Consumption for Analytics Engineers
Consumption is how governed metrics and models reach the people who use them: through BI tools, spreadsheets, APIs and notebooks connected to the semantic layer or marts. The goal is that every tool queries the same definitions instead of copying logic into separate reports.
Stage: Metrics and Semantic Layers · Part of the Analytics Engineer roadmap · Last reviewed 19 November 2026
What to learn in Consumption
- Connect a BI tool to a mart or semantic layer
- Query a governed metric by dimension from outside dbt
- Expose a model to a spreadsheet user safely
- Set access so users only see permitted data
- Check that two tools return the same number for one metric
Consumption interview questions
What is the benefit of serving metrics through an API or semantic layer?
Clients request a metric by name and dimensions and get a consistent result, without writing SQL. This keeps logic centralised, reduces mismatched numbers and lets many tools use one definition.
How do you control who sees what in consumption tools?
Use warehouse roles and grants, row-level policies where needed, and restrict sensitive columns. Give BI tools service accounts with the minimum rights, and review access regularly.
Where Consumption fits in the Analytics Engineer roadmap
Consumption belongs to the Metrics and Semantic Layers stage (intermediate level), which takes about 2 hours. One definition of revenue.
Other units in this stage
Back to the full Analytics Engineer roadmap