Automating Compliance Review for a Global Investment Firm
A global investment firm was processing 10,000+ compliance documents every month. Their review backlog stretched three weeks. Manual analyst work cost $2M a year. We deployed a Document AI pipeline that automated extraction, classification, and risk-flagging. Review time dropped 78%. The compliance team got their time back for actual judgment work.
The client manages $40B+ in assets across equities, fixed income, and alternative investments. Their compliance team processes regulatory filings, trade confirmations, counterparty agreements, and internal audit documents. All of it falls under FCA and MiFID II requirements. Before this project, every document review was manual. Analysts read each document, extracted the data, and typed it into the case management system by hand.
Three problems hit at once. Documentation volume had grown 40% over three years. Headcount stayed flat. A three-week review backlog meant the team was routinely missing time-sensitive regulatory deadlines. That created real compliance risk.
10,000+ documents per month processed entirely by hand
Average review time of 18 days per document batch — regulatory deadlines frequently missed
23% of analyst time went to data entry — none of it required judgment
$2M a year in analyst hours on tasks that could be automated
No audit trail for extraction decisions — the team could not prove data lineage to regulators
We built a multi-stage Document AI pipeline. It ingests documents and runs OCR via AWS Textract. Custom NER models extract financial entities — trained on the client's own document corpus. A classification layer routes documents by type and risk level. The pipeline pushes structured data into the existing case management system via API. Human reviewers handle edge cases and give final sign-off.
Document corpus audit: six weeks of analysis to classify document types, define extraction fields, and set accuracy baselines
Custom OCR + NER pipeline built with LangChain, GPT-4, and AWS Textract — fine-tuned on 2,000 client documents
Risk scoring model that flags documents needing human review based on content and regulatory exposure
API integration with the client's existing case management system — zero workflow disruption
Audit trail layer: every extraction logged with source text, confidence score, and model version
Human-in-the-loop workflow for flagged documents — analyst judgment stays where it matters
Within 90 days of go-live, average review time dropped from 18 days to 4 days. Analyst time on manual data entry fell from 23% to under 5%. The system now processes 10,000+ documents a month at 91% extraction accuracy. That exceeds the client's 88% threshold for straight-through processing.
Review time: 18 days → 4 days (78% reduction)
$1.8M in annual analyst cost savings in Year 1
91% extraction accuracy — exceeding the 88% straight-through processing threshold
Zero regulatory deadline misses in the 6 months post-deployment
Full audit trail now available for regulator inspection
“Norvik didn't just build us a tool — they transformed how our compliance team operates. We went from drowning in paper to having real-time visibility into our review pipeline.”
Sarah Mitchell
VP of Compliance, Global Investment Firm
Client identity is withheld under NDA. The figures reported here were verified against the client's own internal reporting at project close.
Fine-tuning on 2,000 client documents beat a general model by a wide margin. Domain-specific training data matters more than volume.
Keeping humans in the loop for flagged documents sped up adoption. The compliance team trusted a system that didn't try to replace their judgment.
Design the audit trail first — don't retrofit it. The data lineage architecture shaped every other decision.
Key Results
Services Engaged
Technology Stack
Facing a Similar Challenge?
Let's discuss your specific context and what results are realistic for your organisation.
Get in touch