Skip to content

February 21, 2020

Have you ever thought of using split testing to boost the performance of your content marketing campaigns?

Chances are, you already know about it and what role it plays. For most of us, it helps in boosting the conversion rate of a web page, an email or an ad. Split testing can be applied to most situations where you want to boost performance, which includes content marketing campaigns.

As you may already know, most companies are using content marketing to promote their business. This call for ways to boost the effectiveness of a content marketing campaign, and split testing can be one of them.

We will explain how you can use split testing to boost the performance of your content marketing campaigns. 

Split testing for content marketing

Split testing involves pitting multiple variants against each other. Your main aim here is to find which variant performs the best. The A/B test is considered to be the simplest split test as it tests two versions of something against one another.

However, you are free to add extra variants to the mix which is called the multivariate testing. You can also use A/B testing tools to get better results.  

See the Highest-Rated A/B Testing Software, Free →

You will need to send traffic to each version that you are testing. This is a need to collect valuable and accurate data, and will need to run the tests for a long enough period of time so that you can gather enough data to be certain that whatever you are seeing is not happening by chance. Short duration tests often run the risk of leaving you with inaccurate results. 

Here is a more tangible example:

You sent 100 visitors to both your variation page and the control page. You have 4 conversions on your variation page, and 2 on your control page. To you, it may seem statistically significant, but according to the calculator, it’s not.

ab calculator


You need to leave your test running until you get 7 conversions on your control page compared to 4 on your variation page to be sure of the results.

ab testing significance


Another important thing to keep in mind is that A/B split tests only work when you test one thing at a time. In case you test more than one thing to your page, it will become difficult to tell the change that is responsible for the results. This may seem to you as a long process, but you need to keep testing if you want better results.

Now when it comes to content marketing, what exactly should you test? Let's find out in more detail below.  

1. Headlines 

Headlines play an important role in the success of your content because they are the first thing your reader notices, and depending on how engaging your headline is will determine whether or not people click to read your content.

At the same time, you need to check how your audience is responding to your content beyond the headline alone. Are they commenting on your content? Are you getting more shares?

In case you're using a scroll mapping tool, does it indicate that your viewers are scrolling to the bottom of your page? You need to take care of all these things along with the headline.

The best headlines are the ones that are capable of getting more than just clicks. If you think it this way, then you can come up with something that will help you to rank. Headlines are what put your reader in the right state of mind. It tells them what they should expect from your content.

It doesn’t matter whether your content is serious or funny, the headline should be able to put them in that mood. Most people will just test one or two different headlines. That’s because you need a lot of visitors for each of your headlines to make it statistically significant. 

This may not be a problem for a bigger site like Buzzfeed. Sites like these get more than 20 million visitors in a month. But the rest of us have to do with our small pool of visitors.

buzzfeed headers


In this case, you need to rely on the results from a simple A/B or A/B/C test because understanding your customers is very important.

2. Content length 

The ideal length of any blog post should be around 1890 words. Still, every site is indeed different, and so is the audience. This is where you need to use a split test to find out the length of content your readers respond to best. 

To get accurate results, you need to create two different versions of your content. Plus, you need to keep in mind to try with different content lengths in every round. You need to repeat multiple rounds until you reach that “sweet spot.”  However, this is not limited to written content only. You can apply split testing to the length of videos and infographics as well. 

3. Positioning of suggested posts 

You need to make sure that you have great quality content on your page. It should be able to entice people to read more. To make your content more interesting, you should link it to other content. Below is a good example of content linked to other content. 

positioned posts to follow up


The positioning of these links can impact their effectiveness. If you want to increase your effectiveness, then you need to test them. This may include tests for the number of links to include, what kind of links to include, and where to include links in your copy. 

4. Images 

Images play an important role in your content. They not only help to illustrate the points, but also by breaking up long pieces of text. This makes the content more readable and appealing to the reader. But you need to make sure that the images you're using are relevant to your content. If the images are not relevant, it may confuse or distract the readers.

You should try testing images taken by you vs. images found online (i.e. iStock photos); images with people and without people; diagrams vs. pictures; and photos vs. illustrations. 

5. Calls to action (CTAs) 

One of the most important things to do is to wrap your content perfectly. By the end of an article, your readers should know what next steps to take, or what your "subtle ask" is. Calls to action used in content are different than what is used on a service page.

The call to action button on your content may ask visitors to:

  • Comment
  • Share
  • Read more
  • Subscribe to your email list
  • Inquire about your product or service
  • Visit a service page
call to action example


The nature of the call to action is dependent on the subject matter of your content. It is also determined by your goals. To determine the words you should use, you need to conduct a split test.

To make it easier for you to understand, here are a couple of examples:

  1. Do you use any of these SEO hacks? Tell us about your success in the comment section.
  2. We offer a 60-day free trial! Click here to sign up now. 

These are some of the basic examples, but based on your content's subject matter, you should alter the copy as needed. 

Over to you 

Split testing can help you to optimize the performance of your content marketing campaigns. This is one of the most effective ways tried by many marketers.

The things you should try split testing on include headlines, images, your CTA buttons, positioning of suggested posts, and your content length. The number of variants you will put to test depends on the number of visits you get. If you are a small company then you are advised to stick to A/B or A/B/C test results.

You need to make sure that you are receiving a large number of visits to each of your variants to make it statistically significant. This applies to each of the aspects that you are going to test. With the help of split testing, you can find out what works the best for you, your site, and audience. 

Learn more about A/B testing and discover the true value for your content marketing efforts and beyond. Read our statistics rundown! 

Get 1,000's of Data Points Read A/B Testing software stats, updated in real time →

Never miss a post.

Subscribe to keep your fingers on the tech pulse.

By submitting this form, you are agreeing to receive marketing communications from G2.