What challenges did FamilyMart face before AI implementation?

Need for real-time, relevant product recommendations on ecommerce platform to boost sales. Low accuracy of in-app search limiting user experience and feature usage. Scaling data analytics infrastructure beyond on-premises to support growing data volumes and user base.

What was the impact of the AI solution?

Real-time personalized ecommerce recommendations increased sales. Search feature improvements led to 4X higher click-through rate and 2.5X more users leveraging in-app search. Operational efficiency improved through scalability and smoother feature rollouts. Planned further enhancements with voice search, Gemini-powered automatic tagging, and interactive chatbots.

Source

GCP

FamilyMart Enhances Customer Experience with Google Cloud AI and Data Analytics

Use case typeShopping recommendationsUpdated Jun 13, 2026

FamilyMart improved ecommerce product recommendations and in-app search accuracy for its 18 million members in Taiwan. The challenge was to deliver real-time, personalized product recommendations and improve search relevance in the convenience store mobile app. The solution leveraged Google Cloud BigQuery for fast analytics, Vertex AI Search for retail to personalize search results, and Google Kubernetes Engine (GKE) for scalable and smooth service deployment. BigQuery shortens data query times from minutes to seconds, enabling near real-time product recommendation adaptation based on user browsing and purchase behavior. Vertex AI Search improved contextual understanding and personalized search relevance, increasing in-app search click-through rates by 4 times and feature adoption by 2.5 times. GKE supports autoscaling and rapid app updates, improving operational efficiency and reliability.

Organization: FamilyMart
Industry: Retail
Location: Taiwan

Reported outcomes

quantified impactAdoption & scale

2.5xquantified impact

Strategic outcomes

Customer experience & trustImproved personalized product recommendations and searchCustomer experience & trustIncreased in-app search feature adoptionSpeed & agilityEnabled near real-time recommendation updatesScale & capacityImproved operational scalability and reliability

Catalog median for adoption & scale deployments: +85.5% across 46 reported metrics. Compare benchmarks →

Primary read

Use case focus

Showing 3 of 3

1Personalized Recommendations
2AI-Powered Search
3Data Analytics

Need for real-time, relevant product recommendations on ecommerce platform to boost sales.
Low accuracy of in-app search limiting user experience and feature usage.
Scaling data analytics infrastructure beyond on-premises to support growing data volumes and user base.

Migration to Google Cloud for high-performance data analytics and compliance with data security requirements.
Use of BigQuery for fast, parallel data queries enabling product recommendations in 2-5 seconds rather than minutes.
Integration of Vertex AI Search for retail powered by personalized data from Google Analytics and a dedicated BigQuery warehouse.
Deployment of app services on Google Kubernetes Engine for scalable, reliable operation and fast feature release.
Collaboration with Google Cloud Partner Dynacloud for AI model development and tuning.

Technologies

BigQuery Vertex AI Search for retail Google Kubernetes Engine

Real-time personalized ecommerce recommendations increased sales.
Search feature improvements led to 4X higher click-through rate and 2.5X more users leveraging in-app search.
Operational efficiency improved through scalability and smoother feature rollouts.
Planned further enhancements with voice search, Gemini-powered automatic tagging, and interactive chatbots.

Architecture

FamilyMart utilizes Google Cloud BigQuery for fast data analysis, Vertex AI Search for retail for personalized and contextual search relevancy, and Google Kubernetes Engine for scalable deployment of app services, supported by AI model development partnership with Dynacloud.

Implementation partners1

Dynacloud

Sources & evidence1

Groundedness: 4/5

Primary source

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

Explore related AI use cases

Customer evidence