What should you A/B test?

Before we get into the specific elements in your marketing funnels that you should A/B test, let’s take the 10,000-foot view approach first.

The overarching objective of A/B testing is to find ways to get more results from the same (or less) investment.

  • More clicks from the same ad spend.
  • More leads from the same amount of site visitors.
  • More sales from the same number of carts created.
  • More bottom-line profits from the same overall investment.

It’s an optimization game. Once you can improve your click-through rates and conversion rates, you can pour more gasoline on the fire and get better results from more investment. That’s the long game we’re playing here.

So with that in mind—strategically speaking, what should you be A/B testing?

It rolls up to two main things:

1. Positioning and marketing messaging (the what)

Messaging is all about how you present your products or services to the market. You likely have plenty of ideas and hunches about what will resonate, but the best way to answer that question explicitly is to run A/B tests.

Try different angles. Test different value propositions. Experiment with different concepts.

If you can determine what messages and positioning are most likely to pique the interest of potential buyers, you can apply (and benefit from) those findings across the board.

2. Customer touchpoints throughout the user journey (the how)

Alongside the messaging itself is how the messaging is presented.

A typical customer journey will have dozens of touchpoints—some predictable, most fairly random and unique. Throughout this journey, though, there will be critical touchpoints that most buyers pass through. These are the places you want to nail over time. Think core landing pages, checkout pages, thank you pages, etc.

Every lesson you learn from testing how you deliver key messages to potential customers can also be cascaded throughout your marketing engine—even things as simple as color schemes and types of imagery.

Why landing pages are the best place to run A/B tests

The single best vehicles for A/B testing both messaging and touchpoint design are your landing pages.

With the right tools, A/B testing landing pages lets you control most of the variables that could otherwise clutter up your results. The more variables you have control over, the more conviction you’ll have in the winning variants in your tests.

Think about it like this:

You run an A/B test on a landing page where the only thing you change is the color scheme from light mode to dark mode. After a few thousand visitors to build statistical significance, the dark mode variant outperforms the light mode variant by a wide margin. You now get to walk away with complete conviction that light text on dark backgrounds is the ideal design style for your audience.