Case Study: POSCO achieves 53% faster, user-centric enterprise search with Elastic

Moving to Elasticsearch for a More User-centric Enterprise Search Experience

POSCO’s Enterprise Portal serves about 40,000 employees and must index and serve more than 85 million documents with roughly 2 million updates and deletions, but its legacy Google Search Appliance–based engine struggled to meet diverse user needs and handle large-scale, user-centric search requirements. The EP Operations team needed a more flexible, scalable solution to improve search speed, relevance, and operational stability.

POSCO replaced the legacy engine with the Elastic Stack—using Elasticsearch as the main search engine plus Kibana, Logstash, Beats, and Elastic security—which delivered near-real-time indexing, improved Korean search relevance, and 53% faster performance while stabilizing large data volumes. The team now monitors system health, secures indexed data, and plans to add machine-learning–based failure detection and personalized search features.

POSCO

Eui-young Hwang

EP Operations team, IT Business Department