Configure Rule

Now that you have created the rule, follow the below steps on how to configure the rules:

1. After you click the Configure rule, you will be redirected to the rule configuration screen.
2. The left side panel is for configuring the rules, and the right side section is to preview the impact of the configured rules.
   
3. You will see dummy products until you select the product from the search bar.
4. Let’s suppose you selected “Dress” from the search bar, and you will see the following products displayed:
5. Now to apply the merchandising rule or fallback algorithm, let’s see in detail the merchandise rules and fallback algorithm.

What is the merchandising rule?

A merchandising rule is a set of predefined instructions used in e-commerce to determine how products should be displayed, promoted, or organized. These rules are often implemented to optimize the presentation of products, improve the user experience, and drive desired customer behaviors.
Merchandising allows you to control the dynamic arrangement of the products in a search result. There are different types of rules that you can configure to achieve your merchandising goals.

Steps to Implement the Merchandising Rule:

Now let's add the merchandising rule as follows:

1. Follow the merchandising rule by switching the toggle button.
2. Click the Add Rule button you see.
3. You can configure rules like boost, bury, include, exclude, sort, pin, or slot here.
4. Let’s add the Include rule to merchandising as an example.
5. You want to see the earrings, so add conditions accordingly, as seen below:
6. Here, on the left side under Merchandising, you can select “Use Advanced Query Builder.” It allows you to apply advanced queries to that rule using the Query Builder on the top right, below the search bar.
7. In the top right corner, you will see date options, i.e., duration, where you can set the start and end date times for this rule to be applied in that period.
8. Save the rule after configuring it. You also have the option to preview how the rule will look.
   This is what the preview screen looks like:
9. To see each rule in detail, click on the rule name below, which will direct you to its article in detail:
   Inner rules:

• Boost/ Bury To promote or demote products to appear at the top or bottom of search results, you can use the boost/bury algorithm.
• Include - When you want to show only some kind of product in the search results or nothing at all, you can choose this rule.
• Exclude - When you want to remove some kind of product from the search results, you can choose this rule.
• Pin - When you want to pin any products at a fixed place, then you can use this rule, and in the search results, it will always appear at that fixed position you defined.
• Slot - When you want to group products based on some condition to appear in the slot you define, you can use this rule. For example, products with prices less than 50K should appear in slots 1–5, and products with prices greater than 50K should appear in slots 10–20.
• Sort - When you want to apply an ascending or descending sort order to some attribute, like price, for search results, you can select this rule.

Priority Order:

This is the priority order in which rules will work when you configure multiple rules for the same duration and under the same conditions.

The inner rules follow the below-listed priority order.
For example, when you set multiple inner rules like boost and bury, then the bury rule will take priority, and it will be applied to neglect the boost rule.

What is the fallback algorithm?

A fallback algorithm is a mechanism used in a system to provide an alternative course of action or solution when the primary algorithm or method encounters an issue, fails, or is unable to produce a result. When your primary recommendation encounters issues, the fallback algorithm takes over, ensuring a seamless continuation of operations.

Why Do You Need a Fallback Algorithm?

A fallback algorithm can be implemented to handle situations where the primary recommendation engine doesn’t have sufficient data or encounters errors.

The fallback algorithm could offer generic recommendations or other pre-determined suggestions to ensure a user still receives recommendations even in less-than-ideal circumstances.
User Engagement: Users expect seamless experiences. A well-implemented fallback algorithm keeps users engaged, reducing bounce rates even in the face of unexpected hiccups.

Implementing the Fallback Algorithm

Let’s see an example to understand easily:

1. You have selected the “Rose Red Self Smocked Weave Tiered Dress” product, and you are getting the result “No products found.”
   
2. In this case, follow the fallback algorithm by clicking on the toggle button
3. Select the relevant algorithm; select the “Hot New Releases” algorithm in this case, and you will see the products being displayed.
4. Now save the rule and preview it.