⭐️ Testing catalog ads in Meta ⭐️

Everything you need to know about testing catalog ads in Meta

Table of Contents

Why test your catalog ads?Which type of test is right for you?How to testA/B testingSimple testing

Why test your catalog ads?

Testing is one of the most important things in catalog ads.

Testing different designs allows you to continuously see what performs well and adjust your design based on the learnings - always improving design quality and performance.

It is never possible to predict what design will work best for a specific brand - so it is only by testing that we can prove which design concepts work and which don't.

  • With each test, you learn more about what performs well
  • Once you find a winner, you can edit it into i.e. campaign designs
  • Answer specific questions, i.e. "Does including price improve ROAS?"

Which type of test is right for you?

You can test in a few ways, so choosing the right one is essential for optimal learnings.

If you just started with Confect and want to test - use A/B testing.

If you're using Confect and want to try new designs - use simple testing.

A/B testing
 Allows you to see the exact performance differences between the designs.

Details:
- Quantifiable results (i.e., "Confect design 1# performs 31% better")
- High reliability: no biases or influences
- Best suited for your first A/B test
Simple testing
 The Meta algorithm will use multiple designs and distribute the budget into the best-performing ones.

Details:
- Effortless setup
- Actively optimized for performance
- Best suited to try out new designs quickly

How to test

A/B testing

In A/B testing, all the designs must have an equal budget, assigned separately per each design.

This can be done through:

  • A new Manual sale campaign - one AD SET per design
  • New Advantage+ campaigns - one CAMPAIGN per design

Also, using the "A/B test" tool inside the Ads Manager is vital for reliable results.

We recommend using one Manual sales campaign and testing between the ad sets; but you may also use multiple Advantage+ campaigns and testing between campaigns.

Simple testing

As the name suggests, this type of testing is very easy to set up.

Simple testing should not be seen as a direct replacement for A/B testing - it can be a great way to try multiple designs simultaneously and see how they perform. Still, it cannot provide a reliable and quantifiable difference (i.e. "Design #2 performs 31% better than Design #1").

You can conduct a "Simple test" by creating an Advantage+ campaign, simply adding one ad per design, and setting the budget on the campaign level.

This will lead to Meta running all these different designs and trying to give the budget to the highest-performing creative.

Do this for simple testing:

  1. Create a new Advantage+ campaign (don't use an old/existing campaign)
  2. Set the budget for the campaign
  3. Add one catalog ad for every design you want to try
  4. Switch the design variants (set up and publish the ads before switching)

Simple testing is especially handy if all the designs have the same theme - then, you can quickly and easily optimize for the best-performing version.

(This example of True Classic's designs comes from facebook.com/ads/library)