Optimize Your Conversion Rates With A/B Testing

Optimize Your Conversion Rates With A/B Testing

Which would you choose — Blue or red call to action button? Male or female photo on your campaign landing page? Photo slide show or video for your website homepage?

You have your preferences, and so do the rest of the population. Our individual likes and dislikes are what make us special, and what can make attempting to please the masses seem impossible. Optimizing websites, apps, emails, and other communications that reach, engage, and drive large groups of people requires data-driven decisions resulting from what your target individuals respond to. Not what you or your lead designers like. Instead of relying on what your team “feels” is the right way to go, ask.

A/B testing is a simple experiment through which you will not only get qualitative data that will lead to optimized results, but you can start the journey to increased conversion. And it will put you in good company — businesses and campaigns of all sizes and scale use this method every day to ensure they are releasing and promoting their best product. President Obama raised an extra $60 million during his presidential campaign using A/B Testing. Google ran more than 7,000 A/B tests on its test algorithm in 2011 alone. Now it’s your turn.

It does not matter if your company sells goods or services because what you sell is of value to your customers. You can measure the value you generate through every interaction your customers have with your brand. Online, for every interaction there is data available that you can analyze. If you use this data to run tests then you can produce even more value to your business and your customers.

Are you sure about what will work best for your site? Does the user prefer A or B? Upload and push live two versions of a web page or other digital asset, Version A and Version B. These versions should be identical aside from the one element that you are testing. Then sit back and watch. Which version is seeing clicks, responses, registrations, or orders?

For example, let’s say that a nonprofit organization has a goal of increasing its online donations. They prepare two versions of their donations page: one where the donation amount is fixed and one where the visitors can choose the amount they want to donate. A key decision here is how long will the experiment run. It can be either defined in days or in unique visitors, e.g. 3 months (90 days) or 30,000 unique visits for both options. At the end of the experiment, the results are analyzed and if they are statistically significant, the option that achieved better results is selected.

In its simplest form, you only test one element at a time so you can be sure what is driving the difference. Start with one, then move on to another. A/B testing isn’t a sprint, it’s a marathon. Continually study your audience, adjust your presence, and refine again.

Some examples of areas to run tests on:

  • The call to action’s (i.e. the button’s) wording, size, color, and placement
  • Headline or product description
  • Form’s length and types of fields
  • Layout and style of the website
  • Product pricing and promotional offers
  • Images on landing page and product pages
  • Amount of text on the page (short vs. long)

It’s always a good time to test. Just because you “think” your site is working fine doesn’t mean it couldn’t use some tweaks. Take a page out for a tune-up and see what happens. Even small changes, if they engage more viewers, will smooth the path to increased sales, revenue, leads, downloads, or any other goal of your web presence.

A/B testing not only gives you information about your page, but about your visitors. Analytics can help you see if visitors from a certain geographic area prefer different language, or if new visitors opt for a different call to action than returning customers.

And no matter the response, you’ll have data — hard, undeniable information that helps you better understand your target and, in turn, optimize and increase your customer’s conversion rates.

Popular A/B Testing tools: