Skip to main content
Back to E-commerce Dictionary

Data Model

Data management and qualityAdvanced Level

A data model defines what data a system stores and how it relates; a PIM data model structures products, variants, attributes and categories.

Image by · CC BY 4.0

What is Data Model?

A data model describes which types of data a system stores and how they relate. A PIM data model, also called a product data model, defines the entities in your catalog, such as products, variants, families, attributes, categories and assets, plus the rules for each field. In practice the model answers questions like: is color an attribute of the product or of each variant, which attributes are required for a category, and which values are allowed. A product master data model in MDM covers the core identity of a product across the business, while a PIM data model adds the marketing content and channel fields needed to sell it. Getting the model right early saves rework later, because every import, filter and feed depends on it. The attribute modeling guide walks through the choices step by step.

Why Data Model matters for e-commerce

The data model decides what your catalog can express. If size lives in the product title instead of a size attribute, shoppers cannot filter on it and a feed cannot send it as a separate field. If variants are stored as separate products, stock, reviews and content are split over many records. A clear model also makes channels easier. Each marketplace and feed has its own required fields. When your model already stores those values as structured attributes, mapping to a new channel is a matter of linking fields instead of rewriting content. Finally, the model sets the limits for growth. Adding a new category with its own attributes, a new language or a new channel is simple when the model was built to be extended, and painful when every product type was squeezed into the same flat list of columns.

Examples of Data Model

  • 1Entity: Product family Running shoes, Attributes: Brand, Drop, Weight, Surface, Variant axes: Size, Color
  • 2Entity: Attribute Screen size, Type: Number with unit, Unit: inch, Required in: Laptops, Monitors
  • 3Entity: Category Outdoor furniture, Parent: Garden, Inherited attributes: Material, Weather resistance
  • 4Entity: Asset Product photo, Linked to: Variant Navy / Size M, Rule: At least 3 images per variant
Image by · CC BY 4.0

How WISEPIM Helps

  • Product families and variants let you decide which attributes belong to a product type and which ones differ per variant.
  • Attribute management and the Attribute values page keep attribute types, units and allowed values consistent across the catalog.
  • The Category map links your own categories to channel categories, so one model can feed several channels.

Common mistakes with Data Model

  • Storing structured facts such as size, color or material only in titles and descriptions.
  • Creating separate attributes for the same thing, such as color, colour and kleur, so values never line up.
  • Using free text where a fixed list of values is needed, which leads to Navy, navy blue and dark blue side by side.
  • Modeling variants as separate products, which splits stock, content and reviews.

Tips for Data Model

  • List the questions customers ask per category and turn each answer into an attribute.
  • Decide per attribute whether it belongs to the product or to the variant before you import data.
  • Use fixed value lists and units wherever a filter or channel depends on the value.
  • Document the model in one place so everyone who adds products follows the same rules.

Trends around Data Model

  • Models built around channel requirements, so each attribute maps to the fields marketplaces and feeds ask for.
  • More attention to attributes that AI shopping assistants read, such as compatibility and use cases.
  • Extensible models where new product types get their own attributes without a database change.
  • Use of shared classifications such as ETIM, GS1 GPC or the Google product taxonomy as a starting point.

Tools for Data Model

  • PIM systems such as WISEPIM, Akeneo and Salsify, which model product families, attributes and variants.
  • MDM platforms, which model the master record of a product across the whole business.
  • E-commerce platforms such as Magento and Shopify, which offer attribute sets or metafields.
  • Diagram tools such as Lucidchart or dbdiagram, which help you sketch the model before you build it.

Related Terms

Access Control (PIM)Attribute Dependency ManagementAttribute InheritanceAttribute ManagementAttribute mappingAttribute Scoping

Also Known As

PIM data modelproduct data modelproduct information modelcatalog data model

Frequently Asked Questions

A PIM data model is the structure a product information management system uses to store your catalog. It defines product families, attributes and their types, variants, categories and assets, and how they link. It decides what you can filter, translate and send to each channel.

A product master data model, usually in MDM or ERP, covers the core identity of a product used across the business, such as codes, units and supplier data. A product data model in a PIM builds on that with marketing texts, rich attributes, images and channel fields. Many companies use both and sync the master fields into the PIM.

A typical ecommerce model has products grouped into families or types, each with its own attributes. Variants hold the attributes that differ, such as size and color. Categories organize products for navigation, and assets such as images link to products or variants.

Every channel asks for specific fields in a specific format. When your model stores those values as separate, structured attributes, you can map them to each channel. When they are buried in free text, someone has to extract them by hand for every channel.

A rough model helps a lot. Knowing your product types, key attributes and variant logic lets you test whether a PIM can represent your catalog. You can refine the details during implementation.

Still have questions?

Can't find the answer you're looking for? Please get in touch with our team.

Contact Support

Keep exploring

Hand-picked next steps to go deeper.