Technical interview guide

Semantic Data Modeling for BI

The metrics layer that defines business terms consistently, so 'revenue' means the same thing in every report.

Read: 45 min
Practice MCQs: 25
Interview QA: 25
Edition: v4
Editorial status: Reviewed
Relevant for: Business Intelligence DeveloperData Analyst

Scope: Current dbt MetricFlow, LookML, Power BI, Tableau, Cube, OpenMetadata, BigQuery, and Snowflake guidance reviewed 2026-09-04.

Interview QA

Treat each question like a live interview question: answer out loud first (structure, assumptions, tradeoffs), then open the model answer to spot gaps and rehearse a tighter follow-up.

Curated: 2026-10-05 · Written: 2026-10-05 · Reviewed: 2026-10-03

QA-1

Define grain for an orders semantic model.

QA-2

Prevent metric fanout across semantic joins.

QA-3

Model balances, ratios, and distinct counts correctly.

QA-4

Define a governed conversion-rate metric.

QA-5

Design and validate semantic relationships.

QA-6

Model a many-to-many attribution relationship.

QA-7

Define metric time, calendar, and incomplete-period behavior.

QA-8

Model current-state and historical dimensional analysis.

QA-9

How do you evaluate whether to model business logic in a decoupled headless semantic layer versus directly within a BI platform's native modeling layer?

QA-10

Create metric governance and certification.

QA-11

Handle null, unknown, and not-applicable dimensions.

QA-12

Design a governed multi-currency revenue metric.

QA-13

Implement secure multi-tenant semantic queries.

QA-14

Protect sensitive aggregates and dimensions.

QA-15

Design semantic pre-aggregations safely.

QA-16

Cache semantic query results correctly.

QA-17

Build a semantic-layer test strategy.

QA-18

Design conformed customer, product, and calendar dimensions.

QA-19

Resolve ambiguous and bidirectional relationships.

QA-20

Design an approximate metric responsibly.

QA-21

Build useful semantic lineage and impact analysis.

QA-22

Resolve competing metric definitions across teams.

QA-23

Model late facts, corrections, and period close.

QA-24

How do you handle role-playing dimensions and shared conformed dimensions across heterogeneous fact tables at different granularities?

QA-25