you can test are:
Layout
Color scheme
Images
Copy (headings, subheadings)
Special offers
Product pricing
CTA button design
Now that we understand the parameters let’s get to the actual procedure involved in conducting a successful A/B test.

Start applying the scientific method. You know the one, from primary school. It’s this simple:

Step 1: Identify a Problem
The first step is identifying a specific problem. For example, if your problem is that there are not enough conversions on your eCommerce store despite having numerous ads on various platforms, you’ll have to be more specific in finding out what exactly you need to look at to start fixing the problem. Let’s say that your store has plenty of online sales, but only very few of them ever come from your Instagram ads. This is a great place to start your analysis. You can look at your analytics to pinpoint how much you’re spending on your ads versus how much you get from conversions. You can imagine how jarring it would be to see a Ferrari advertisement filmed by an amateur college team, with bad lighting and cheap bad acting. Everyone already believes that Ferraris are luxury, sporty, cool and expensive works of art. Making sure that your creatives align with the goals for your ad is paramount for having a successful Instagram campaign. For example, if your ad is designed to get people to make a purchase, you wouldn’t want your call to action to read “Learn more”. The same goes for the creative itself, if your ad is about a specific product, you don’t want to have your image portray something general about your company, you want it to be of that specific product. Keeping your ads in line with your overall brand image and promise helps customers feel like they are not being sold to, but just looking at a new product, style or emotion that they want. Creating that product pricing and creative quality dissonance helps keep all of your creatives in line, as well as helps keep customers comfortable with the price of your product. If they see a premium ad, they will feel better about paying a premium price.

Step 2: Analyze User Problems
To pinpoint where you can make improvements, it’s helpful to analyze user data. Looking at how people interact with your existing materials will allow you to identify gaps and areas for improvement. For example, if you run an ad banner and find that people are opening your links but ‘bounce’ once they click through to your website, you might find that there’s a problem somewhere in the customer value journey from your advertisement to your website. This could be something that doesn’t require A/B testing, such as a slow loading page, or it can be a poorly-designed welcome banner or unfriendly website design that may be driving customers away.

Step 3: Formulate And Test Your Hypothesis
Start with one that you want to change. For example, you might want to test whether or changing your “Buy Now” button from Orange to Green will increase click throughs? After you’ve formulated a hypothesis, it’s time to test it. Develop a new version of the said test item implementing your idea. For example, the abovementioned ‘Shop Now’ version can be tested by running an A/B test between the existing version and a new one. In the new version, for example, you can create a version of the ad or email that places the ‘Shop Now’ button on a more visible part of the page. You don’t need to change the font or color just yet, only the positioning. You can then run this test for a day to see if anything changes.

Step 4: Analyze And Report Data
After you’ve concluded your test, you should now look at the results to determine if the new version of your material resulted in your desired changes. For example, if placing your ‘Shop Now’ button at the top of your email or ad banner did not lead to any changes in conversion, you can now proceed to test it in another location. You can test placing it at the middle part to see if engaged readers will click on the button once they’ve read through a part of your email.

Step 5: Find New ‘Challengers’ For Your ‘Champion’
A/B testing refers to the ‘champion’ and ‘challenger’ as the current best option and other new possibilities. The champion is the more successful option among two or more options you’ve previously tested. You can then test the champion against other options, called the challengers. Ideally, if you find a challenger that beats your champion, then you’ve found a better way to execute your ads or website materials. If not, then you know your champion works well. Remember that split testing is a constantly ongoing process, and as your teams come up with new and exciting ways to engage customers, it’ll be important to size these up against each other with every new product, campaign, or idea. We hope that this simple A/B primer has helped you better understand how the practice works and how essential it can be for any online business.