No Stone Unturned: Voice Interaction Analytics in the Contact Center

Calls between agents and clients can be a treasure trove of information. However, it takes some effort to extract the real value from interactions. What supervisors really want is to glean actionable insights — something a supervisor can use to improve the performance of the agent in reaching a goal — from the interaction.

In going through the rote and manual process of reviewing phone calls, a supervisor might not be able to derive a significant amount of insight; listening for particular clues that are used to provide feedback to an agent will take up most of their time. Many details can be missed during this manual process and, especially for longer calls, it can be difficult to stitch together how different parts of a discussion affect an outcome.

Voice interaction analysis usually starts by creating a transcript of the conversation. Once this is completed, it’s possible to analyze for topics that were discussed and for the sentiment of the customer and agent during the call. With just the simple addition of search capability, the supervisor can quickly search a call for critical information. This means no longer having to wait and listen for something to be said or having to skim through a long transcript to find the relevant part of the discussion that could affect performance.

Voice Interaction Analytics for Supervisors
It’s important for supervisors to have the most impact with the time they spend either reviewing calls, looking over metrics of interactions, or from the one-on-one time that’s spent with agents. Supervisors must balance the act of providing critical feedback to improve agent performance and praise on the things that agents are doing right. To do this, they need to understand which actions agents perform that lead to the best outcomes — and which behaviors can be avoided. For these supervisors, it’s important to know if calls are going well, if agents are saying what they’re supposed to say in a way that they’re supposed to say it, and what particular tips can have the biggest impact on agent performance.

Systems that perform topic analysis look at which words are spoken in a transcript and group these into topics to serve as a summary of a conversation. By analyzing for topics, supervisors can see whether particular items were mentioned in a call or how an agent handled an objection — without having to read through an entire transcript. Supervisors can see which topics were spoken, which topics are more likely to be grouped together in a call, and in which order the topics were spoken.

By diving into specific areas for topics, supervisors can quickly evaluate calls that are relevant to where they can provide coaching. This way, they don’t waste efforts listening for long stretches at a time to portions of the call where they can’t help. They also can see previous feedback affects the order of topics an agents covers — and how that feedback affected particular agents.

Voice Interaction Analytics for Agents
Honing assessments creates autonomy in how agents work. Assessments become more directed toward where agents can actually improve, rather than just a binary assessment of complying to a script. When agents can access the same information about topics in interactions they’ve had with customers, they can glean insights into how their responses lead to certain outcomes. This gives them further control over their work and where to make improvements.

When agents are assessed against criteria based on topics analysis, they gain a better understanding of what’s expected of them. This can lead to higher satisfaction and lower turnover rates. Because these tools can provide more autonomy and trust between supervisors and agents, they also lead to a much happier workplace. And this usually translates into a better customer experience.

Check out this infographic for more information on how to provide great customer experiences. And read more to find out why Gartner recently named Genesys as a Visionary in Workforce Engagement Management.