Whether you’re a marketer who helps clients with their ad campaigns, or a business owner who handles your own paid advertising efforts, you understand just how fine the line is between a successful ad variation and an ad variation that flushes money down the drain.
And while you can spend your time guessing and making assumptions based on emotions and past experiences, the reality is that you will never maximize ad spend without mastering A/B testing.
Understanding A/B Testing
Optimizely defines A/B testing as a, “A method of comparing two versions of a webpage or app [or ad] against each other to determine which one performs better. A/B testing uses data and statistics to validate new design changes and improve your conversion rates.”
In other words, A/B testing directly compares two or more variations of the same experience against each other to see which one customers prefer. And, when done properly, split testing can remove the guesswork from the equation.
While A/B testing is largely used to optimize ads, it can also be used to validate ads, which is very important in the initial stages of a PPC campaign. Here’s the difference between the two primary uses:
- A/B testing for optimization. Most commonly, A/B testing is used to optimize something. In this case, it’s used to optimize different ad campaigns until the highest-converting variation is identified and implemented. A/B optimization occurs when a PPC campaign has been in use for an extended period of time.
- A/B testing for validation. The less common use involves split testing to determine which ad variation will be best before the campaign is even rolled out. For marketers who really want to maximize ad spend, A/B testing for validation is incredibly important.
“A/B split testing is important if you want to validate which particular PPC campaign to use and optimize it to get the best results in terms of higher conversion,” says Alana Schwamberger, Senior Digital Marketing Manager at YourHearing. “We’re constantly A/B testing everything from landing pages to ad copy to ensure we’re being as efficient as possible.”
Related Article: The Comprehensive Guide to A/B testing for E-commerce
The Importance of A/B Testing
“If you are a digital marketer working in the online advertising space, split testing needs to become a part of your vocabulary (if it isn’t already),” writes Jason Parks.
The reason is that a slight difference in the CTR between one ad variation versus another can equal hundreds or thousands of dollars per day, depending on your budget.
Often you’ll hear business owners talk about how unfortunate it is that their PPC ads don’t perform as well as they’d hoped. But after you talk to them for a while, it becomes apparent that they’ve never tried different variations, let alone invested in A/B testing.
Would you count yourself among this contingency? If so, you need to reevaluate and begin investing your time into A/B testing. It could fundamentally change how you approach paid advertising.
The A/B Testing Process
Before you can invest in A/B testing and benefit from the results, you need to have a clear understanding of the process involved. Unfortunately, this is where many people go wrong.
They assume that split testing is as simple as taking two variations, making some tweaks, and studying the respective conversion rates. As experts will tell you, this is a serious oversimplification of A/B testing.
For accurate results, you’ll need to follow a replicable process that involves the following stages:
- Identify and hypothesize. The first step is to identify the problem, issue or desired outcome. You then need to develop a hypothesis regarding which elements will allow you to accomplish these goals.
- Test the hypothesis. The next step is to test the hypothesis using different variations, while retaining a control variation to compare against the alterations.
- Analyze the results. You’ll then analyze your results to determine which of the different variations performed best according to your desired conversion goals.
- Report and implement. These results are then reported to the key decision maker, whether that’s you or someone else, and then implemented according to plan.
This is obviously an oversimplified look at the split testing process, but is merely designed to show you that a process is important. You can’t approach A/B testing without a plan and expect to reach scientific and data-proven conclusions.
The good thing about A/B testing PPC ads, or bad, depending on how you look at it, is that there are really only four different components that can be tested.
They are: the headline, description line one, description line two, and the display URL. Other options exist, such as ad extensions, location in headline vs. no location, English vs. native language, and more, but you’ll likely dedicate most of your time to the four basic components.
4 Rules You Can’t Ignore
When A/B testing anything, ad variations included, it’s important that you follow a set of standard rules in order to achieve accurate results that give you actionable advice. Here are a few of those worth remembering:
- Never forget your goals. The number one rule is to always keep your goals in mind. As soon as you forget the goals you identified, your split testing efforts are rendered useless.
- Only test one thing at a time. The second important thing to remember is that you can only test one element at a time. While it can be tempting to test a couple of different components in order to accelerate the process, the reality is that this leaves you with subjective conclusions.
- Use the right settings. In order to conduct a fair A/B test, ad impressions need to be equal across all variations. When using AdWords, make sure you access the campaign settings tab and update the setting to “Rotate Indefinitely.” For Bing, the setting is “Rotate Ads More Evenly.” This ensures you get even distribution.
- Be patient. Finally, you have to be patient. Depending on your budget and/or bidding strategy, it can take time to get the results you need to reach an objective conclusion.
When it comes to any paid advertising campaign, A/B testing needs to be at the forefront.
Not only is it a good idea to use split testing to optimize an ongoing campaign, but you should also use it to validate the initial campaign. Keep these tips and strategies in mind and you’ll have an advantage over most of the competition.