Optimizing Customer Service with Intelligent Automation

In today’s world, customers expect an ever-greater depth of experience and value from the services they seek. However, overcoming the obstacles to delivering a superior customer experience has never been a greater task than now.

To do so means investing in a digital future. Automation must be deployed for organizational efficiency and technology. Systems also need to be upgraded to serve customers in better ways. With initiatives such as updating policy systems and integrating with partners, automating processes will be key to unlocking the next wave of productivity.

Our client is a prominent general insurance firm. They faced challenges impacting customer satisfaction and agent productivity in their contact center. This center handles nearly 35 million conversations annually, grappling with inefficiencies that bogged its operations.

The goal was to reduce wasted agent hours and customer wait times significantly. This necessitated a streamlined approach that could automatically discern caller intent. This way, more straightforward queries could be resolved more quickly. This would free human agents to focus on more intricate issues.

Breaking through contact center operational inefficiencies

The contact center was overwhelmed by the volume of inquiries. It was handling over 20,000 hours of agent time daily. This inefficiency also meant many queries were misdirected to incorrect departments, causing avoidable delays and extended wait times for customers seeking support.

Streamlining security and authentication

A cumbersome security authentication process further impeded the flow. Substantial time was needed to verify customer identities. This added to the overall wait time for customers. For the agents, it detracted from the time they could spend addressing more complex customer needs.

Our client needed a solution to create better digital value for its 20m customers and make the most of its 19bn data points.

Our design and technology expertise made us an ideal partner for dissecting and classifying the underlying intents behind customer inquiries. Our solutions approach progressed to the meticulous training and evaluation of 16 identified categories to model for automated responses. We tapped into a comprehensive dataset of ~12,200 recorded dialogues. We distilled them into 9 distinct categories to inform the creation of automated response models. The model was designed using machine learning and NLP neural network-based techniques before undergoing several iterations to achieve greater accuracy.

What we provided:

AI for efficient customer service automation

Leveraging GCP, we deployed a voice bot that could identify and handle more straightforward queries. We created a conversational design flow for three intent categories: policy updates, cancellations, and document requests. Together, they drove 35% of the total call volume. Our solution created a more straightforward, streamlined process for handling customer calls, freeing up agent time.

The immediate impact

We found that nearly 78% of customer conversations could be grouped within four simple intent categories. This meant that the automated voice bot could effectively manage most requests. Rolling out the solution saved an estimated 40% time for the client’s customer care agents.

The long-term gains