This content is part of the Essential Guide: How unified communications is getting down to business

AI in contact centers offers agents behavioral analytics

Contact center AI offers tools to help agents provide better customer service through behavioral analytics. But AI may also present some privacy concerns.

These are unusually disruptive times in the contact center space, and it's hard to keep up with the latest technologies and trends. The implications of this rapid change are not fully understood, however, especially around AI and its growing role in the contact center. While the benefits of AI in contact centers will be quickly welcomed, we may be giving up a few things along the way. The specter of Big Brother is everywhere with AI, and you should expect to see a growing push for ethical best practices as those roots go deeper.

One area at issue is behavioral analytics, a rather clinical-sounding term that can be applied in several ways. The most common use case is to map out online traffic patterns and buying behaviors, which generally falls under the benign umbrella of customer journey.

An implicit social contract exists with e-commerce, where it's understood that every move we make online is tracked, and chances are good that our digital trail is being closely followed in real time. We're no longer surprised when the "Can I help you?" chat window pops up within seconds of browsing whatever is on offer. Shopping online may be fast and easy, but we pay a price -- namely forfeiting most of our privacy.

Your energy is flagging -- get a coffee

Take things a step further, where AI-driven behavioral analytics can detect the psychological state of your agents and the anxiety level of your customers in real time. That's how good AI in contact centers is getting, especially with speech recognition and applying it to what's called sentiment analysis. Just as humans pick up signs from body language, tone of voice, speech patterns or pace of talking, AI is now able to identify those social cues. The capabilities aren't perfect, but for some, they're good enough to be used in contact centers with live customers and agents.

The appeal is understandable for contact centers embracing AI to find new ways to get measurable improvement in agent performance, with the belief it will translate into a better customer experience. This is the brass ring most businesses are chasing -- not just for a genuine desire to delight customers, but also to adopt new metrics for agent performance and remuneration.

Imagine a scenario where an agent is nearing the end of his or her shift and is taking a little longer than usual to respond, and the customer is expressing some impatience. The agent may be a bit drowsy and unaware of these cues. And from there, the interaction could go downhill fast. That's where behavioral analytics comes to the rescue.

These applications use Deep learning algorithms to not only detect warning signs, but also prompt the agent in real time with calls to action, such as getting a coffee to become more alert or using scripted responses to defuse a particular situation.

AI in contact centers isn't quite doing all the thinking here, so it's too soon to view the technology as an agent-replacement scenario. Rather, these behavioral analytics serve to keep agents performing at a high level. They can be a valuable tool, as long as they are deployed in a constructive, transparent manner. If done in a heavy-handed manner, behavioral analytics will look more like eavesdropping, and its use could backfire badly. AI in contact centers shouldn't be viewed as a silver bullet, but if responsibly deployed, the benefits can be worthwhile.

Dig Deeper on Communications platforms and integrations