Customer Support is a crucial part of every organization. With the proliferation of social media, customer centric organizations are using social media platforms likeX(formerly twitter)to provide customer support. Irrespective of the platform, Support Automation has been actively pursued by organizations to improve customer experience and loyalty while reducing the workload on support teams.
While automated prioritization and routing have been solved using predictive models, Automated resolution is still an active area of research. Majority of support requests are primarily requests for information or Level 1requests. These requests aren't the best use of Customer Support personnel's time and need to be automated.
About this Accelerator
In this accelerator, we will illustrate how to use Generative AI models to cater to Level 1 requests, allowing support teams to focus on more pressing and high visibility requests. Learning from historical communications, Generative AI Agents can maintain the same standard of support communication that the customers are used to.
Usage of Generative AI comes at a cost which includes cost per request, compute, cost per token which can balloon up in no time. To optimize this, we will show how to use Predictive ML models for routing standard requests to Generative AI and high priority severity requests to Human staff.
Hence, using Predictive ML and Generative AI models we illustrate the idea of optimizing customer support workflows.