He's famous! We expect paparazzi to be chasing Prodigal's Scott Hamilton down the street after his appearance on Troutman Pepper's Consumer Finance Podcast.
Scott's our Banking and Lending Strategy Leader, and he chatted with Stefanie Jackman and Chris Willis of Troutman Pepper's Consumer Financial Services Regulatory Practice.
Here's what they had to say:
Chris talked about old-fashioned call monitoring, which involved pulling a few calls randomly each month for each customer service rep and having a compliance manager listen to and score them. This method was obviously labor-intensive and only allowed for a limited number of calls to be monitored, which raised questions about the fairness and thoroughness of the process.
Then came basic voice analytics technologies, which work by transcribing voice conversations to text and allowing the client to define what to look for.
This was an improvement, but has left all kinds of problems, including the limitations of transcription accuracy, false positives and negatives, and and endless workload of training to come up with every possible keyword and phrase the software needed to look for.
"Many clients are really bright, but they can't come up with every way that customers can articulate a particular issue. So, you end up in this never-ending brainstorming cycle of different words or short phrases that identify the situation that you're trying to uncover," Scott explained.
The good news? Voice AI and ML can accomplish what those old-school technologies tried to do, but faster and with human-level accuracy. Even better? They learn as they go.
"The models are leveraging the learning that they've made because they understand lending or banking call types and conversations that they're then able to, with a lot more accuracy, identify very nuanced questions, complaints being a great one," Scott continued.
Pro tip: If you're looking for an AI tool to replace outdated voice analytics, make sure it uses machine learning and is trained on consumer finance instead of something that overgeneralizes.
Stefanie pointed out that regulators, particularly the Consumer Financial Protection Bureau (CFPB), have been increasingly expecting companies to have some form of voice analytics technology as part of their compliance management system.
"I do see the CFPB, at least historically, it has asked, why don't you use call analytics? Or now we have text analytics and all sorts of things, and I'm seeing it raises a little bit of a flag. I think there is an expectation from regulators that a compliance management system should have these types of tools involved in it," she said.
She also mentioned the challenges and questions that might arise from using machine learning and AI in the context of voice analytics.
"How are you teaching the machine? How do you unteach a machine? If it turns out that you've put in data or inputs that are resulting in the machine coming out with the wrong conclusion, you can't erase things from a person's head. How do you undo that in the AI context with some level of confidence?"
Pro tip: Ask how an AI solution you are considering satisfies existing and new regulatory requirements, and how it's trained.
Scott discussed the potential of real-time machine learning voice analytics in improving call center efficiency and effectiveness.
While older solutions delivered only post-call (typically next-day) compliance reporting - and sometimes not even that - because AI can process information in real time, it can offer additional information to support agents.
An AI solution, for instance, can provide real-time prompts to agents on what to say next during a call and or provide supervisors with real-time insights into ongoing calls.
"These same models can be run during the call as it's happening and can understand how the conversation is turning left, turning right, and can apply these learnings that the models have achieved to understand the context, can understand the intent of the evolving conversation, and as an example, can recommend to the agent in the call what they should say next," Scott explained.
He also mentioned the potential of these models in accurately identifying and logging complaints which is a challenge many financial services companies are facing. Unlike transcription-dependent solutions, AI models can understand the intent behind conversations, which helps in identifying nuanced questions and complaints more accurately and efficiently.
"The models are leveraging the learning that they've made because they understand lending or banking call types and conversations that they're then able to, with a lot more accuracy, identify very nuanced questions, complaints being a great one. You can load in, for example: "Identify calls with frustration or disagreement." Those aren't words that they're searching for. Those are intents that the models are searching for," Scott said.
Pro tip: When looking for a support solution, ask whether it is transcription- or keyword-dependent, or if it can leverage sentiment and intent.
What Scott's conversation with Chris and Stefanie shows is that there is an incredible amount of potential for AI and ML in contact center interactions and beyond.
Finding a solid AI solution can deliver far more than the compliance table stakes of voice analytics - it can provide agent coaching, opportunities for managers to support and evaluate their teams, capture complaints and therefore improve customer experience, and provide analytics to improve business.
.svg){kind=icon}
.jpg)
.png)
.png)
.webp)
-min.png)