No Jitter is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Boosting Contact Centers with Conversational AI (Part 1)

AdobeStock_334669730.jpeg

Creating an AI bot
Image: sdecoret - stock.adobe.com
If you aren't currently exploring conversational AI (CAI) for your contact center operations, there is a good chance that you will be soon. With the pressure many agents are facing due to coronavirus-related call volume increases, CAI can provide agents with some much needed relief and boost in call center productivity.
 
CAI allows contact center operations to automate messages and speech-enabled applications for interactions between humans and computers. CAI bots can communicate similarly to humans, recognizing speech and/or text. Additionally, these bots can go so far as deciphering customer intent in different languages and respond accordingly.
 
To learn more about CAI, I contacted Amy Allen, product manager for CSG’s conversational AI solution. Below is the first part of our conversation.
 
What is conversational artificial intelligence?
CAI is a new and innovative way to communicate, leveraging AI, and delivers a next-gen platform. CAI isn’t just about voice; it’s a generational shift beyond chatbots to meet customers where they want to interact, regardless of communication platforms (SMS, voice, text, chat, IVR, smart home devices), and in a conversational, smart, and personalized way.
 
The hallmark of an AI system is that it’s always learning. A CAI platform has accelerators that speed up AI deployment and training. This solution uses a single AI “brain,” allowing enterprises to reuse interface logic and integrations across several channels or multiple Interactive Voice Responses (IVRs).
 
How does it apply to contact centers and unified communications?
CAI improves customer satisfaction (CSAT), call containment, first call resolution, and cost control and shortens customer service representative (CSR) training cycles. Contact centers can deploy an AI-powered virtual assistant in their IVR, chat, social, or text interfaces to resolve a large chunk of inbound inquiries. By combining multi-intent understanding with contextual awareness, CAI has a better grasp of the customer’s intent than standard voice-activated call trees. It can communicate internally and externally through applications and web sites using natural language (voice, text, or gesture inputs).
 
Once the virtual assistant is fully integrated and trained, it becomes a true asset for the agile contact center. It adapts to changes in the business’s offerings, processes APIs, and data with minimal manual adjustments and improves the customer’s ability to self-serve through the platform over time.
 
CAI can also be a valuable training tool. The AI can assist a CSR through the desktop by listening in on the call, populating the CSR’s screen in real-time with pertinent knowledge based information. It can also give the CSR on-screen prompts of what to say or what required disclaimers to give the customer. The result is that contact centers can shift time away from training CSRs and toward training the AI brain that delivers consistent information — interaction after interaction.
 
Is this an out of the box solution or do you have to build it? If it’s the latter, what is the process?
CSG works with the enterprise customer to develop and build the CAI platform to support the customer’s desired use cases and channels. This is supported by pre-built accelerators that enable rapid deployment into multiple verticals for common use cases and pre-built frameworks for easy integration into back-office and third-party applications.
 
The process for building a custom CAI solution includes:
  • Discovery: CSG helps the enterprise take massive amounts of raw, unstructured data from multiple sources and classify customer intents and key issues. This conversational data mining and analysis suite unlocks knowledge held in immense volumes of natural language conversations, delivering previously unprecedented levels of big data insight and true “voice of the customer” understanding. Recent projects have sourced data from call transcripts and ticketing systems as examples of inputs.
  • Studio: This tool allows non-experts to easily construct dialogues and business logic using visual flow chart structure. The graphical interface makes it easy to understand what is and isn’t working with the dialogue flow, with the ability to make rapid adjustments as necessary. Adding new dialogues or updating responses is as simple as drag and drop, while one-click publishing ensures any changes are live instantly. The studio tool enables customers to do as little or as much of the dialogue management as they desire.
  • Deployment: Tailored implementations can be completed in as little as six to eight weeks, depending upon scope and specific business goals. Next, the CAI platform is fine-tuned and optimized after delivery to ensure the best possible customer experience.
  • Management: Enterprises can deploy CAI as a fully managed service running in the cloud, including help with expanding the suite of use cases. Alternately, customers can self-manage all or parts of the system.
 
For more of this conversation, make sure to check out part two next week, where we will explore the business consideration for deploying CAI.