A/B Testing Your Landing Pages - [Marketo Success] Guide Sneak Peek

Marketo Employee
Marketo Employee

This week, we prepared another excerpt from our brand new Marketo Champion-authored ebook, titled [Marketo Success] Guide. The [Marketo Success] Guide will give marketers detailed insights to navigate Marketo like our most successful customers and help you prove your Marketing team’s impact to your organization. Check out an excerpt from the ebook below written by Alex Greger, Marketing Automation Manager at Hitachi Vantara & Lauren Aquilino, Senior Consultant at Revenue Pulse to learn how they A/B test their landing pages in Marketo.

A/B Testing

What is it

Only 52% of those that use landing pages also test them to improve conversions. (1) Testing allows you to optimize conversion rates. This is a native, and frankly underused, feature. In Marketo, A/B test groups allow you to compare conversion rates of specific landing pages and optimize for the best results.

Why use it

The real question is - why not use it? A/B testing allows you to quickly gain insight into what’s working and what’s not. You’ll be able to gain valuable insights on the UX preferences of your audience.

How to use it

Landing pages are tested using a feature in Marketo called Landing Page Test Groups. First, you’ll create the landing pages you would like to test against each other - two or more. Ideally they would have forms. While still unapproved, right click one and select convert to test group.


Add the landing pages that you would like to test, and approve your test group. Once your landing pages are approved and live, you’ll be able to compare stats of the landing pages in your test group.


When testing, every component is up for adjusting. However, there are some factors that may affect conversion more than others. Some examples include:

  • Form placement and length
  • Social proof
  • Headline
  • Copy
  • Images
  • CTA
  • Button Color

4 Keys to testing include:

  1. Define goals and baseline metrics
  2. Keep it simple and choose one variable at a time
  3. Test often
  4. Maintain a log of results

We hope you enjoyed reading this excerpt from the upcoming [Marketo Success] Guide: Best Practices from Marketo Champions. Keep an eye out for more tidbits from the new ebook that is launching in early 2019. Have questions about the content? Let us know in the comments!


(1) http://www.wordstream.com/blog/ws/2017/08/02/conversion-rate-statistics

Level 1

Really helpful! Thanks for breaking it down step-by-step Katie - we can't wait to test our LPs now  

Level 2
Level 2

Within the Landing Page Test Group  - In order to roll out the winning Landing Page - do you just Unapprove one of the Landing Pages and leave in the Winning Landing Page? 

Level 2

Yeah, my question exactly @JLM . What do you do once a winner is chosen? Unapproving the loser(s) seems like a messy way to handle it, though I could see it being useful to have it remain a test so it's a running log.

Level 4

Any insight on what happens to the loser?

Level 2

Wouldn't you just redirect the loser to the winner's URL?

Level 1

@Katie_Pedroza2 will it be okay if my LP are built on two different templates and i put them in AB testing?

Marketo Employee

@Manahilsami1 I believe you can apply two different landing page templates. You still have to add them to the same Landing Page Test Groups as stated in this product documentation. Hope this helps.

Level 5

HI @Manahilsami1 ,

Technically YES you can use two or more different templates to A/B test as mentioned by @achiu

But this way it become very tough for you to understand which component of your LP is working best with your visitors since you are changing the entire template.

As also indicated in this article its always advisable to test One component at a time, this will give you most insight on what is going well with your design.


For example, you are using the same form on the LP but have changed the position of form & also the Button text (because of different templates) then you will not be able to judge why is your conversation increased as there are two variation on your LP it might be because of position of form or because of the button text.


So i will advise to always use one component at a time to A/B test.