Product Types and Attributes

Learn how to manage Product Types and organize Attributes.

Ask about this Page
Copy for LLM
View as Markdown

A Product Type is a collection of Attributes that acts as a template for Products of the same type or kind. An Attribute is a characteristic of a Product such as color, size, fabric, or any other specification.

You can model Product Types and Attributes either from scratch or with the help of an AI-driven tool, the Smart Data Modeler.
Learn more about Product Types in our self-paced Product data modeling module.

Add a Product Type

  1. Go to Settings > Product types and attributes.
  2. Click Add product type.
    • To create a Product Type manually from scratch, open the drop-down menu and select Add product type.
    • To use the AI assistant, open the drop-down menu and select Smart Data Modeler.
  3. In the General Information section, follow these steps:
    • For Name, enter a unique name for the Product Type.
    • For Description, enter a description for the Product Type.
    • Optional: For Key, enter a unique identifier value for your Product Type.
  4. Click Create product type.

Add an Attribute

  1. Go to Settings > Product types and attributes.

  2. Click the required Product Type.

  3. On the Product Type details page, click Add attribute.
  4. On the Add attribute page, follow these steps:
    1. For Attribute identifier (name), enter a unique identifier for the Attribute. This is an internal identifier and will not be displayed in the storefront.

      Two Attributes can share the same identifier across different Product Types, if they have the same Attribute type and constraints.

    2. For Attribute label, enter localized names for the Attribute. This will be displayed on the Product details page.
    3. Optional: For Attribute description, enter localized descriptions for the Attribute. This will be displayed on the Product details page.
    4. For Attribute level, select any one of the following options:
      The Attribute level option cannot be changed after saving the Attribute.
      • Product: to share the values of an Attribute (or a set of Attributes) across all variants of a Product.
      • Variant: to individually store the values of an Attribute (or a set of Attributes) in each Product Variant.
      Attributes with Product level are not supported by Product Projection Search.
    5. If you selected Variant as the Attribute level in the previous step, then you can choose an Attribute constraint depending on how you want an Attribute (or a set of Attributes) to be validated across all variants of a Product:
      The Attribute constraints field is visible only if you select Variant as the Attribute level.
      • None: the value can be entered individually for each Variant because the Attribute has no constraints.
      • Unique: the value must be different for each Variant of a Product.
      • Same for all: the value is defined at a Product level and is identical for all Variants of a Product.
      • Combination unique: the combination of values of all Attributes of a given Product must be different for each Variant.
    6. For Attribute type, select the data type of the Attribute:
      • Yes/No (boolean): allows to select Yes or No. A boolean Attribute could be, for example, if a product is child-friendly or not.
      • Text: allows to enter a string or (localized) text. A text Attribute could be, for example, the name of an author (if the product is a book).
      • Number: allows to add a numeric value. A number Attribute could be, for example, the weight of a product.
      • Money: allows to set a currency and its numeric value in cents. A money Attribute could be, for example, the value of a stamp (as a product).
      • Date/time: allows to set a date and/or time in ISO 8601 format. A date/time Attribute could be, for example, to display a specific time for Products related to events.
      • Reference: allows to reference another resource by ID. The available resources are: category, channel, key-value-document, product-type, product, review, shipping-method, state, and zone.
      • List (enum): allows to select from predefined values. The values can be set by providing a unique enumeration key and a label for the list item that will be a visible and selectable option in a list. At least one Enumeration Key and Line Item Label must be defined. A list Attribute could be, for example, the different colors of a Product.
        Key and label values defined for an enum.
      If an Attribute can contain multiple values at the same time, select Create a set with this type.

      If selected, the Attribute can only be optional for a Product.

  5. Optional: For Attribute settings, select the settings for the Attribute:
    • (If you selected Reference in the previous step) Type of reference: choose the resources to reference.
    • Searchable: if selected, the Attribute is indexed for storefront search and affects the Product search results.
    • Required: if selected, the Attribute is mandatory and must be filled for all Products or Product Variants assigned to a given Product Type.
    • (If you selected Text in the previous step) Allow multiple line field (an expandable field recommended for longer texts): if selected, allows to use a multiple line field for the Attribute.
  6. Click Add attribute, and then click Save.

Edit a Product Type and Attribute

  1. Go to Settings > Product types and attributes.
  2. Select a Product Type and update the information for the Product Type or Attribute. You can only update the values for Attribute label, Attribute description, and Attribute settings (depending on the Attribute type).
  3. If you updated an Attribute, click Update attribute.
  4. Click Save.

Use Smart Data Modeler to suggest Product Types and Attributes

Smart Data Modeler uses the Gemini AI model. Uploaded content is not used to train Gemini. Results that come from a large language model (LLM) may produce inaccuracies. Always review and validate before use.
To enable Smart Data Modeler for your Project, submit a support request and include your Project key.
Use the Smart Data Modeler, an AI-powered assistant, to set up your Product data model in your Project. It analyzes your existing product catalog and suggests a structured set of Product Types and Attributes, detecting Product Attributes that can be defined at the Product or Variant level. Review the Product and Variant level assignments to verify that shared, non-differentiating characteristics (such as brand or material) are modeled once at the Product level.

Smart Data Modeler works best for simple product catalogs, such as those in retail, rather than for complex scenarios like bundles or composite products.

Generate a Product Type proposal

  1. Go to Settings > Product types and attributes.
  2. On the Product types tab, open the Add product type drop-down menu and select Smart Data Modeler.
  3. Upload your existing product catalog. Supported formats are csv, json, or xlsx. Prefer the original xlsx or json export if available.
  4. Provide business context. Before adding further instructions, select one of the predefined options to help the assistant tailor the proposal.
  5. Optional: For additional instructions, add any specific requirements or context, such as required Variant Attributes, localization needs, and Attributes to exclude.
  6. Click Generate proposal and review the suggested Product Attributes. You can select which generated Attributes to include before accepting the proposal. Minimizing your total Attribute count ensures better storefront performance. Attributes to review are typically non-essential for standard models.
  7. Click Confirm list and continue analysis and review the suggested Product Types and Attributes grouping. You can download it as a json file with all enum values for each Attribute to import the new model to your Project.

Refine the Product Type proposal

Enter further guidance in Your instructions and click Generate new proposal. Examples of effective prompts include the following:
  • Product detail: "Product detail pages show: brand, material, care_instructions...".
  • Search/filter: "Users must search by brand and filter by color and material".
  • Downstream: "ERP requires style_code on every apparel variant".

You can compare versions and return to earlier steps at any time.

Prepare your data sample for Smart Data Modeler

Provide a small, clean, representative sample that shows structure, not a full export.

Include:

  • Representative products for each distinct product family/category root.
  • Unique identifiers to group Product Variants, such as product key and product name.

Include SKU if it is the only reliable Product Variant identifier, for example when there is no product key or other Variant grouping column.

  • Differentiators, such as color, material, size, voltage, fit, brand, format.
  • Evidence of constraints (repeated identical values for sameForAll; unique combos for combinationUnique).
  • Complete or representative enum value sets.
  • Multi-value examples where a Set type is needed.

Exclude or trim:

  • Long marketing prose, FAQs, promo copy.
  • Internal-only utility/audit fields.
  • Pricing and core platform fields already modeled, such as price, slug, and sku.
  • Fields only relevant to downstream systems if not needed for discovery, pricing, compliance, or checkout.
  • Localized columns you do not plan to localize.

The maximum file size for the uploaded file is 4 MB. Remove non-essential columns (category paths, images, stock, prices) to reduce size.

Optimize large catalog data samples

  • Run Smart Data Modeler separately per product family to stay within limits.
  • Trim only excessively long free-text fields keeping enough context to infer structure.

Provide further instructions

  • Specify whether Attributes are mandatory or optional.
  • Specify whether Attributes are Product‑level or Variant‑level.
  • State required localization, even if only one locale appears.
  • List intentional exclusions, such as pricing and analytics fields.

Attribute search best practices

  • Mark an Attribute searchable only if it must drive user search/filter.
  • Over-indexing can slow search.
  • Maximum length for a searchable Attribute value: 10,922 characters.

Test the Product Type model before adoption

  • Review and adjust keys, enum completeness, and constraints.
  • Validate results in a staging Project first.
You can create Product Types in your Merchant Center project from the generated JSON via the API Playground.

Delete an Attribute

After deletion, the data of the deleted Attributes cannot be recovered.

  1. Go to Settings > Product types and attributes.
  2. Select a Product Type and, in the corresponding row of the Attributes table, click the delete icon.
  3. Confirm the action and click Save.

Import Product Types

For more information about importing resources into the Merchant Center, see Import data.
  1. Go to Settings > Product types and attributes.
  2. Click Import product types.
  3. In the Import product types by CSV window, choose a CSV file on your computer to upload, and click Upload and preview.
  4. After the file is uploaded and validated, verify the rows and columns to be imported, and click Continue.
  5. Click Start import to confirm the import process. As the import process is a background task, you do not need to wait for it to complete. To check the import status, go to Operations > Import.

Export Product Types

See export Product Types.

Manage Attribute Groups

By grouping relevant Attributes together, you can alleviate data enhancement workflows—for example, if a set of Attributes must only be edited by the Marketing team, you can create an Attribute Group and grant access to the team for that Attribute Group.

Attributes can be shared among several Attribute Groups for easier management.

If an Attribute is assigned to an Attribute Group, you must have editing permissions to edit the Attribute in the Attribute Group.

Create an Attribute Group

  1. Go to Settings > Product types and attributes.
  2. Click the Attribute groups tab, and then click Add attribute group.

  3. Enter the details for the Attribute Group.

    • For Name, enter localized names for the Attribute Group.
    • Optional: For Description, enter localized descriptions for the Attribute Group.
    • Optional: For Key, enter a unique identifier value for the Attribute Group.
  4. From the Add attributes drop-down list, add Attributes to the Attribute Group. You can add both Variant Attributes and Product Attributes to the Attribute Group.
    You can add Attributes to the Attribute Group only if the Project is indexed. To index, go to Products > Product list and click Index my product catalog now.
  5. Click Create attribute group.
Attribute Groups relevant to a Product Type are also displayed on a Product's details page as individual collapsible panels (in the order of their creation). Unassigned Attributes are displayed under the Ungrouped panel.
Attribute Groups displayed on the Product details page.

Manage Attributes in an Attribute Group

  1. Go to Settings > Product types and attributes.

  2. Select the Attribute Group.

    • To add an Attribute, from the Add attributes drop-down list, add Attributes to the Attribute Group. You can add both Variant Attributes and Product Attributes to the Attribute Group.
    • To remove an Attribute, click the Remove attribute from this group icon in the corresponding row of the assigned Attributes table.
  3. Click Save.

Delete an Attribute Group

Deleting an Attribute Group will not affect the Attribute definitions.

  1. Go to Settings > Product types and attributes.
  2. Select the Attribute Group and click the Delete attribute group icon in the Attribute Group's details page. Attributes that were assigned to the deleted Attribute Group are automatically moved under the Ungrouped panel.
  3. Click Save.