What is a data product?

IBM Data Product Hub as a Service is a self-service solution that is used by data-driven enterprises to share data products. On Data Product Hub, data producers can publish curated data products to share with data consumers in their community. Data consumers can easily access data products for their business needs.

Data producers use the full-service Data Product Hub solution to package, productize, and share their data-rich assets. Data producers manage data as a product to provide more value to teams across the organization. When a producer publishes their data product, they assign a short description, business domain, and recommended usage to guide consumers who are searching for a data product. Producers ensure that published data products are accurately represented. Producers ensure that the items in a data product are accurate and available. Producers are also responsible for upholding any relevant clauses in the data contract.

Data consumers use Data Product Hub to discover and access data products for their business needs. Consumers can read descriptions of the data product and decide whether it fits their needs. Consumers are responsible to read and uphold any relevant clauses in the data contract. Data consumers quickly and easily access the right data by using an intuitive interface.

Data products can contain one or more data or data-related assets. They are curated, packaged, and distributed to be easily accessible and reusable. Unlike data assets in governance catalogs, data products are managed as products with lifecycle management, wide distribution, and multiple purposes to provide maximum business value.

Packaged and curated

A data product is a collection of curated data or data-related assets that are packaged for reuse and distribution on Data Product Hub. Data products can contain data, as well as models, dashboards, and other computational asset types. To provide flexibility, you can use several methods to add items to a data product. For example, you can add items using queries, from a direct source, or from a connected catalog.

Managed as a product

Data products are assigned to a designated owner who manages them through the draft, published, and retired stages. Similar to other products, data products have a market demand, value proposition, and defined terms and conditions. The typical workflow for creating and managing data products iterates through use case definition, preparing the data, defining the data contract, testing, and publishing.

Using Data Product Hub, you can apply product management principles to your data products, including:

Define the use case for the data product: Clearly define the purpose and objectives of the data product, including what problem it solves or what value it provides to users.
Assign an accountable owner: The owner manages the lifecycle of a data product from initiation to end of life.
Restrict distribution: Designate approvers for data products to restrict distribution to approved consumers.
Prepare the data: Ensure that the data assets are high quality, accurate, and secure. Remove any Personal Identifiable Information (PII). Choose an appropriate data source and delivery method.
Include a data contract: The data contract establishes transparency by describing the Terms and Conditions and Service Level Agreements.
Testing and Validation: Thoroughly test the data product before publishing.
Publish updates as needed.

Optimized for large-scale distribution

Data products are highly reusable across any number of consumers who are community members. They can be delivered through multiple delivery methods for maximum availability.

Associated with a data contract

Data products are associated with a data contract that outlines the terms and conditions of usage. The data contract provides assurance on both ends for data products that are distributed across teams.