Testing Merchandising

Intelligems enables your org to run real-time split tests on different offers. This article will walk you through how to run a Merchandising Test with Intelligems and some best practices. Watch the Looms or follow the steps below.

Setting Up a Merchandising Test from Test Creation Flow

If you'd rather follow written steps, read below.

Step 1: Create a new experience

Navigate to the "A/B Tests" tab in the menu on the left-hand side of the Intelligems app. Once you're there, click Merchandising. Select your type and then "Create".

Step 2: Create your test groups

Create between two and five groups to include in the test by clicking on the ‘+’ button. Name the groups for the test and use the slider at the bottom of the page to allocate what percentage of traffic will go to each group.

The more groups you have, the longer it will take to get statistically significant results. You’ll need about 300 orders for each group in the test to detect a 10% change in conversion with 90% confidence.

Step 3: Select which Offer each group will be exposed to

Once you've created your groups, go to the Modifications step. Here, you will choose what kind of Offers you're testing for each group. You can:

• Leave a group untouched, as a no-offer group

• Build an offer by choosing an offer type and other optional modifications

• Or start by copying the contents of a Personalization that contains an Offer. See more here on how to set up a new Personalization.

These details can be modified anytime during test setup and while the test is live.

Step 4: Set up targeting if needed

Targeting is an optional step. By default, a visitor will be immediately assigned to one of the test groups using its random split-test mechanism. This assignment is determined at the first visit and is stored via a first‐party cookie, ensuring that the visitor remains in the same group on subsequent visits during the offers test period.

The targeting tool allows you to apply specific conditions to certain site visitors. There are a few different ways you can do this:

• Audience Targeting: limit your users based on their device, country, UTM parameters, landing page URL, new/returning, cookies, and much more.

• Currency Targeting: Limit your test to a single currency and/or a list of specific countries.

• Mutual Exclusion: Prevent users from being targeted by related experiments to reduce undesired interactions under the Mutually Exclusive Tests option.

You can learn more about targeting here.

A note on targeting: If you had any targeting options set on the Offer Personalizations you used to define your test groups, these will be ignored in lieu of test-level targeting.

• Personalization-level audience targeting is ignored. The audience targeting used on the test (or lack thereof) is used in its place.

• Personalization-level currency targeting is ignored. The currency targeting used on the test (or lack thereof) is used in its place.

• Personalization-level page targeting is ignored. The page targeting used on the test (or lack thereof) is used in its place.

Step 5: Save your test and start it

Once you have completed all the steps, you'll be able to save your test with the button in the top right. You can then start it whenever you're ready.

Don’t worry, this won’t set the test live yet and you can come back and edit if you need to make changes!