Setting up rule-based recommendation
On this page
Understanding rule-based recommendation
Rule-based recommendation are generated by the rules set up by your store. The recommended products will base on product attributes such as collection, vendor, tag, metafield, product type.
There are 2 recommendation types that can use rule-based recommendation Frequently bought together and Related items.
- Note: This feature is only available for App Lib V3, please contact us for further instructions.
Set up rule-based recommendation
As a basic flow to set up a recommendation widget, to create an rule-based widget in the app’s dashboard:
- Go to Recommendation > Manage recommendation widgets.
- Click button Add new widget.
- On the Page type section, select Product page or Cart page> button Create (currently we only allow the rule-based for Frequently Bought Together and Related items type on the Product page and Cart page).
- On the step General, select recommendation type Frequently bought together or Related items > button Create.
- Toggle the button Rule-based recommendation to enable the feature.
- Select the rules to display recommended products. The recommended products will have the same attribute(s) as the source product.
- Same collection
- Same product category
- Same product type
- Same vendor
- Same tags
- Same product metafield value
- You can set up up to 3 rules of “Same product metafield value”.
- The list of product metafields is linked to the metafields added on the Tool > Metafield feature. Remember to sync recent updates when you add a new metafield from Tool > Metafields.
The maximum number of rules that can be enabled is 4.
When multiple rules are enabled, it will apply logic AND. The more rules a recommended product matches to the source product, the higher ranking it is.
To change the rule priority, you can drag and drop the selected rules. The highest rule will be considered to show the products matching this rule in the first position on the storefront.
- Products in the same collection as the source product will appear first in the recommended products list.
- After that, products from the same vendor as the source product will be displayed.
You are enable two rules: same collection (1st) and same vendor (2nd)
→ All products that match the same collection and same vendor will be considered as recommended products.
We have ignored fields so that you can exclude specific collections, tags, and metafields from the model. The input data for the model will not include these ignored values.
Source product has 3 tags A, B, C. Then you ignore tag B ⇒ Model will understand that the source product only has tag A, C → it will recommend the products that have the same tag A and C as the source product.