Transforming Contact Centers with a Unified Knowledge Base and Generative AI

We unified knowledge fragmented across channels and integrated it with generative AI. An increase in self-resolution rates and reduction in human operator response ratio were achieved, enabling operators to shift to value-added tasks.

Business Process Reform (BPR)
Retail & Distribution

Transforming Contact Centers with a Unified Knowledge Base and Generative AI

ISSUE
- Rising inquiry volumes and fragmented knowledge across channels had begun to limit further improvements in operational efficiency and self-service resolution rates.
- While the company explored generative AI as a potential solution, balancing response accuracy with cost efficiency remained a key challenge.
SOLUTION
- Built a unified knowledge base common to all channels and centrally linked it with self-resolution tools such as Visual IVR
- Optimized answer accuracy and cost efficiency with a hybrid configuration in which routine Q&As are supplemented with knowledge and non-routine Q&As are supplemented with generative AI.
RESULT
- Reduced the telephone inquiry ratio and greatly improved the self-resolution rate
- Reallocated operator resources to value-added tasks, such as online customer service

ISSUE

Growing Inquiries and Knowledge Fragmentation

A major retail and distribution company operated a large-scale contact center, but fragmented knowledge across channels was limiting growth in self-service resolution rates. In addition, the company had concerns about the risks associated with adopting generative AI, including inaccurate responses (hallucinations) and rising costs.
Beyond improving operational efficiency, the company needed to enhance the customer experience at the same time.
SOLUTION

Hybrid Implementation of a Unified Knowledge Base and Generative AI

We supported business transformation by integrating knowledge management with generative AI. First, we built a unified, cross-channel knowledge base, enabling the system to instantly deliver optimal responses. In addition, we introduced a Q&A ecosystem designed to resolve repeat inquiries through automated channels, along with visual IVR, significantly improving self-service resolution rates.
We also designed a solution that leverages generative AI to handle non-routine inquiries, such as questions related to product specifications. By combining structured and unstructured data, the system delivers AI-generated responses that balance both accuracy and flexibility.
RESULT

Reduction of Telephone Inquiry Ratio and Shift to Value-Added Tasks

As a result of the contact center transformation, the share of phone-based inquiries decreased by approximately 10 percentage points, significantly reducing the workload on operators. Resources freed up through these efficiency gains were reallocated to revenue-generating areas such as online customer engagement, transforming the contact center into a function that actively enhances customer satisfaction.
In addition, generative AI enabled the center to handle long-tail inquiries related to a vast range of products, allowing customers to resolve issues on their own even outside business hours. This led to a substantial improvement in overall customer convenience.