AI-Driven Modeling to Improve the Agent and Customer Experience

Are traditional analytics and contact center practices enough to drive customer satisfaction? During this live Virtual Conference webcast, Larry Skowronek and Michelle Carlson from NICE Nexidia lead a conversation about how AI-driven data modeling can be the key to achieving greater success. To further explain, Larry and Michelle walk through the state of analytics today, an overview of sentiment analytics, an overview of predictive behavioral routing, and how to combine sentiment and predictive behavioral routing to maximize customer satisfaction and drive progress.

Today, we generally see a large disconnect between business and how they evaluate customer interactions. Eighty percent of companies claim they deliver “superior” customer service, while in reality, only eight percent of their customers actually agree. This is partially because the state of measuring customer satisfaction is deeply flawed. Manual reviews of calls that require a human to evaluate transactions lead to highly subjective, interpretive, and inconsistent feedback, which not only requires higher costs, but also fails to move the needle forward.

Customer contact centers are a dynamic and evolving animal. The only way to respond to change is with change. Enter: Sentiment Analytics. Sentiment Analytics is a way to use machine learning to train a model that measures whether our customer interaction was positive, negative, or neutral on a granular scale. The machine can take our otherwise subjective behaviors and turn them into subjective data that is highly valuable and actionable. This data is consistent, accurate, and without bias. Most importantly, because it is a machine, it can do as much work as we throw at it, so we can receive and analyze data for every single customer interaction.

This AI-based model has proven to be statistically accurate, according to several CX centers that use it. But how exactly does this model measure customer satisfaction. The model reliably measures every interaction, including:

  • Spoken words, like “Awesome”, “I’m annoyed”, and “This is ridiculous”.
  • Laughter detection.
  • Pitch and tone.
  • Cross talk: customer and agent interrupting one another.

These models may also differentiate calls that start positively and end negatively, indicating worst practices, as well as calls that start negatively and end positively, indicating best practices. The reliability and accuracy of these models have allowed businesses to gain deep insights on the overall customer experience and quickly translate those insights into action. Finally, these models create a hyper-personalized customer experience. This is a monumental advantage, as eighty-four percent of customers say that personalized customer experiences are key to winning their business.

For a perfectly personalized customer experience, sentiment models can aid in Predictive Behavioral Routing (PBR), which uses sentiment analytics to match the customer to the appropriate agent and therefore improves the overall customer experience. By bringing Sentiment Analytics and PBR together, businesses can seamlessly operationale their sentiment data by:

  • Calculating customer sentiment on 100% of interactions
  • Using this sentiment combined with personality, make the best connection for the customer.
  • Immediately improve customer experience with AI-powered routing.

So, what does this process look like in real time? In one example, a Fortune 500 company’s customers were initially being transferred all over the contact center. They then optimized their customer calls based on sentiment dada. Here’s what happened:

  • They saw a 15% decrease in negative sentiment on PBR (predictive behavioral routing) routed calls.
  • They saw a 13% increase in positive sentiment on PBR routed calls
  • They saw a 6.4% decrease in average handle time in PBR routed calls
  • This required 0 hours of coaching, training and employee change management.

The combination of sentiment and behavioral routing will improve customer satisfaction metrics, reduce costs for manual listening and surveys, improve customer satisfaction via targeted coaching and performance management, and increase employee satisfaction. Your analytics practices are valuable, but should be evolving to keep up with dynamic consumer expectations. Your employees and customers alike will thank you for it.

To listen to the full webcast click here: https://bit.ly/2ULJgPB

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s