Composable data product ecosystem¶
Context¶
Organisations often respond to inconsistent reporting by creating another central integration model or very wide analytical table.
Pattern¶
Keep source-aligned data products small enough to retain their business context and authority.
Make their semantics, contracts and relationships explicit so that higher-order products can compose them without discarding provenance.
Where shared dimensions, metrics or process views are required, derive them from these governed building blocks rather than recreating meaning independently.
Result¶
The ecosystem behaves more like a set of composable semantic building blocks and less like a sequence of opaque ETL layers.
Guardrail¶
Composition is not permission to create a new God table. The combined view should preserve grain, lineage and the boundaries of the underlying processes.
Join the argument
Agree, challenge or add evidence
Architecture improves when its assumptions are challenged. Say which kind of contribution you're making:
- Challenge I disagree because…
- Evidence We've seen this too…
- Question How would this work when…?
- Alternative Another way to approach this…
- Extension This also implies…
Please keep clients, employers and colleagues unidentifiable. Strong contributions may be quoted and credited on this site. Read the contribution guidelines and the privacy notice.
Comments are provided by giscus and stored on GitHub. Posting requires a GitHub account.