Data Pipeline Summary

This multi-dataset join operation combined 4 input tables (orders, sessions, support, and marketing) with a combined 31 rows, resulting in only 6 output rows—an 80.6% row reduction. The significant loss suggests an inner join was performed, which retains only rows with matching keys across all datasets; if an outer join had been used instead, the output would contain more rows (up to 31) with null values for non-matching records.

The 25 rows removed raises important data quality concerns: this indicates substantial key misalignment or missing values across the joined datasets, meaning most records lack corresponding matches. Before proceeding with analysis, investigate whether this high removal rate reflects legitimate data filtering (e.g., only complete customer journeys) or problematic data quality issues like incomplete key populations, inconsistent identifiers, or upstream data collection gaps.