What challenges did Shop Global face before AI implementation?

Replace keyword search with conversational, multimodal product discovery that understands user intent and personalizes results. Integrate the shopping experience into LINE while supporting fast responses and high traffic. Deliver accurate product discovery across a large catalog and diverse customer contexts.

What was the impact of the AI solution?

Provides instant, smarter support for product queries and style advice. Delivers hyper-personalized discovery across more than 1,000 brands. Supports conversational searches with accurate results in 1 to 2 minutes. Scaled to support about 150,000 visitors during the Saha Group Fair '25 event.

Source

GCP

Shop Global: agentic LLM search and AI Personal Shopper built with Vertex AI and Gemini

Use case typeProduct discoveryUpdated Jun 13, 2026

Shop Global, part of Thailand's Saha Group, implemented an AI-powered agentic and LLM search solution to replace keyword search with conversational and multimodal product discovery. The solution supports text and photo search, extracts product attributes, and uses user profile, session, trend, and weather context to generate hyper-personalized recommendations through a chatbot and dynamic results page. It is integrated into the LINE shopping experience as an AI Personal Shopper for personalized deals and in-app shopping, and supports Thai and English search at high event traffic.

Organization: Shop Global
Industry: Retail
Location: Thailand
Published: May 2026

Reported outcomes

2 minutes

timeTime & speed

Strategic outcomes

New product / capabilityIntroduced conversational multimodal product discoveryCustomer experience & trustCreated hyper-personalized shopping recommendationsCustomer experience & trustDelivered instant product and style supportScale & capacitySupported high-traffic event visitors

Primary read

Use case focus

Showing 3 of 4

1Conversational Search
2Product Discovery
3Personalization

Replace keyword search with conversational, multimodal product discovery that understands user intent and personalizes results.
Integrate the shopping experience into LINE while supporting fast responses and high traffic.
Deliver accurate product discovery across a large catalog and diverse customer contexts.

Shop Global partnered with Tridorian to implement an AI-powered agentic and LLM search solution on Google Cloud.
The workflow uses Vertex AI Search for vector search, Vertex AI Conversation for agentic interactions, and Gemini multimodal AI to interpret text and image inputs and infer product attributes.
Datastream, Dataflow, Cloud Composer, BigQuery, Looker, and Cloud Operations support data transfer, analytics, orchestration, and monitoring.
The AI Personal Shopper uses contextual signals such as profile data, session activity, trends, and local weather to tailor recommendations and promotions.

Technologies

Vertex AI Search Vertex AI Conversation Gemini BigQuery Looker Datastream Dataflow Cloud Composer Cloud Operations

Provides instant, smarter support for product queries and style advice.
Delivers hyper-personalized discovery across more than 1,000 brands.
Supports conversational searches with accurate results in 1 to 2 minutes.
Scaled to support about 150,000 visitors during the Saha Group Fair '25 event.

Implementation partners1

Tridorian

Sources & evidence1

Groundedness: 5/5Type: Customer StoryPublished: May 26, 2026Publisher: Google CloudEvidence: PrimaryConfidence: High

Primary source

AI-generated summary. Verify important details with the linked sources before relying on this case.

Shop Global: agentic LLM search and AI Personal Shopper built with Vertex AI and Gemini

Use case focus

Challenges

Solution

Impact

Comments

Similar cases