We empowered one of the largest Energy provider in Europe by optimizing phone and email customer service processes with tailor-made AI-enabled tools. Explore how machine learning (ML) models can increase customer satisfaction and decrease average handling time.
Key challenges & Context: Don’t Compromise Quality for Efficiency
Our client is one of the largest energy suppliers in Germany and Europe. They provide more than 5.5 million people with gas, power, water and energy-related services and products.
Operating in a highly homogenous market by nature, our client identified customer service quality as a key differentiator. In consequence, they started to and continue to increase customer support efficiency and quality by harnessing Natural Language Processing (NLP) technology as a competitive advantage.
Our client was aiming to:
- Decrease average handling time of customer requests;
- Increase quality of customer interactions;
- Improving the understanding of customer needs and requests.
Our Approach: Tackling Customer Support Processes to Build AI-Enabled Tools
We helped our client automate a part of their customer service processes and build AI-enabled tools that support customer agents and customer analytics to achieve their goals in two different ways:
1. Email-based processes automation:
- We first get an in-depth understanding of the business processes, identified relevant use cases, validate an valuing their potential.
- We then, took care of building and launching cloud-based Minimum Viable Products (MVPs) in the shortest time feasible to support agent.
- Finally, we continuously iterated to improve performance against clearly defined, measurable KPIs.
2. Phone-based processes automation:
- To better understands customers request, we helped our client to build, improve, operate and maintain an intelligent system. A ML-based tool that asks customers the reason for their call, classifies it and tracks the customer’s answers.
- The system responds by offering several self-services options that either shorten the following interaction with a support agent or render the entire interaction obsolete.
Benefits: 1.8M calls minutes representing €900,000 saved
- Automation of six phone-based processes within 15 months
- Automation of three email-based processes in 8 months
- Delivered two ready-to-use AI tools improving customer support agents performance in eleven months
Our client now automatically classifies 2M customer calls per year according to 50 different reasons. This saves nearly 1.8 million minutes annually, representing approximately €900,000 in cost savings.
Team involved
The project’s success is largely due to the dedicated efforts of our diverse and specialized team. Over 22 months of intensive collaboration, our multidisciplinary team worked closely with the client to deliver this comprehensive solution.
The Collaborative Team:
- ML Engineers: Our machine learning specialists brought deep expertise in model development, training, and optimization. They were responsible for designing and implementing the core AI algorithms, ensuring robust performance and scalability of the machine learning components.
- Data Scientists: The data science team played a crucial role in data analysis, feature engineering, and model validation. Their expertise in statistical analysis and data mining techniques was instrumental in extracting meaningful insights and ensuring data-driven decision making throughout the project lifecycle.
- NLP Experts: Our Natural Language Processing specialists, with their profound understanding of language models and conversational AI, were key in developing sophisticated text processing capabilities. They focused on optimizing language understanding, response generation, and ensuring seamless human-computer interaction.
- Cloud Architects: The cloud infrastructure team designed and implemented scalable, secure, and cost-effective cloud solutions. Their expertise ensured optimal performance, reliability, and seamless integration with existing systems while maintaining high availability standards.
- DevOps Engineers: Our DevOps specialists orchestrated the continuous integration and deployment processes, automating workflows and ensuring smooth operations. They were instrumental in establishing robust monitoring, logging, and maintenance procedures for long-term system reliability.
- Project Managers: The project management team coordinated all aspects of the development process, ensuring timely delivery, effective communication between stakeholders, and adherence to project requirements. They facilitated knowledge transfer and maintained project momentum throughout the 22-month engagement.
Technologies Used
The successful implementation of this project relied on a carefully curated technology stack and strategic partnerships. Each component was selected for its specific capabilities and seamless integration potential:
- Microsoft Azure served as our primary cloud platform, providing scalable computing resources, advanced AI services, and robust security features. Azure’s comprehensive ecosystem enabled efficient deployment and management of our machine learning models and data processing pipelines.
- Python was our core programming language, chosen for its extensive libraries, machine learning frameworks, and rapid development capabilities. Its versatility and strong community support made it ideal for building complex AI applications and data processing workflows.
- LUIS (Language Understanding Intelligent Service) powered our natural language understanding capabilities, enabling sophisticated intent recognition and entity extraction. This Microsoft cognitive service was crucial for creating intuitive conversational interfaces and improving user experience.
- Dialogflow complemented our conversational AI architecture, providing additional natural language processing capabilities and conversation management features. Its integration allowed for more flexible and contextually aware dialogue systems.
Contact our experts to discover how we can help you as an Energy Provider or a organization looking for an AI & Data Platforms expertise for your customer service.
