``````
granify_nodes.png

The Granify E-Commerce Blog

Prioritizing A/B Tests for the Highest Impact

Conversion optimization is one of the key techniques to getting the most out of your traffic, especially in eCommerce, where even a marginal improvement to conversions can dramatically affect your revenue. The issues of what and when to test are critical to developing a strategic approach to A/B testing for retail eCommerce professionals. Every test run has an opportunity cost; if your efforts do not target the areas with the highest possible impact then you are not getting as much out of your tests as you should.

This article outlines strategies for prioritizing A/B tests for the highest impact to ensure that you're making your tests count.

Ask Yourself: "Where's the money?"

Marketers are too often focused on conversions in the wrong places. When dealing with matters of eCommerce revenue, you should take your conversion funnel and turn it upside down. The upward pointing triangle you're looking at corresponds to the expected value of each user in the funnel based on their probability of purchasing. More accurately, the expected value of each user = (probability of converting * the value of items in their cart). As they move closer to checkout, they signal their intent to continue with the purchase.

inverted funnel.jpg

This method of visualization demonstrates the value to be gained by testing the areas closest to the money. Even a small improvement to the checkout process can make a significant increase to your revenue.

Okay, let's turn the conversion funnel back around.

The top of the funnel can be seen as the persuasive end. Only the most deliberate of shoppers will focus wholly on purchasing what they came to the site for. Those who clicked your site at a time when they weren't ready to begin shopping will bounce. The vast majority of users at this point still need to be persuaded to buy from you and move forward with their purchase. You have their attention, but need their interest and desire.

The bottom of the funnel is the action end. This is where the intention to purchase is put to the test. The timeliness and fit of the checkout experience with the user's context is the most important factor here. Things that they are concerned about will be further deliberated at this stage, and the users will keep mentally weighing whether the purchase is a good idea based on a multitude of factors.

Below is an infographic highlighting some of these common shopping objections for retail eCommerce.

Top-Objections-by-Industry (1).png

How Should You Approach Prioritization?

There's not necessarily one 'correct' way to prioritize your tests as the various challenges and levers facing your business at any given time will give some credence to one method or another at different times.

First, we want to list out different methods of prioritization for your consideration:


  • Prioritizing based on perceived opportunity: Testing based on the expected impact of the changes on conversion rate and other key metrics (such as revenue per customer). This includes changes to template pages that can have a site-wide impact.
  • Prioritizing based on pain points: Unexpected heatmap activity, lower than normal metrics, or other data uncovered suggesting a problem. See the top "% Exit" pages on Google Analytics for the most obvious areas of improvement.
  • Prioritizing increases to revenue per customer: Focusing exclusively on increasing order value per customer, which means identifying areas where upselling may occur and facilitating browsing and product discovery. Time on site is a key metric for this.
  • Prioritizing pages based on cost of traffic: Optimizing landing pages for paid campaigns and other areas where traffic has a high cost and a high volume.

This gives you a launching pad to dig deeper in to what's happening with your shoppers, and which tests to invest your immediate time in to quickly identify the underlying issues.

So now that we've identified where the money is, what pain shoppers are feeling, and different ways in which we can prioritize the test, how do we decide which to go with first?

Finally, 3 Strategies for Prioritizing A/B Tests

For the majority of eCommerce marketers, the following three strategies are the best for finding the pages with the most opportunity for improvement:


  1. Funnel bottlenecks: Map out your conversion funnel and find the bottlenecks, or the pages with the lowest micro-conversions. If you're using a common URL slug for various page types, you can figure this out in Google Analytics. Otherwise, plot the data in Excel to figure it out manually!
  2. Ranking your pages by traffic vs. conversion rate: For eCommerce sites, take your product pages and sort them by the highest traffic and lowest converting product pages.
  3. Testing closest to the money: For the reasons mentioned above, testing closest to the money can have the highest impact on your bottom line. Start with your checkout and move backwards through the process to see the most direct improvement in conversions and revenue.

These strategies will allow you to plan out your A/B tests and make significant impact with your optimization efforts.


At Granify, we constantly geek out on conversations like these, and we really love when we get to apply what we love and learn to work with our clients.

If you'd like a partner and subject matter expert to roll their sleeves up and dive in to the A/B tests that will drive results for your business, we'd love to chat with you for 15 minutes or so and see how we can help.

Request A Consultation

SHARE THIS STORY | |

Looking for more? Our readers also liked...

Want weekly insights in to the world of e-commerce optimization?

Recent Posts

Search