Datum is a master data management platform designed for corporate law firms that need a consistent view of clients, matters, organisations and people across disconnected systems.
Key features and functions include:
Entity resolution and golden records
Datum identifies duplicate records across source systems using matching logic and standardises incoming data. Firm-defined rules determine which source supplies each field of the master record, with human overrides available, and each field's provenance (source system, external register, steward decision, or matching inference) is shown separately.
Corporate hierarchies and matter relationships
The platform links companies to corporate groups and parties to matters, with each relationship dated and sourced. This supports queries such as identifying a firm's total work for a client group.
Data quality and stewardship
Configurable rules turn data issues into queued work items with owners and deadlines, covering match review, hierarchy assertions, and data quality exceptions. The platform supports the SALI Legal Matter Standard Specification (LMSS) for classifying legal work, alongside firm-specific extensions.
Confidentiality, retention and audit history
Ethical walls are enforced on every request, including for administrators and AI agents. The platform maintains retention schedules, a legal hold register, and a change history for every record field, and tracks both when information was true and when the firm recorded it.
AI and agent integration
Through the Model Context Protocol (MCP), AI agents can look up records, resolve entity identities, and retrieve matter or corporate-group context, subject to the same access controls as human users. Agents can propose changes, which enter a human review queue rather than updating records automatically. The platform can also run without AI integration enabled.
Integration and deployment
Connectors are configured as settings files and support databases (SQL Server, PostgreSQL, Oracle), web APIs, file shares, and spreadsheet imports, with working examples for Elite 3E and a generic CRM. Records can be checked against Companies House, CIPC, and OpenCorporates. Data is made available downstream via API, a change feed, and webhooks. Deployment is available in a firm's own data centre or as a managed single-instance service.
Loading...