Integration
Overview
Narrative is designed to work with data law firms already have — without requiring complex integrations or IT projects to get started.
Start simple, expand over time:
Start Here
Data Lake
CSV export
1-2 weeks
Enhance
iManage / NetDocs (DMS)
Read-only API
4-6 weeks
Expand
Elite 3E / Aderant (PMS)
Read-only API
6-8 weeks
Phase 1: Data Lake (Start Here)
Most firms have a data lake or warehouse that aggregates exports from source systems. This is the fastest path to value.
What We Need
Time Entries
Time entry ID, Date, Timekeeper ID, Timekeeper Name, Level,
Client Number, Matter Number, Hours, Standard Rate Fee,
Client Rate Fee, NarrativeMatters
Matter ID, Matter Name, Client Number, Practice Area,
Matter Type, Open Date, Close Date, StatusDeal Data (if available)
Matter ID, Deal Value, Value Band, Jurisdiction,
Transaction Type, Industry SectorIntegration Approach
CSV Export
One-time or periodic file export
POC, pilot
Direct Connection
Service principal to data lake
Real-time
What We Don't Need
Access to source PMS directly (we use the lake)
Complex ETL pipelines (we handle transformation)
Perfect data quality (we clean and normalize)
Phase 2: Document Management System (iManage)
Adding DMS integration enriches matter profiles with document context and outcome information. In some instances, firms will have the relevant information already available in their data lake.
What We Access
Workspace Metadata
Workspace ID, matter reference, dates
Link to matter
Document Counts
Count by type (contract, memo, filing)
Complexity signals
Activity Signals
Last modified, version counts
Engagement patterns
Key Document Metadata
Engagement letters, closing memos
Deal characteristics
What We Do NOT Access
Email bodies
Privileged materials without guardrails
Author identities (unless needed)
Note: Document content is accessed strictly for context building and entity extraction, respecting all privilege and security boundaries.
Integration Approach
Protocol: iManage Work API (REST)
Authentication: OAuth 2.0 service account
Permissions: Read-only, scoped to specific workspaces
Refresh: Weekly (metadata changes slowly)
Security Controls
IP allowlist for live infrastructure
All API calls logged in iManage audit trail
Access limited to matter IDs in scope
No data extraction — only metadata queries
Phase 3: Practice Management System (3E / Aderant)
Direct PMS integration provides real-time financial data and richer analytics. This will helpful for any lifecycle management solutions we roll out during our partnership.
What We Access
Live Time Entries
Same as data lake, real-time
Current matter tracking
Rates
Standard rates, client rates, effective dates
Rate accuracy
WIP/AR
Unbilled hours, receivables, aging
Cash flow signals
Write-offs
Amounts, dates, reasons
Margin analysis
Realization
Billed vs collected
True profitability
Budgets
Original, revised, actual
Variance tracking
Integration Approach
Protocol: 3E API (REST) or read-only database views
Authentication: Service account with SELECT permissions
Scope: Filtered by practice area or matter list
Refresh: Nightly (time entries), weekly (rates)
Why This Matters
Data lake exports are typically periodic (daily or weekly). Direct PMS integration enables:
Real-time matter monitoring
Up-to-date pricing for active matters
Immediate visibility into WIP and AR
Continuous profitability tracking
Architecture Overview
Security & Compliance
Data Residency
All data hosted in the relevant Azure or AWS region
No data leaves the dedicated tenant
Compliant with UK, EU and US data sovereignty requirements
Access Controls
Read-only: All integrations are read-only
Scoped: Access limited to agreed matter sets
Audited: All data access logged and available for review
Encryption
At rest: AES-256 encryption
In transit: TLS 1.3 for all connections
Compliance
ISO 27001 compliant (audit actively underway; target certification: December 2025)
SOC 2 Type 2 certification (target: Q1 2026)
Penetration testing active; all critical/high findings remediated
DPIA completed for each engagement
Zero retention for any LLM
Implementation Timeline
Typical POC (Data Lake Only)
1
Kickoff, security framework deployed
2
Data lake access provisioned, schema aligned
3-4
Initial data export, ingestion, transformation
5-8
Review cycles, accuracy validation
9-12
Refinement, POC assessment
Adding DMS Integration
1-2
iManage access scoping and approval
3-4
API connector configuration
5-6
Integration testing, validation
Adding PMS Integration
1-3
3E/Aderant access scoping and approval
4-6
API/database connector configuration
7-8
Integration testing, validation
IT Effort Required
POC Phase (Data Lake)
Data Engineering
6-10 hours
CSV export, schema documentation
InfoSec
4-6 hours
Security review, access provisioning
Total
10-16 hours
DMS Integration
KM / IT
6-10 hours
iManage access scoping
InfoSec
4-6 hours
API access approval
Total
10-16 hours
PMS Integration
Finance Systems
8-12 hours
3E access scoping
InfoSec
6-8 hours
API/database access approval
Total
14-20 hours
Interoperability
Narrative outputs are designed to be useful beyond our platform:
Export formats: JSON, CSV, API access to structured data
BI integration: Feed enriched data into Tableau, Power BI, or internal dashboards
CRM sync: Push matter insights to client relationship tools
SALI alignment (Planned): Future standard classifications for matter types, practice areas
Your investment in data transformation pays dividends across your technology ecosystem.
Last updated