The father of modern advertising, Claude Hopkins, first introduced the concept of A/B Testing (also known as split testing) in his book Scientific Advertising back in 1923. And despite the drastic changes that have taken place in the marketing world since his book was first published, this concept still remains a cornerstone of modern, data-driven marketing.
By utilizing one version as the control (A) and a duplicate version that's been modified based upon the testing criteria (B), marketers are able to gain incremental insights into the best ways to target and engage their audiences.
For years, direct marketers have used A/B testing for marketing in newspapers, magazines, mail order, classified ads, infomercials, and on the radio. However, with the rise of the Internet, more and more direct marketers have made the switch to digital marketing because it enables them to track every sale from the first click to the final sale-- making A/B testing easier to implement and monitor than ever before.
In order to remain competitive, businesses that use marketing automation as part of their digital marketing efforts need to incorporate A/B testing. If you're not performing A/B testing, you're flying blind. Though an email or blog post may perform well, without A/B testing, you'll never know why and you'll never be able to accurately replicate that effect.
A/B Testing is a Process of Continuous Improvement
The basic idea is simple: Test all elements of a marketing or sales asset, one-by-one, and keep the version that performs best. At the end, you have a piece of content or sales document that is fully optimized, from the headline to the final call-to-action.
In practice, there are a huge number of possible variables, so marketers concentrate on items that typically make the most difference: headlines and leads, calls to action, prices, offers, bullet points, and email subject lines.
In real life, testing is never complete because even the best controls become "tired" when your entire target market has seen them too often. i.e. The website design that converted customers in 2008 probably won't convert as well today.
Incorporating A/B Testing into Marketing Automation
Founded in 2006, Hubspot is one of the pioneers of inbound and content marketing, supplying the leading software platform for a marketing automation solution.
Hubspot clearly defines inbound methodology and how to use marketing automation: taking "strangers" and transforming them, a step at a time, into "promoters." It's a simple yet clear 4-step marketing stage process.
I. Attract Stage
Your content attracts strangers and turns them into visitors.
This is done by publishing engaging blog posts that utilize relevant keywords. It also includes publishing videos, infographics, and other interesting content and then sharing that content on social media. These inbound marketing strategies will help you attract interested strangers in your target market.
A/B Split Testing in the Attract Stage:
- What kind of content attracts the most visitors? Short blog posts? Long blog posts? Videos? Infographics? List blog posts? Find out what your audience responds to most, and use that smart content on your pages.
- What sites attract the most visitors? Your own domain? Facebook? Twitter? LinkedIn?
- What keywords attract the most visitors?
II. Convert Stage
Generating more online traffic is a great start, but it's just a start. You should assume that all those new visitors are not going to become customers the first time they visit your site. They will likely want more information about you, your business, and your products and services, before they'll make a purchasing decision. Keep your brand top-of-mind by sending them content in the form of email, blog posts, videos, or anything else that keeps them engaged and interested.
This step works to convert visitors into leads by encouraging them to opt into your email list or sign up for your newsletter.
The emphasis at this step should be on the calls to action throughout your website. You should also focus on optimizing the sign-up form for email and all landing pages.
A/B Split Testing in the Convert Stage:
- Landing pages. This includes headlines, images, and bullet points.
III. Close Stage
Once you've converted a visitor to a lead, the next step is to convert them to a buyer. This process will require the help of your sales team or, at the very least, a series of informative and persuasive sales documents.
Autoresponder emails can grow trust and familiarity. You can also send out additional content such as white papers and case studies to help people make more informed decisions.
A/B Split Testingin the Close Stage:
- Test your marketing and sales emails in every way. Text or HTML? If HTML, which design? What works best for subject lines? Do readers respond best to pure content or sales messages? What about personal interest stories? A/B tests will help your better optimize your emails. Have several versions of the email autoresponder series, emphasizing different product benefits. Use links within the emails to test which emails draw the most buyers.
- The additional content. Test your white papers and case studies.
- Test each part of your sales documents, including the headline, the basic message, the offer, the pricing, and the images. Would a video sales letters convert better than a written one? Which font types and colors are most effective? Do people prefer bullet points or blocked text?
IV. Delight Stage
At this stage of your online marketing process, you should have already closed the sale. Now it's time to "wow" your customers into becoming brand evangelists and promoters.
A/B Split Testing in the Delight Stage:
- Additional content
- Social media engagement
Which Step Should You Optimize First?
Ideally, all of them. If that's not possible, look at your current metrics, try to identify your weakest link, and analyze how that step of the process might be improved. If you are failing to convert prospects into customers, you are wasting all that effort attracting them in the first place. However, without traffic you cannot run a statistically significant test.
Use A/B testing as a means to grease the chute of your sales funnel so a larger percentage of the "strangers" in your target market become prospects and eventually, customers.
Once you've tested and optimized your automated marketing funnel, you have a valuable business asset. But, it's still not complete! There's one more step.
Which Strangers Eventually Become the Most Valuable Customers?
It's one thing to test, track, and optimize the content that attracts the most visitors, but not all visitors are equally valuable. "Strangers" drawn by one keyword may be a lot less valuable than "strangers" drawn by another keyword. Some keywords are more likely to draw freebie seekers. They may opt in to your list in larger numbers, but they may not actually be your target market.
The same thing applies to your content. Some kinds of blog posts may draw more visitors, and maybe even get liked and shared a lot, but only a small fraction of those visitors will likely ever become customers.
When you have enough data, start tracking the value of your keywords and the related content. Focus your efforts on building more of the right content that attracts the right type of visitor using the most effective keywords.
The end goal of any A/B testing is better lead generation and finding the most effective way to move a prospect to a customer. You will find you will continue these testing concepts as new content or opportunities emerge on your site, but always keep in mind to test against some of your existing so that you continue to further optimize.
If you'd like some thought starters and ideas, we recommend downloading our free guide The 30 Greatest Lead Generation Tips, Tricks & Ideas.