Marketing a business, product, or service online is a different breed when compared to traditional marketing mediums. With digital marketing, everything can be tracked and updated in real-time, which means you have the ability to test every campaign to see what’s most effective. In the digital marketing world, this is called A/B testing, and it can help you constantly optimize your best practices.
Planning An A/B Test
The reason why it’s called “A/B testing” is because you compare one campaign against one alternative campaign of the same element. As far as best practices, we strongly advise that the things you test be very specific, like changing the color, location, shape, or wording of the exact same call-to-action button. There are a few reasons why you would do this:
- If you test multiple things at once it is impossible to tell what change is responsible for any increase or decrease in clicks, submitted leads, and so on
- If you create several tests where each one has a different specific element you won’t have a large enough sample size of data to use
- If you test two things that are too general it’s more likely that any difference in the data will be a fluke or negligible
So when you plan on making an A/B test you should know exactly what you want to test, how you want to test it and why. For instance:
- Do you want to test more than one variant to see if it gets more clicks?
- Is your goal to generate more phone, email, or form leads from your ad campaigns?
- Are you considering running an email campaign that produces better open rates?
- Do you just want to increase your website rankings and search positioning?
Once you answer these questions, you can set up the test with a clear idea of what to look for when you are tracking your results. The great thing about digital marketing is that as you see the data come in, you can make any changes and adjustments necessary. You can choose to run a different test as soon as you want if the data wasn’t what you expected.
Things To Test
The truth is, with digital marketing you can test anything: visual ads, text ads, email blasts, website design, color schemes, fonts, sizes, and so on. If you do something differently, try testing it out. However, you shouldn’t necessarily test every little thing. Instead, focus on the elements you were thinking of changing that may have a positive impact. However, if you’re questioning the way you’re currently doing things, then here’s a general list of things you could test:
- Your website — headlines, titles, buttons, graphics, content, fonts or color schemes
- Email campaigns — subject lines, call-to-action buttons/links or content in the body
- Display ads — call-to-actions, the offer itself or image composition
- Social media — hashtags, images usage or not using any at all or headline text
For example, on your website you might test out different wordings in your headline or page title to see what gets more clicks and traffic; for email blasts, you can play around with subject lines for the same campaign to see what gets people opening more often; and for display ads and social media posts you might use a different image, remove the image altogether or change the image composition to see which one produces more clicks and/or click through rates.
Refining Your Efforts
Once your A/B tests have enough data, you can take a look at the metrics you’ve targeted for improvement when you first created the test. You should be looking for large, noticeable differences between the levels of success of the two alternatives so you can be sure that one is definitely better than the other. If the number of clicks, or leads, or SERP rankings over time is only slightly higher then it’s just as likely due to random variance then because the B-alternative was actually better.
Once you find a successful alternative, make sure you run another test of the same element, but this time test the new alternative with something else. You might find that there’s still an even better option after you play around with it. Trends in online design, search engine optimization , and consumer behavior is always changing, so annual A/B testing of the same things will help to make sure your best practices stay up-to-date.