Research

Key References

• Fanout deep dive — canonical explanation of fanout, GROUP BY pitfalls, and sidecar pattern
• Fivetran dbt packages fanout analysis — 25-model analysis across 5 Fivetran packages identifying real fanout patterns
• Inspector output schema — current profiler output format (grain/fanout fields will extend this)
• Profiler contract — versioning policy for inspector output

Key Findings from Fivetran Analysis

1. Pre-aggregate then join is the dominant safe pattern (17/25 models). This is what fanout warnings should recommend.
2. NetSuite has real structural fanout risk — multi-period currency conversion creates grain expansion. Grain detection would catch this immediately.
3. No packages document grain explicitly — validates the need for automated grain inference.
4. 5 risk patterns identified: pre-aggregate-join (safe), multi-period expansion (risky), SUM-after-dimension-join (watch), UNION-then-aggregate (safe), star-join-of-sidecars (safe).

Existing Inspector Signals Useful for Grain Detection

From the current profiler output: - key_role=primary_key → direct grain signal (100% unique + name matches id pattern) - key_role=foreign_key → relationship signal (points to parent table grain) - uniqueness_ratio → composite grain candidate detection - role=identifier vs role=dimension → grain column vs grouping column - is_incremental + role=time → temporal grain component - distribution=unique → likely grain column - row_count → denominator for uniqueness checks