All Collections
Enriched Catalogs
Testing Enriched Feeds
Testing Enriched Feeds

Methodologies for testing your enriched feed designs

Austin Alexander avatar
Written by Austin Alexander
Updated over a week ago

To identify which enriched feed design is lifting performance we will need to run campaigns setup to compare design treatments with each other and a control. Once the testing concludes we will scale the winners into our BAU campaigns.

01. Uploading designs to your catalog

When testing, we will be creating a Multi-Treatment feed that contains all of the looks that we want to run against each other. To do this we will create a new output feed and select multi-treatment mode. Then we will create a new treatment for each design option in our test. After we've added each design to the correct treatment then we will publish the output feed so that we can copy it's URL and upload it to Meta.

After clicking new output feed you should see the new feed modal where you can choose multi treatment mode

To upload your new output feed into your catalog, press the copy URL button and then navigate to Meta Commerce Manager. From commerce manager open your catalog and then click data sources. From the data sources page click "Add Supplementary Feed" and choose data feed. Paste your enriched catalogs URL and choose the main data feed from the following page. Once the upload is complete, you will see that the design treatments have been added as additional image options for each product.


02. Campaign Setup

After our catalogs are created we will setup our test in Ads Manager. We that recommend most brands test using Advtange+ Shopping Campaigns because it improves ROI faster and proves out scalability.


ASC+ Setup

Create a new Advantage+ Shopping campaign in ads manager. In the campaign create an ad for each treatment design being tested and another ad to serve as the control.

In each ad, all settings should be identical. Choose the same catalog for each ad so that the creative is pulling from the catalog that the enriched URL was uploaded to. The quickest way to set this up is to make one initial ad, then duplicate and rename the ad for the rest of the treatments.

After setting up the shell campaign, make sure it is paused and then Navigate back to Marpipe. Open the output feed page for the feed that you're testing with. Click the assign ads button.

Choose the ad account and then the campaign that was just set up. Now you will see the ads for each treatment + the control. Click each ad and assign the relevant treatment. Leave the control ad alone as it will use the plain product feed image by default. Once you've assigned each treatment, click the publish button. After publish is complete you will see the correct treatment when previewing from Meta.

Finding a Winner

Run the campaigns for 14 days and then take the top performing treatment and assign it to the dynamic product ads that you're using in your BAU campaigns.

Scaling the winner

As time goes on you can move new design and treatments out of your main feed from Marpipe. Use the assign ads dashboard to control the look of the DPA creative throughout your entire ad account.

The ROI of enriched catalogs is not fully realized in the testing period. Performance efficiencies emerge when winning designs are scaled into the existing BAU campaigns. Use the scaled performance to determine the ROI of using enriched catalogs.


Additional Notes

Things to Avoid

  • Avoid testing only one design. Performance is going to be highly dependent on the creative. It often takes brands multiple design iterations before they find the combination of elements that unlocks a major win. If your first design doesn’t work, try something completely different.

  • Avoid comparing newly created catalog to an existing catalog. New catalogs take between 30-60 days to completely warm up with audience data.

Did this answer your question?