By Ryan Malone

ab_testing_content_strategy.jpgAs the Internet grows increasingly crowded with content in all shapes and forms, it’s more essential than ever for marketers to help their website – and the content on it– to stand out as much as possible. While it goes without saying that creating buyer persona and buyer’s journey-targeted pieces are the first step to successful content, where do you go from there?

Once you’ve already got creating targeted content under your belt, what other tactics can be used to improve individual content pieces, or even your overarching strategy? Enter HubSpot’s A/B tests.

 

What is an A/B test?

how_ab_testing_improves_content_split_testing_graphic.jpgSometimes called split or bucket testing, A/B testing is the process of comparing two versions of an asset (for instance website page, graphical element, copy, or otherwise) to see which performs better. The assets are compared by showing two variations (i.e. variation A and variation B) to similar visitor populations at the same time. The asset that performs the best is the winner.

Often, the most successful A/B tests are often those that change only one thing at a time – for instance, tests that involve only changing the heading on a landing page or the color of a CTA button instead of showing a completely different page or design. Why? Because if you change too many variables between the A and B variations, it becomes impossible to determine which elemental change actually contributed to changes in asset performance.

 

What can you A/B test?

Frankly, just about anything. But for the purposes of this post, we’ll mainly be discussing how the three types of automated A/B tests built into HubSpot – CTAs, emails, and landing pages– can impact your content strategy. However, bear in mind that social media can also be a powerful tool for running A/B-type tests, as well. For more information, check out this great post from Buffer.

 

First off, let’s look a little more at what each type of test entails in HubSpot:

CTAs: A/B testing in HubSpot’s CTA tool allows two different variations of a CTA to be shown to determine which is most effective. Variations typically involve different designs, text, or colors.

Emails: HubSpot’s A/B email testing feature allows you to test two versions of emails “to your audience so that you can obtain information on which triggered the most engagement,” and then automatically choose the most effective email variation as the final email send. 

Landing Pages: As with emails and CTAs, landing page A/B tests randomly show different versions of a page to determine which page performs best.

 

Regardless of which type of asset (CTA, email, or landing page) is being tested, the general idea behind running a successful A/B test is the same: change one variable at a time to see which has the greatest impact on the metrics that are most meaningful for you. As a rule of thumb, though, it’s generally a good idea to measure the impact of a piece as far into your buying process as possible – you may find that the landing page with the lower visitor-to-lead conversion rate actually lead to more sales qualified leads.

 

What does A/B testing have to do with improving your content?

The premise behind A/B testing is that by trying different variations on a concept, you can identify which best resonates with your audience and will ultimately be most successful. If the concept you’re testing is, for instance, a topic you’re interested in blogging about or creating a premium content offer with which to convert leads, A/B testing it will allow you to determine which permutation of that concept will make the biggest splash when it ultimately gets transformed into a post, eBook, whitepaper, infographic, or otherwise. 

Let’s say, for example, you’re interested in creating an eBook about on-page SEO, but you’re not sure exactly which direction to go in with the copy. Before heading down a dead-end road with your content, you could run an A/B test to see which of your ideas best resonates by sending your target audience two variations of an email containing SEO-related blog posts, and altering the subject line of each variation to cover your two potential eBook topics. Whichever email subject line gets the most opens would then determine the topic your eBook should be about.

 You could also run a very similar test on social media, by sending out tweets promoting existing posts and testing which positioning - whether you’re talking about topic A or topic B – gets the most interaction. Here’s a great example of Buffer, who are avid fans of the A/B test, doing something similar to test their content marketing strategy:

buffer_example_b_testing_content_strategy.png

In the above example, Buffer tested two variations of an existing blog post title. Their test revealed version B, “The Research Behind Hashtags,” actually performed better than the existing title (version A). As such, based on the findings of their simple A/B test, the title of the existing post was changed.

 

How can HubSpot’s A/B tests improve your content?

Because A/B tests allow you to assess how well something is going to perform before you go “all in” on it – and because HubSpot’s built-in A/B tests will perform the analytical grunt work with little to no effort on your part, they can be a great asset to refining your content marketing strategy. 

Here are a few examples of how HubSpot’s A/B tests can be used to improve your content: 

  1. Choosing between two similar keywords to target with a new piece of content – or looking to breathe new life into an existing offer? Run a landing page A/B test and vary which keyword you use in the page headings. Whichever performs best is the one that should be integrated into your content offer.
  2. Vary the headings used on your CTAs to identify which positioning for an existing offer (or future blog post) works best. For instance, a test between “Increase Your 2016 Revenue” versus “How to Increase Revenue in 2016” could indicate whether “how to” or more expository posts around this particular topic are a better content target.
  3. Not sure if your persona prefers “guides,” “eBooks,” “guidebooks,” “digital booklets,” or otherwise? Play around with the wording on your CTAs to determine how to package your offers.
  4. Created a checklist no one is downloading? Try varying your positioning of the offer packaging – i.e. as “guide + checklist” or “quick start guide” – on your CTAs or landing pages to see which gets the most clicks. Then, use your findings for the positioning of future offers – and even blog posts.
  5. Test topic positioning in email subject lines to see which messages get the most opens, clicks, and/or conversions. Use those topics that get the greatest engagement and results as post or lead generation content ideas.
  6. Beyond A/B testing email subject lines, consider testing email copy itself. Specifically, test out different variations of anchor text to see which get the most clicks (or which drive the most qualified traffic), then create content around those topics.

 Whether it’s HubSpot’s built-in tests or those you conduct on your own, A/B tests can be powerful tools in helping to refine your content strategy. If you’re in a content rut, out of ideas, or just looking for a way to take your content to the next level, A/B testing ideas, positioning, and even keywords can be a great asset to help ensure your content strategy is working as hard for you as you are for it.

marketing automation comparison guide 

The-Ultimate-Guide-to-Inbound-Marketing-Personas-cover

Refine your inbound marketing efforts with:

The Ultimate Guide to Inbound Marketing Personas

Check It Out
Ryan Malone

About the author

Ryan Malone is the founder and CEO of SmartBug Media and is a veteran of Deloitte & Touche, Seagate and several venture-backed technology companies. When he's not leading SmartBug and helping clients build high-octane marketing organizations, he's loving his wife and daughters and unsuccessfully learning the guitar. Go Terps! Read more articles by Ryan Malone.

Subscribe to get our new blogs delivered right to your inbox