How do you know whether your inbound marketing efforts are actually getting you the best possible results? You test the alternative. For years, A/B testing has been a crucial part of the toolset for advanced digital marketers. But if you’re new to either marketing or HubSpot, you may never even have thought of the technique as a way to optimize your efforts. This article will help you make sure that changes.
The Basics of A/B Testing
At its core, it’s a simple concept: you set up two slightly different variations of a marketing effort, and show them to two random samples within your audience. The variation that performs better can be implemented going forward.
The deeper you dig into the concept, the more you will find top-level marketers and thought leaders using A/B testing to optimize their marketing efforts. Google alone, for example, runs more than 7,000 individual split tests per year to optimize its product and user experience. That’s at least in part due to the versatility of the strategy and its universal applicability.
What Can You Test?
The short answer is everything. Any aspect of your digital marketing can be evaluated and optimized using this testing methods. From email subject lines to social media account name and website CTA buttons, simply trying out two variations may help you determine which one will serve your marketing goals better moving forward.
On his personal blog, industry star Neil Patel (he of Amazon and NBC marketing fame) detailed earlier this year just what you can test using this methodology. For example, have you considered the language you used in your navigation bar? As he points out, when Optimizely changed its Why Choose Us page to How it Works, the brand achieved 50% higher clicks and 8% more conversions from the page.
Understanding A/B Testing as a Continuous Optimization Tool
On its surface, A/B testing can be limited. The name suggests a simple truth about the concept, which is the fact that you can only change a single variable in two variations at the same time.
In other words, you don’t want to run a split test in which you send two completely different emails. If you do, and one outperforms the other, how will you know which part of the email led to more success?
The key, then, is understanding A/B testing not as a one-time implementation, but a continuous optimization tool. By testing one variable within one channel at a time, you can ultimately overhaul and improve your entire marketing output.
To stay with the above example, you can start by changing an email subject line. Then, take the iteration that performs better, but test the sender. Repeat the process as often as necessary to make sure every decision within your email is made with data in mind.
Don’t be satisfied with one-time results, either. You may have unearthed an improved subject line, but who’s to say that other subject lines don’t perform even better? At its best, A/B testing never ends, always working toward maximizing your digital marketing success.
Setting Up A/B Testing Within HubSpot
One of the many beautiful features of HubSpot is the fact that the software recognizes the need for marketing optimization. So it probably shouldn’t be a surprise that when you dig a little, you can find A/B testing tools built into the platform that can help you optimize your marketing efforts.
A/B Test Your Emails
As outlined in its Knowledge Base, HubSpot allows you to split test your emails in an automated way. When creating a new email, simply click on the A/B symbol in the left-hand navigation. Choose a name for your variation, from which point you can edit A and B simultaneously.
After creating the actual emails, it’s time to configure the test. By clicking on the A/B symbol again, you can set your sample size. If you plan on sending the email to less than 1,000 recipients, the split will have to be 50/50. But for larger sizes, you can choose alternatives such as showing both variations to only 15% of your audience each, with the remaining 70% getting only the variation that won the test.
How does the system know the winner? On the same screen, you can also set the criteria that determines just that. Finally, you can set up contingencies, such as deciding which version will be sent if the winner is inconclusive. After the send, you’ll find the results of your test right next to your regular email send statistics in the Content > Email screen.
A/B Test Your Calls to Action
Similar to your emails, HubSpot allows you to set up split testing for your website call to action buttons, as well. Simply create a CTA, then click the gear icon and select create a test variation. Edit the B variation, and the software will randomly substitute each onto your website.
Once the sample size is large enough, you can view the results of your test by clicking on the gear icon of the GTA again and selecting Details. Having analyzed the results, you can easily disable the underperforming CTA with the Disable variation selection in the same screen.
Performing Manual Split Tests in HubSpot
If you’re an enterprise customer of the software, you can also take advantage of other A/B tests, such as testing various aspects of your landing page. But even if you’re not, you can set up manual tests that help you at least approximate similar results to the above.
For example, if you’re looking to understand whether a question or definitive statement works better on social media, you can schedule posts on Facebook or Twitter with an identical message and link, but phrased differently. The post that receives more results gives you an idea of what type of content works.
Implementing A/B Testing in Your Marketing Efforts
Regardless of the software or channels you use to market and promote your business, A/B testing should be a part of your repertoire. Fortunately, if you use HubSpot, doing so is deceptively easy. We’d even go so far as to say that if you use HubSpot and don’t split test at least your emails, you’re not getting the most out of your subscription.
Of course, especially if you’re new to marketing or simply don’t have the time to implement split testing efforts, you may still be hesitant to use it. In that case, talk to us! We’d love to work with you in making sure that your marketing efforts are performing as well as possible with the help of a continuous set of A/B tests.