Transforming customer service with agentic AI and Amazon Nova at Ryanair

Ryanair operates at massive scale, supporting over 700,000 customers per day. As passenger volumes grew, customer service inquiries increased proportionally. The traditional approach of scaling human agents to match demand created an unsustainable cost structure for a low-cost carrier where operational efficiency drives competitive advantage.

The airline faced a fundamental business challenge: customer service volume was growing in line with passenger numbers, but scaling agents proportionally made no sense from a cost or efficiency perspective.  Adding more agents to the mix would not have solved the problem and would undermine the cost discipline central to Ryanair's business model. Handling routine inquiries through traditional channels as the airline continued expanding was not a sustainable solution.
 

Ryanair had already achieved significant progress in voice automation, using Amazon Connect to redirect 80%  of inbound calls to digital channels and saving over 500,000 agent hours. However, the chat channel remained largely manual, relying on a legacy chatbot that provided static responses. Customers frequently abandoned these interactions in frustration, creating downstream volume in other channels.

The airline identified three core opportunities; First, reduce cost per contact in the chat channel by moving routine inquiries away from human agents without sacrificing customer satisfaction. Second, extend AI-driven automation from voice into chat to create a unified omnichannel experience. Third, deploy multilingual support across all markets through a single scalable architecture rather than maintaining separate systems for each language.

"We needed to move from linear cost scaling to a model where customer service costs remain largely fixed as passenger volumes grow," said Tracy Kennedy, Director of Customer Services. "The question was whether AI could handle the complexity and volume of real customer inquiries across multiple languages while maintaining the service quality our passengers expect.”

Working with AWS Partner Cation Consulting, Ryanair built a customer service AI assistant on Amazon Bedrock AgentCore, a platform for building and deploying AI agents for complex workflows, powered by Amazon Nova 2 Lite, a fast and cost-effective reasoning model. The solution handles all query categories including baggage, refunds, bookings, and flight disruptions within a unified architecture.

The airline conducted a rigorous model evaluation, testing five foundation models against 12,000 real production customer questions. Amazon Nova 2 Lite delivered an 84% latency improvement, reducing response time from 18 seconds to 2.9 seconds, while achieving a 25% accuracy uplift. These results were validated in production within weeks of deployment.

Amazon Bedrock AgentCore replaced an earlier architecture of 12 domain-specific agents with a single consolidated agent in February 2026. This consolidation reduced latency, simplified orchestration, and provided a more scalable foundation for ongoing development. The agent evaluates every answer through a multi-dimensional, multi-model guardrail scoring system before it reaches the customer, ensuring response quality and accuracy.

For multilingual support, Ryanair integrated Amazon Translate, a neural machine translation service, alongside Amazon Nova 2 Lite to handle six non-English languages across all European markets. This approach eliminated the need to maintain separate systems for each language while ensuring consistent service quality.

The solution integrates with Amazon Connect, which Ryanair uses for voice customer service, creating a unified omnichannel platform. This integration allows the airline to route customers between voice and chat channels based on inquiry complexity, while maintaining context across interactions.

"Amazon Bedrock AgentCore with Amazon Nova 2 Lite gave us the performance, cost efficiency, and scalability we needed to transform customer service at our scale," said Alan Kiernan, CTO Cation Consulting. "The single-agent architecture simplified our operations while the guardrail system ensured we maintained service quality across millions of interactions.”

Since launching in October 2024, Ryanair's AI-powered customer service solution has generated 10 million chatbot answers, achieving a 94% accuracy rate. The assistant now handles 120,000 daily customer answers and delivers a reduction of Customer Services contacts per passengers carried by 70%. 

Response latency improvements directly enhanced customer experience. The 84% reduction in response time, from 18 seconds to 2.9 seconds, meaningfully improved the quality of digital interactions. Customers receive immediate, accurate answers across seven languages without waiting for human agent availability.

Combined with the Amazon Connect voice transformation the platform has contributed to a saving of over 500,000 agent hours to date, the omnichannel platform delivers actual savings. This allows Ryanair to scale customer service capacity without proportional headcount growth, fundamentally changing the economics of customer service operations.

The unified architecture also creates a foundation for post-booking ancillary upselling, a net-new revenue stream currently underway. By combining voice intelligence and chat commerce within a single platform, Ryanair can convert service interactions into commercial outcomes in ways that competitor customer service platforms cannot match.

"The AWS-powered solution transformed our customer service model from a cost centre that scaled linearly with passenger growth into a strategic asset that improves both efficiency and customer experience," said Stephen Keane, Head of Business Transformation, Ryanair. "We're now positioned to handle continued growth while exploring new revenue opportunities through the same platform."

Looking ahead, Ryanair continues to refine the AI assistant's capabilities and expand its role in the customer journey, building on the scalable foundation provided by Amazon Bedrock AgentCore and Amazon Nova 2 Lite.