A/B Testing: what is it, why it’s important, and how to start
A/B testing is something that many people have heard of, but don’t really know what it is, how to use it or don’t have enough time to actually start a testing strategy. Unfortunately, A/B testing is often thought of as a luxury for businesses, afforded only to ones that have dedicated marketing staff that can put large amounts of time towards the execution of big A/B testing plans. But that couldn't be further from the truth. A/B testing is something that any business can engage in to help increase the effectiveness of their marketing efforts and create more ROI for their business.
What is A/B Testing?
A/B testing is really simple. It’s picking two specific elements of a website, email, social post, etc. (any marketing activity really) to test against each other to see which one has higher performance. A/B testing is very helpful when you have a couple ideas for content, headlines, etc. that you think might all work well, but want quantifiable data that will prove which one will actually work the best. Simply put: A/B testing proves that something is statistically higher performing than another thing it’s competing against. And knowing this can reap huge rewards for your business.
What should I A/B test?
There are many different things that your business can A/B test to improve your marketing performance but here, we’ll focus on a few of the most common A/B tests that you can run without spending too much time and effort on a large testing plan.
Website headlines and hero text
Some of the best things you can A/B test to start, that often has some of the highest ROI, are different headlines and text in the hero sections on your website (these are the sections at the very top of your web pages). Usually, you only have a few seconds to catch the attention of a website visitor - according to techjury, people spend less than 6 seconds reading website content. If you have such little time to really get someone’s attention, then you need to make sure the first content they see on your site is a great hook. By A/B testing different headlines on your website, you can look to see which ones resulted in higher conversions, higher time spent on page, high CTA clicks, and other actions that will show you which ones your visitors like more.
Website pictures
Another great website A/B test is different imagery. It’s very common for a website to be built and none of the graphics, images, or other visuals change for months, or even years. But with consumers and markets changing every day, you should periodically test new visuals to see if they resonate more with visitors. Try different backgrounds, color contrasts, real world stills vs. animations, and other variations to see if by making simple changes to background images or hero graphics result in increased page performance.
Email subject lines
If you are using email in your arsenal of marketing activity, email subject lines are one of the easiest things to A/B test. There is an incredible amount of data on email subject line performance that shows performance really depends on the industry, person, product, and many other elements - there are no “one size fits all” subject lines. Because of this, A/B testing different subject lines (and as we discussed in a previous article, introducing audience segmentation) is the best way to get to your best email performance. Many low-cost email marketing tools (such as MailChimp, Hubspot, and Constant Contact) allow easy A/B testing that will show you which combinations of subject line length, action-based words, key phrases, and more work best for your audiences.
Social media posts
Most people think a social media post is a social media post...and that’s it. But there are several elements of a social media post that can influence its performance, including a key metric: click through rate. You can try different headlines and subtext combinations, character lengths, graphics, and more to see which posts perform better. There is even data to suggest that when posting on LinkedIn, instead of posting a link to an article in the actual post body, placing the link in the first comment of the post will increase KPIs like overall reach, clicks, and CTR. Try creating a few different versions of a post that only has a change to one element of the post and try them all to see which post works best.
The key to proper A/B testing: make sure to limit your variables!
As mentioned above, when you’re creating A/B tests, make sure you’re only testing two versions of the exact same element. This will ensure that the amount of variables you have to impact performance is limited to only the element you’re trying to test. If you change the headline and an image in a social media post to compare it to another post, how do you know if the headline OR the image that was changed had the biggest influence on performance (good or bad)? By testing only the image or only the headline and keeping everything else exactly the same, you’ll have definitive data that show which one works better.
By starting an A/B testing strategy, even if you don’t have a comprehensive one from the outset, you can begin to zero in on what activities will perform the best for your specific industry, audiences, and channels. If you are interested in starting some A/B testing on your website but don’t know if your site is set up to accommodate it, reach out to the Moby team to discuss how to create a website that will allow for easy A/B testing to achieve maximum performance.