You'll learn about the different recommendations strategies you can choose from, how to create rules to customize the products in your carousel, how to draw visitor attention to your carousel, and how to create an audience to define who will see it.
Before configuring a recommendation, you need to add it. Go ahead and click Add recommmendations
If your property has multiple catalogs, as is the case for most multi-region retailers, you need to assign a catalog to each recommendtaion. This lets us know which set of products are eligible for a visitor with a speciific language and currency.
INFO We will only show recommendations to users whose language and currency match the catalog you've selected. If your users have multiple languages and currencies, be sure to configure a recommendation for each matching catalog you want to serve products to.
To review the catalogs set up for your property, refer to the product catalog viewer.
If your property only has one catalog, we will skip this step for you.
An audience is simply the group of customers that you are looking to target with your experience.
If you or someone in your team has created audiences for other campaign types, you could use one of those, but you can also build one from scratch.
▸ If you want to use an existing audience and see it in All audiences, select it. If you can't see it, trying searching for it
OR
TIP: To help others in your team find audiences, we recommend using a name that describes what sort of people you are trying to target. An example might be, "People that have less than five conversions that are viewing jackets and are in London."
Select the attributes you wish to use and enter any details required by the attribute like is less than and a value to go with it, like 5
Here's an example for the first part of our audience:
INFO: A super important thing to remember is that audience attributes are And ed together. So in our example, only people with less than five conversions AND are viewing jackets AND are in London will potentially see the experience.
When you are happy with the audience you have created, click Next
Deciding which products to show starts by choosing a strategy. This choice depends on your business goals. For example, you might be looking to increase conversions, build engagement through exposing more of your product catalog, or highlight trending products to inspire your customers.
In the Merchandising Hub, you can choose from one of Google's recommendations strategies or one of Qubit's. For each strategy, we provide details about where it is best used and the business goal. You can find a bit of extra information about each one in Recommendations strategies available in the Merchandising Hub.
Click either Google Cloud or Qubit and then the card for the strategy you want to use
By default, your carousel is introduced with a default headline based on the strategy you selected. You can change this by altering the recommendation's heading.
We suggest tailoring the message to the strategy you've chosen for your campaign. Here are a few examples:
TIP If you're creating recommendations for multiple catalogs that have different languages, remember to translate the headline in each recommendation to match the catalog's language.
Rules allow you to customize your chosen strategy and fine-tune the products shown in the carousel. Rules will help you to:
Each rule consists of a simple condition:
"if people in the audience are viewing"
To this, you add an attribute, for example, "product name," "product category," or "price":
"if people in the audience are viewing products in the dresses category"
Then an action, for example, "blacklist" or "promote":
"if people in the audience are viewing products in the dresses category then promote"
And finally, the products you want to action, in this case, promote, again based on product attributes:
"_if people in the audience are viewing products in the dresses category then promote products in the category shirts"
Let's have a go at creating that rule.
TIP To get a heads-up on the potential impact of your rules before going live with your campaign, we'll update the carousel preview as you add rules.
Look what happens to our carousel when we blacklist products below £20:
Click Edit rules and then Create new rule
Give your rule a name. We'll use Promote shirts with dresses
By default, the rule condition will be people in the audience viewing any product. We'll need to change that, so click any product
Now click Category, is, and enter Dresses in the field provided:
Click Save
Click Promote to increase the likelihood that shirts will appear in the carousel
Now, in Products, click the Category attribute, the operator is, and enter shirts in the field provided
That's it! Click Save and continue to finish up
TIP: Of course, you can create far more complex rules than the one we just added, but we recommend starting with some simple rules and observing the results on the carousel before adding others. Don't forget to check out the FAQs at the end of this article to find out more about this.
So, now we've chosen a strategy and added some rules to customize our carousel:
As we've mentioned, a campaign can have more than one recommendation. This is a great way to differentiate your carousel's behavior based on a visitor's audience or to manage multiple catalogs. To do this, simply click Add recommendations and repeat the steps above for your next set of visitors.
Remember that only customers that are members of an audience and that match the selected catalog(s) can see the recommendations.
When you have a campaign with more than one recommendation, it is possible that a person could be a member of more than one audience in the campaign and potentially see both recommendations. Why?
To get around this potential conflict, you can order your recommendations by dragging and dropping. Customers that fall into more than one audience will be shown the recommendation at the top of the list.
Here you can see a drag and drop in progress:
Once you've set up your recommendations campaign, you'll want to ensure it behaves as you'd expect it to. You can use the in-app preview to test behavior for different audiences, catalogs, and seed products, for seeded placements.
INFO Personalized strategies, like the Google strategies that account for a visitor's history, can still be previewed in the app. The preview will not be personalized, however, and will reflect the experience of a brand new user. Preview the strategies on your site to get a feel for how the personalization behaves
Previewing is a great way to test your campaign by replicating the customer journey. When you preview, you can see the campaign as one of your customers would see it.
We offer two ways to preview:
Click Preview on an experience card to preview the experience
For mobile, click ...on mobile and scan the QR code with the mobile device you want to preview on
INFO: It can be confusing to preview your experiences when your audience condition targets URLs. This is because we use the placement's default preview URL, which may not match your audience conditions. The "Preview campaign" button is a great way to validate that URL matching conditions are working as expected.
Click Preview campaign to preview your campaign
Here's an example of our carousel on web:
Global rules are rules added by your Qubit team when your specific use case is not covered by the rules available in the Merchandising Hub. Please reach out to your CSM at Qubit to discuss global rules.
It is important to understand that rules created for specific campaigns will interact with global rules and any custom rules added by Qubit for your property. What does this mean?
Well, for example, suppose you have a global rule to promote products in the category Bed Sheets - Valences and a second rule for a campaign that also promoted a smaller set of products, perhaps by product attribute, in the same category. In this instance, we would promote all products in the category and promote even further that smaller set of products targeted by the campaign-level rule.
TIP: A good rule of thumb is the more a product is promoted, the more likely it is to be recommended.
It is also possible that a rule for a specific campaign could blacklist a product promoted in a globally-applied rule. In this instance, the product would be blacklisted- blacklisting wins.