Big Data 13 min read

ZhongAn Insurance Data Platform: Digital Transformation, 4633 Framework, and Real‑time Data Warehouse with StarRocks

This article details ZhongAn Insurance's digital transformation through its 4633 data‑centric framework, the architecture of its JiZhi data platform, the challenges of its original ClickHouse‑based real‑time warehouse, and how migrating to StarRocks improved performance, scalability, and operational efficiency across advertising and insurance use cases.

DataFunSummit
DataFunSummit
DataFunSummit
ZhongAn Insurance Data Platform: Digital Transformation, 4633 Framework, and Real‑time Data Warehouse with StarRocks

ZhongAn Insurance, a fast‑growing internet‑focused insurer, achieved 200 billion CNY premium in 2021 and relies heavily on data, with engineers comprising about 50% of its staff.

The company’s digital framework, called 4633, consists of four layers—Application, Algorithm, Platform, and Guarantee—emphasizing business value over pure performance and guiding data‑driven initiatives such as incremental growth, loss reduction, and market expansion.

The JiZhi data platform implements this framework by integrating data processing, machine learning, and governance into a unified pipeline, supporting both internal analytics and external marketing systems while providing fine‑grained, real‑time data access for various stakeholders.

Initially, the real‑time warehouse relied on ClickHouse for both offline and streaming workloads, but increasing concurrent queries, multi‑table joins, and operational overhead (ZooKeeper dependency, lack of automatic resharding, limited DML support) caused severe performance degradation.

To address these issues, ZhongAn migrated to StarRocks, which offers superior multi‑table join performance, better concurrency handling, and simpler operational management, achieving 3‑5× speedups in Flink‑on‑StarRocks pipelines and comparable single‑table performance to ClickHouse.

The platform’s core modules focus on answering three questions: what happened (fine‑grained, second‑level visualizations), why it happened (attribution analysis across metrics), and what will happen (predictive modeling for proactive decision‑making).

Data synchronization components enable rapid ingestion from external sources (e.g., Douyin, Kuaishou, Baidu, Tencent) and internal systems, feeding a unified OLAP engine that abstracts technical complexity for analysts, marketers, and risk teams.

In advertising scenarios, the end‑to‑end data flow spans multiple touchpoints, and the real‑time capabilities of StarRocks allow sub‑10‑second query responses, dramatically improving operational agility.

Overall, ZhongAn’s data team demonstrates how a well‑designed, scalable data platform can accelerate digital transformation, enhance business insight, and support innovative insurance products.

big datareal-time analyticsStarRocksdata platformdigital transformationinsurance technology
DataFunSummit
Written by

DataFunSummit

Official account of the DataFun community, dedicated to sharing big data and AI industry summit news and speaker talks, with regular downloadable resource packs.

0 followers
Reader feedback

How this landed with the community

login Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.