How to Do A/B Testing—And Why You Should

A/B testing is an extremely powerful tool for marketers. In fact, it should be an integral part of every marketer’s arsenal. The truth, however, is that many marketers have yet to take full advantage of A/B testing—in fact, only 44 percent of marketing professionals use the technique.

If you’re a marketer, this is great news for you. A/B testing allows you to refine and re-refine your marketing efforts so that they are fully optimized to your audience’s preferences—and the fact that the tool is underused means that if you do it, you’re likely to get ahead of the crowd.

What is A/B testing?

A/B testing (or split testing) is a near-perfect example of the much coveted simple yet powerful idea. For marketers, A/B testing makes it possible to test two versions of a particular piece of content (for example, a landing page, newsletter, email, etc.) or a function (say, a call to action) by randomly displaying to a specified percentage of traffic two different versions of that content or function.

By monitoring which piece of content or function performs better, marketers can alter variables to see which has a positive effect on the desired output of that content or function (downloads, click-throughs, replies, and so on). By repeating this process, and continually altering the best performing versions, marketers can refine their work until it has maximum impact.

For A/B testing to be effective:

• Changes must be made incrementally—It’s important that you don’t make too many changes to your content or function at once since, with too many variables, you won’t know what’s making visitors happy and what isn’t. Only by making small changes will you be able to see exactly what your audience prefers, and then build on it. (Note, this means that A/B testing will take time—but who said it was going to be easy or that there were quick fixes?!)
• It must be well timed—As a digital marketer, you already know the importance of timing when it comes to posting online content. Web traffic comes in peaks and troughs, and in order to make the most of any split test you need to test at (your) peak times. The optimal time varies for every business; fortunately, your social media marketing tool should be able to tell you this information.
• It must be taken to completion—An A/B test is only complete when the piece of content or the function you’re testing can no longer be improved. There will inevitably be a point at which the improvements you see with each modification will start to plateau. That’s natural— no growth can be sustained forever. However, stopping short of this point only does you a disservice. To get the most of A/B testing, you need to see it through to the bitter end.

These may sound like simple pointers, but since only one out of eight A/B tests drive significant change, I’d suggest that not everyone is sticking to them.

A/B testing: What can it be applied to?

Unfortunately, not everything is a candidate for A/B testing. However, I’d recommend that you apply the technique to as many areas you possibly can. Earlier I mentioned some different candidates for A/B testing, but here is a more comprehensive list of some of the best applications:

• Call to action buttons (CTAs)—Variables include copy, button design, colour, position on site, etc.
• Newsletter sign-up buttons—Variables include copy, button design, colour, position on site, etc.
• Landing pages—Variables include page layout, palette, copy, call to action placement, etc.
• Email marketing—Variables include email layout, how the recipient is addressed, message copy, design, etc.
• Content marketing – variables include updating copy, headlines, images used in conjunction, variations to the actual content, etc.
• Social Ads – variables include advertising copy, accompanying image, advertising format, etc.

The key benefits of A/B testing

As you can imagine, the process of constant refinement to your marketing materials will lead to many advantages. By constantly optimizing, tweaking, and upgrading your materials, you can expect to see real improvements to:

Sales optimization—Sales are necessary to the survival of any business. Using split testing, marketers can expect to generate more hot leads from their marketing materials for their sales team to pick up. This means more business, more revenue, and more money. Everyone’s a winner!

Click-through rates (CTRs)—Getting traffic to your site (or landing page) is often crucial to generating more customers. Using A/B testing, marketers can expect to see noticeable improvements to click-throughs from social media posts and email campaigns.

Conversion rates—Currently, only 22 percent of businesses are satisfied with their conversion rates. By A/B testing ads, marketers can expect to see conversion rates improve dramatically. This means that digital advertising budgets can be justified and maximised, which (hopefully) means bigger budgets and that, in turn, means more optimized ads and more leads and sales.

Content downloads—Putting together content for download is time-consuming and difficult. So it’s incredibly annoying when very few people actually download that content. By using A/B testing to optimize landing pages and call to action buttons, marketers can expect to see the amount of downloads they get shoot up.

The benefits of A/B testing are clear. However, it’s not the case that by refusing to do A/B testing you’ll simply stay where you are; in truth, failing to conduct A/B testing may actually damage your business. Those who do so may lose business to companies with more refined marketing campaigns; fall behind the competition; waste time, money, and resources on ineffective campaigns; and have no appropriate metrics for benchmarking future campaigns.

If You Do Nothing Else, A/B Test These Three Things!

Hopefully you’re already convinced of the benefits of good A/B testing. However, if you’re new to the concept, or if you want to experiment before thoroughly committing, here are three things that every marketer should be A/B testing and the tools you’ll need to do it:

• Email marketing—By split testing your email marketing, the efficacy of your marketing emails will improve drastically. Wishpond has a great blog, which goes into some detail about how to do this effectively.
• Content marketing—The role of content is only increasing in importance. This means that failure to A/B test yours could lead to a lot of wasted time and effort. If you use a social media marketing tool (such as Hootsuite) it is easy to set up and monitor A/B content tests. Neil Patel wrote a detailed blog about content marketing and its process.
• Calls to action—The CTA is ideal for an A/B test. Small changes made here can lead to massive results, and those results are usually very important to your business. HubSpot’s article on the subject is incredibly helpful, and Google Website Optimizer can help put HubSpot's advice into practice.

A/B Testing Takeaways

A/B testing is a great tool. Unfortunately, not enough marketers are making use of it—and of those that are, many are doing it quite poorly. The canny marketer will take advantage of this situation by learning about and experimenting with A/B testing. As the state of online business continues to develop and change, A/B testing may turn out to play an important role in who wins and who loses.