Why this matters
An answer to a business question spans several tables, a few transformations and some notes on how to use them. A data product packages that set as one thing with one owner.
Before you start
Draw a product around a consumer question
The test of a well drawn data product is whether a consumer can describe what it is for in one sentence. Grouping assets because they came from the same source system produces a folder. Grouping them because they answer the same question produces a product.
A product names an owner
A data product names someone accountable for it. An asset with no owner has nobody to ask and nobody responsible when it drifts.
Products are how a mesh becomes navigable
In a domain oriented estate, products are the unit consumers browse. They sit above individual assets and below the domain, which is the level of granularity at which someone outside your team can actually find what they need.
The interactive demo
Work through the demo below. It runs in your browser with nothing to install. Watch for:
- Creating a data product
- Managing the assets inside it
- How data products integrate into the wider data ecosystem
This demo is also available on its own at the Learning Center.
After this module you should be able to
- Draw a product boundary around a consumer question rather than a source system
- Assign and express ownership of a product
- Explain how products relate to domains and to individual assets