Business Intelligence for Analytics Engineers
Business intelligence is the layer where data becomes dashboards and reports. For analytics engineers it covers how BI tools model data, define joins and calculations, and stay fast, because a well-shaped mart and sensible extracts matter more than clever dashboard features.
Stage: BI and Enablement · Part of the Analytics Engineer roadmap · Last reviewed 19 November 2026
What to learn in Business Intelligence
- Connect a BI tool to a mart and define relationships
- Keep heavy calculations in the warehouse instead of the dashboard
- Use extracts or caching to speed up slow dashboards
- Reduce the number of queries a dashboard fires on load
- Explain why joins in the BI layer can give wrong totals
Business Intelligence interview questions
Why do dashboards get slow and how do you fix them?
Common causes are wide tables, many visuals each firing queries, heavy calculated fields and unfiltered scans. Fix them with pre-aggregated marts, fewer visuals, filters on partitions and caching or extracts.
Should business logic live in the BI tool or in dbt?
Prefer dbt or the semantic layer, where logic is versioned, tested and shared across tools. Logic in a BI tool is hard to review and gets duplicated, which creates conflicting numbers.
Where Business Intelligence fits in the Analytics Engineer roadmap
Business Intelligence belongs to the BI and Enablement stage (advanced level), which takes about 3 hours. Make data used.
Other units in this stage
Back to the full Analytics Engineer roadmap