Data Quality Controls, Validation, and Enrichment
All data governed by RDRS is subject to data quality controls designed to ensure that terminal-ready outputs are evidence-backed, internally consistent, and suitable for downstream analytics and reporting.
Validation and enrichment workflows are applied across both asset-level and policy-level data. Where required fields are missing, incomplete, or fail statistical or qualitative validation checks, values are reviewed and validated against primary source documentation retained within the platform.
Only data that meets RDRS validation and traceability standards is approved for use in Market Terminal outputs.
Validation and Enrichment Methods
Data validation and enrichment may include one or more of the following methods:
- Automated validation and re-extraction using a purpose-built AI extraction and validation system
- Iterative improvement of automated workflows through multiple human-in-the-loop (HITL) training cycles
- Manual review by the data team where automated validation is insufficient or where policy or asset structures require additional scrutiny
All validation workflows prioritize traceability to primary evidence. Data values that cannot be conclusively validated against documentation are excluded from terminal-ready datasets.
Broker and Brokerage Data Controls
Broker and brokerage identifiers are captured at the time of data collection and sourced from insurance documentation and/or documented communications with agents or brokers.
Where broker or brokerage entries appear inconsistent, ambiguous, or anomalous, they are subject to additional verification, which may include validation against external reference sources, prior records, or supporting documentation. Only verified broker and brokerage records are eligible for inclusion in Market Terminal outputs.