Conversations are the nuts and bolts of human communication. Yet we’re still exploring precisely how they work and how to gain maximum value from them.

Conversation analysis straddles the academic disciplines of sociology, psychology and linguistics, with research continuing to yield insights into the patterns and infrastructure of human dialogue. Since Gridspace artificial intelligence (AI) and machine learning are similarly centered on speech science, a study reported in the Harvard Gazette caught our eye recently. It addresses a topic we’ve worked hard to perfect in our own software—how to recognize when it’s time for the talking to stop.

The study, published in the Proceedings of the National Academy of Sciences, concludes that most conversations don’t end when the participants want them to. It’s maybe a coordination problem that seems to happen when people don’t want to appear rude. Or, of more concern in the commercial context, when one side feels they haven’t fully achieved the result they desired.

“Social connection is essential to physical and psychological well-being, and conversation is the primary means by which it is achieved,” the report explains. “And yet, scientists know little about it—about how it starts, how it unfolds or how it ends. Our studies attempted to remedy this deficit, and their results were surprising: Conversations almost never end when people want them to! At a moment in history when billions of people have been forced to curtail their normal activities and to reimagine this one, a scientific understanding of conversation could hardly be timelier.”

Mastroianni told the Harvard Gazette that this happens for two reasons. “One is that people don’t want to talk for the same amount of time, so they have that coordination problem to solve. Two, they don’t know what the other person wants, so however they want to solve that problem, they can’t do it because to do it you need to know what the other person wanted, and they would have to know what you wanted.”

The study also provides data on what people talk about. As we might expect, many respondents (44%) reported the purpose of their talk was to “chat or hang out.” But a majority (49%) said their conversations had a more transactional purpose, focusing on the need to “solve a problem,” “make a decision together” or because one of them needed something from the other, such as advice or information.

This certainly rings true for us at Gridspace. People are comfortable and well-practiced at solving problems, making decisions or asking for help utilizing verbal exchanges. Speech remains central to customer interaction for most organizations, and tech advances of the past few years are nurturing this preference, not eliminating it.

“Conversation is common, but it is not simple,” the report concludes. “To converse, people must generate and comprehend language in real time, alternate turns in rapid sequence, infer what their partners know, remember what has and hasn’t been said and much more. Conversation is a bundle of complex tasks.”

Mastering these complex conversational tasks so that our clients don’t have to is central to the Gridspace offering. Our software makes it easy for contact centers and developers to quickly understand customer requirements, synchronize agent responses and drive successful outcomes. Put simply, legacy voice systems were not designed for the intense demands of modern speech services. So, we built our own speech processing system, in house, to enable low-latency speech analytics, live agent guidance and natural-sounding, interwoven voice bots to work alongside them.

But, let’s get back to those endings. As the song goes, it’s not where you start that matters most (although picking up calls promptly is an initial hurdle many contact centers are still failing to surmount). It’s the quality of the conversation itself, and its framework of opening, feedforward, business, feedback and closing—that will ultimately lead to a great finish for your business.

Our virtual agent, Gridspace Grace, provides consistent, friendly and natural interactions that align perfectly with human conversational infrastructure. These interactions are not just phrases, but real-time interactive engagement with callers, including smooth navigation towards those all-important timely conclusions.

Business is moved by conversations. It’s time to make sure yours are going in the right direction.