We use cookies for our website. By continuing to browse the site, you agree to our use of cookies.

Home Clients AI Call Analytics with Gemini for a Logistics Company Call Center

AI Call Analytics with Gemini for a Logistics Company Call Center

AI Call Analytics with Gemini for a Logistics Company Call Center

REQUEST:

The logistics company team was looking for optimal AI tools to analyze phone calls and turned to Wise IT to compare the reliability and cost-effectiveness of different solutions.

RESULT:

A successful pilot project confirmed the advantages of using the Gemini AI model for the tasks at hand. Efficiency, moderate cost, and ease of configuration became the key criteria for selection.

Business Challenge

The logistics company team aimed to implement artificial intelligence to automate the analysis of call center recordings. The key AI tasks were tone analysis, classification into one of over 300 topics, determining the call’s purpose, and summarizing it. Such analysis was needed to improve customer service quality, optimize operator performance, and identify systemic trends in inquiries.

Initially, the client considered building a custom AI model for these tasks based on the Vertex AI Platform. However, recognizing the need for an objective comparison of approaches in terms of efficiency, cost, and deployment speed, the company sought expert advice from the Wise IT team.

Solution

The Wise IT team, with experience in cloud technologies and AI, suggested exploring ready-made solutions on the Google Cloud platform. In particular, they recommended testing Gemini 1.5 — Google’s newest AI model at the time.

To objectively assess the benefits, Wise IT and the client ran a pilot project. The pilot compared two approaches: potentially developing and configuring a custom model versus using the Gemini 1.5 API.

Key Results

The pilot project clearly demonstrated the benefits of using Gemini 1.5 for call analysis tasks. The model showed high efficiency in detecting tone, topics, and other call parameters.

The key factors that led to the choice of Gemini were:

  • Efficiency: The model successfully handled analysis across all required criteria.
  • Cost-effectiveness: Testing showed that the cost of processing a single call via the Gemini API was less than 1 cent.
  • Ease of implementation and management: Using a ready-made API greatly simplifies integration and ongoing system administration.

Client Feedback

Thank you to Wise IT for helping us choose the right solution for our call center analytics. The billing for Gemini AI turned out slightly higher than expected, but it’s clear that estimating the operational cost of such a system is quite difficult. We are satisfied with the tool’s performance and cost-efficiency and are currently working on further optimization.

— CTO of the client company