Choose Your Reading Style
A professional-level summary covering key definitions, frameworks, and exam-relevant points.
The Five Pillars of Data Observability
| Pillar | What It Monitors | Example Issue Detected |
|---|---|---|
| Freshness | Data update recency | Sales table not updated for 6 hours (pipeline failure) |
| Distribution | Value distributions and anomalies | Average order value drops from $150 to $0.01 (data error) |
| Volume | Record counts and completeness | Daily transaction table has 100 records instead of 10,000 (partial load) |
| Schema | Structural changes | Column "customer_id" renamed to "cust_id" without warning (breaking change) |
| Lineage | Data flow and dependencies | Upstream source change impacts 15 downstream reports (impact analysis) |
CDMP Exam Relevance
Data observability is an emerging topic in the CDMP exam, primarily relevant to the Data Quality knowledge area (11%) and Data Integration & Interoperability (6%). Key exam topics include: the definition of data observability and its five pillars, the difference between data observability and data quality monitoring, and the role of data lineage in observability. As data observability becomes more mainstream, its presence in the CDMP exam is likely to increase.