Coforge

Who We Are

About Us Newsroom Leadership Partners Locations Careers Awards & Recognitions ESG Learn more about Coforge
Coforge: Where AI engineering meets industry expertise.

Learn about our company, our vision and values, and the 45,000+ professionals enabling businesses to harness the power of AI.

Learn more about Coforge
Composable Enterprise

CASE STUDY

Achieved 90%+ Query Resolution Accuracy for Travel Subscribers Across 11 Languages with a GenAI Support Bot

Industry

Travel

Our Contributions

Generative AI Development, AI-driven Enterprise Search, Chatbot Integration, GCP Data Pipeline Automation

Location

USA

Achieved 90%+ Query Resolution Accuracy for Travel Subscribers Across 11 Languages with a GenAI Support Bot
To improve customer satisfaction for its global user base, a leading US travel technology firm needed to streamline information retrieval, resolve queries faster, and reduce dependency on support teams. They turned to Coforge to transform its subscriber support experience using Generative AI to enhance search capabilities and embed an AI chatbot into its existing support channels.

Transformation timeline

Phase 1
Discovery and approach evaluation.
Phase 2
Automated data pipeline to bring Salesforce knowledge documents into GCP.
Phase 3
Gen AI bot deployment across homepage, finder search, and case creation pages.
Phase 4
API integration, multilingual enablement, and roadmap for customization.

Drag

Providing_Intelligent__Multilingual_Subscriber-Mid-page-image

The Challenge

The client’s subscribers were struggling to quickly find answers to their questions. Traditional search surfaced a high volume of results and forced users to wade through lengthy knowledge articles, slowing down resolution and driving up support case volumes.

The existing chat channel lacked intelligent automation, which meant customer queries frequently escalated to support teams, increasing operational load and impacting customer satisfaction.

The client needed an AI-driven search and chatbot capability that could reduce the volume of search results, minimize the need for extensive review of knowledge articles, efficiently resolve customer queries, and decrease dependency on support teams, all at enterprise scale and across multiple languages.

Our Approach

Dual-Approach Evaluation

We proposed two implementation options based on our analysis: a no-code GCP solution, and a customizable platform with a bespoke UI with a pipeline built using Python SDK on Google Matching Engine.

No-Code GCP Solution

The client selected a no-code approach on Google Cloud Platform, enabling rapid configuration of embeddings, vector store, and LLMs through the GCP GUI for faster time-to-value.

Automated Salesforce-to-GCP Document Pipeline

Built and automated a pipeline with scalable REST API integration to continuously bring Salesforce knowledge documents into Google Cloud Platform, ensuring the Gen AI engine always draws from fresh, enriched enterprise content.

Gen AI Bot Integration

Implemented a Generative AI bot on the homepage, finder search, and case creation page to efficiently resolve customer queries within the existing chat channel and reduce dependency on support teams.

Roadmap for Customization

Laid the foundation for subsequent phases where the fully customizable platform, built on Python SDK with Google Matching Engine, will be leveraged to deliver richer, tailored subscriber experiences.

Providing_Intelligent__Multilingual_Subscriber-3rd_image

Impact to Date

90+%

Accuracy achieved in resolving subscriber queries

11

Languages supported via automated translation of knowledge content
Provided intelligent, frictionless access to a 13,000+ document knowledge base

Key Outcomes

 

  • Streamlined information retrieval through AI-driven search, reducing the volume of results and the need to review lengthy knowledge articles.

  • Reduced dependency on support teams by resolving customer queries directly within the existing chat channel.

  • Improved customer satisfaction with fast, accurate multilingual support.

  • Accelerated time-to-value by deploying the no-code GCP solution in Phase 1 while keeping a clear path to a fully customizable platform.

  • Delivered a future-ready architecture with an automated knowledge pipeline and scalable REST API integration.