A/B Testing & CRO: How, Why, & Who
If you're trying to grow your business but finding difficulty in measuring the success of your marketing efforts, utilizing A/B tests can be a great way to improve your strategy and reach your goals faster. In the article below, I will touch on three main points about A/B testing:
- How you do it
- Why you should do it, and
- Who has done it well
How you do it
Simply put, A/B testing is a method of gathering insight. It involves comparing design A (control) to design B (variable) to understand which version performs better. In A/B testing, you could run tests about your call-to-action, landing page layout, font style and colors, and much more. As a relevant example, I could A/B test the title of this article:
Optimizely provides a basic framework for A/B testing:
- Collect Data: collecting and analyzing your data about existing webpages can provide insight on where to start optimizing.
- Identify Goals: these goals are most commonly conversion rates. Whether that be lead generation, purchase or email building an email list.
- Generate Hypothesis: after identifying a goal, create a hypothesis about each test and prioritize.
- Create Variations: create a variable variation for your experiment. This could be changing the color or font or placement of your call to action button on your landing page... even the length of copy on your landing page.
- Run Experiment: utilize the A/B testing platform of use. Half of your traffic will be directed to version A and half will be directed to version B.
- Analyze Results: analyze the test, gather insight and determine if there is a statistically significant difference.
Why you do it
A/B testing allows you to gain valuable insight about how your users and potential customers interact with the content you present.
Why is this important?
Let's take a quick step aside and talk about how conversion rate optimization plays a role in A/B testing. Conversion rate optimization is the process of improving your website or landing page’s ability to convert visitors into leads or customers. A conversion rate is defined by Moz as the number of times a user completes a goal divided by your site traffic.
Conversions are divided into two types: micro-conversions and macro-conversions. Macro-conversions can be in the form of purchasing a product, subscribing to a service or lead acquisition. Examples of micro-conversions are a creation of an account, viewing a product page, or entering the checkout process.
How do you improve these conversion rates? Use A/B testing to test your product pages, checkout process, and account creation process can help you optimize your conversion rates.
Who has done it well
In the case below, Sony A/B tested banner ads for their advertising campaign for their customizable ASIO laptops. Their goal was to identify a better approach to optimize click-through rates and shopping cart adds. They tested two variations: variation 1 emphasized a promotion and variation 2 emphasized customisability. After running the A/B test, Optimzely provided the results. Variation 1, the banner that focused on customization, saw a 6% increase in banner CTR and led to a 21.3% increase in visitors that reached the shopping cart compared to the original banner. Variation 2, the banner focused on the sitewide promotion, only increased click-throughs by 1.8% and actually performed worse than the original in terms of shopping cart views, decreasing conversions by 2.9%. This is a great example of running a simple A/B test that provides insight to optimize the effectiveness of your marketing efforts.
In the last case study, we are going to take a look at sign-up forms.
Which form below do you think achieved the best sign-up conversion rate?
You might guess its version 1 because of the TRUSTe privacy certificate but surprisingly, version 2 performed 12.6% better. Why? If the customer has reached this form, chances are they already trust your brand/company enough to consider signing up. The TRUSTe badge might even make them reconsider singing-up or question the authenticity and trust of your brand. Meticulous A/B testing like this may shine a light on some aspects about your website that you might've looked over in the past and present CRO opportunities where you'd last think.
What can we take away?
If you're operating any type of conversion website, A/B testing is a must. Understanding how your users interact with your content is crucial and in today's digital climate it's more important than ever.
Well, that's it for this week, if you missed my article about Google Analytics you can check it out here.
Judd Bobin