For your product catalogue to be available in your commerce store, you need to index your products in the search engine. Indexing your products means providing the search engine with a feed of products’ information recognizable and usable by the Search microservice. Basically, indexing represents the input point for a successful search and discovery experience. However, product data indexing has implications on multiple levels, as by feeding your product information to the search engine, you’re also designing your store’s basic structure and navigation.
To help yourself design your feed and its structure, you start by defining your commerce store's extensive data model. Then, you use it as a guideline to create a feed file, a structured inventory of your product information. When the feed file is ready, you can start the indexing process by sending it to the Index microservice. The Index microservice processes the feed file and transforms its information into a feed, or rather, a product’s data collection stored in the search engine. During the search and discovery experience, the Search microservice accesses the product information to retrieve at query time relevant products to the query.
To get started with the indexing process, you need to:
Understand the indexing process
The product catalogue data is the core information for the search and discovery experience in your commerce store. For your catalogue to be available at query time, its data must be stored in the search engine and be accessible by the Search microservice. Indexing the product catalogue means transforming, processing, and storing the product data in a way that the Search microservice can quickly find and retrieve accurate product information.
Do you want to learn more? See Catalogue indexing process.
Prepare the data model
The first step for indexing your catalogue is defining which data you need for your search and discovery experience, and which purposes it serves. To do so, you create your commerce data model, where you define the fields with the related attributes and specify how the search features (or any other related Empathy Platform tools), will use the field data, e.g. as searchable, filterable, facetable, etc.
To help you prepare your data model, check the guidelines about the fields to map in your data model.
Prepare the feed
Once you have your data model ready, you define and organize your product catalogue in a feed file in JSON format In the feed file, your catalogue is represented as a list of objects, where each object identifies one product and describes its properties. In the feed file, mandatory fields are the ones that identify the product (id, name, description, etc.), while optional fields add commercial information that guides the search experience (price, size, brand, stock, etc.).
To learn more about the feed file, the supported fields and languages, and code examples, see Prepare your feed file.
Upload the feed
Once you have your feed ready, you can upload the feed using the Empathy Platform Index API or submit an index job through the Instance Management Console from your Empathy Platform Playboard instance. When you upload the feed, the Feed Validator tool validates the feed and ensures the indexing process can be performed successfully by the Index microservice.
To know more about how to launch the indexing process, see Launching the indexing process.