Case Study: Tokopedia achieves 10x smarter search with Milvus by Zilliz

How Tokopedia Achieved a 10x Smarter Search Experience Using Milvus

Tokopedia, Indonesia's largest e-commerce platform, faced challenges with its keyword-based Elasticsearch system. The search results lacked semantic understanding, as the engine could not interpret the meaning behind user queries, relying instead on statistical word matching. To create a more intelligent search experience, Tokopedia turned to the vendor Zilliz and its vector database, Milvus.

By implementing Zilliz's Milvus as its vector similarity search engine, Tokopedia gained a powerful and user-friendly solution for semantic product matching. The system was configured for high availability to ensure reliability. This resulted in a 10x smarter search experience for users, which in turn led to a 10x higher click-through rate (CTR) and conversion rate (CVR) for its advertising service, significantly enhancing the overall user experience on the platform.

Tokopedia

Rahul Yadav

Software Engineer