Case Study: Cerved (Italy’s credit‑risk leader) achieves 99% faster real-owner identification with Neo4j graph database

Companies and their Economic Connections Are More Transparent Thanks to Big Data Intelligence

Cerved, Italy’s leading credit-risk analyst and one of Europe’s largest rating agencies, manages data for ~34,000 customers and hundreds of terabytes from diverse sources. It faced a critical challenge: quickly and accurately identifying the “real owner” of companies (those with indirect or direct control of ≥25%+1 share) for anti–money‑laundering compliance and real‑time customer services, which proved inefficient on a relational database.

Cerved built the Aracne project on Neo4j’s graph database, replacing its relational system to enable efficient network analysis. The switch cut owner‑identification times from about 12 seconds to 67 milliseconds (≈99% faster for up to 15 ownership links), simplified the architecture, reduced maintenance costs, improved precision and security, and allowed the solution to be rolled out across teams and new applications—boosting customer confidence and operational efficiency.

Cerved

Stefano Gatti

Innovation & Data Sources Manager