At a glance
Stores, versions and retrieves documents such as invoices, contracts and ID scans.
Key data
Receives from
- Billing
- Contracts
- KYC
Sends to
- eCare
- CRM
- Audit
How it works
The main steps, end to end.
- 1Receive the document from a business system
- 2Store with metadata and version
- 3Apply access and retention rules
- 4Retrieve for customers and agents
- 5Archive or delete when retention ends
Core Capabilities
What the Document Management Service provides
Document Storage
- Stores all platform-generated documents in a structured, indexed store
- Document types: contracts, invoices, KYC verification records, dunning letters, payment slips
- Each document stored with metadata: type, subscriber ID, order ID, creation timestamp, source service
- Binary content stored separately from metadata for efficient retrieval
- Supports multiple content formats: PDF, HTML, JSON, image files
Document Retrieval
- Retrieves documents by subscriber account, document type and date range
- Single document retrieval by document ID for direct link access
- Supports document list queries with filtering and pagination
- Streamed download for large documents to avoid memory pressure
- Accessible by internal services (billing, care portal, contract management)
Versioning
- Each update to a document creates a new version — previous versions are never overwritten
- Version history accessible for audit and regulatory review
- Latest version flag for quick retrieval of current document
- Reissue tracking: invoice and contract reissue count per document
Access Control
- Documents accessible only to authorised internal services and care agents
- Subscriber-facing documents exposed only via authenticated portal session
- Role-based access: billing agents see billing docs, care agents see all
- Audit log of every document access: who accessed what and when
- Temporary signed URLs for secure subscriber document download
Retention & Expiry
- Configurable retention periods per document type: contracts, invoices, KYC
- Documents marked for deletion after retention expiry — not immediately purged
- Regulatory hold: retention pause applied to documents under compliance investigation
- Automated archival of documents beyond active retention period
- Deletion audit trail: records document identity and deletion timestamp
Document Audit
- Full access log per document: service name, user ID, action, timestamp
- Document generation log: source service, template used, subscriber
- Tracks document delivery: sent to subscriber, delivery status, channel
- Provides evidence for regulatory document delivery compliance
Storage & Persistence
How and where DMS stores its data
Object Store
- PDF and binary document files stored in object storage
- Content addressed by document ID
- Scalable storage with tiered access: hot (recent) / cold (archive)
Document Data Store
- Document metadata index: type, subscriber, order, timestamps
- Version records per document
- Retention policy state per document
- Access control attributes
Audit Log Store
- Every document access event: user, timestamp, action
- Document generation records
- Deletion audit trail for compliance
Design Principles
Key architectural decisions behind the Document Management Service
Content-Metadata Separation
Binary document content is stored in object storage, separately from document metadata in the structured store. This allows metadata to be queried efficiently without loading large binary files, and enables content to be stored in a cost-optimised tiered storage backend.
Immutable Versioning
Documents are never overwritten. Every reissue or correction creates a new version, with the previous version remaining accessible. This guarantees that the document a subscriber received at a specific point in time can always be reproduced exactly, which is essential for billing disputes and regulatory compliance.
Retention-first Deletion
Documents are not deleted on demand — they follow a configured retention lifecycle. Even when marked for deletion, a regulatory hold can pause the process. This prevents accidental or premature deletion of documents required for ongoing compliance or legal review.
Extract data from IDs, forms and contracts automatically.
Find documents by meaning, not just file names.
More in 30 AI & ML use cases and AIOps for BSS.