GreenKey Technologies will introduce a new product in early October, called Blotter, which is a front-office solution for leveraging a lesser talked-about alternative data set: human voice and the dynamics in communication. Built together by GreenKey and IPC, which purchased a minority stake in GreenKey in March, the tool will be a front-end dashboard that allows clients of both firms to view and analyze audio conversations within their organizations.

The Blotter will be an add-on service with IPC’s late-August release of its real-time audio gateway tool (RTAG), an API that unlocks and creates a data feed from all the telephonic conversations within a firm, which can then be used for analytics, to populate forms, or help sales teams better manage the history of client requests. The Blotter will serve as the visual window into that data, says GreenKey’s founder and CEO, Anthony Tassone, and it’ll allow customers to search and query quotes and trades inside their organizations, and automate parts of traders’ workflow.

Voice and conversational nuance can be used to find alpha, Tassone contends, both in the market and operationally. Tassone, who also sits on the IPC advisory board, estimates millions of quotes per day generated on IPC’s network of more than 100,000 traders aren’t being fully captured—but they could be.

“If you think about the current workflow, these traders are shouting back and forth quotes,” Tassone says. “Nobody’s capturing that. Then they do a trade, then five or 10 minutes later [they] type it into a system. So everyone finds out about these trades 10 minutes after they happen, and nobody knows what quotes actually occur.”

The release is part of a growing trend in adoption of natural language processing (NLP) technologies in over-the-counter (OTC) trading and sales, as well as fixed income and equities, where quote and trade data are also mounting.

“A good analogy would be in the mid-90s when open outcry went electronic,” Tassone says. “OTC sales and trading are very bespoke, and it involves a lot of natural language. The time spent on tedious tasks and communications processes can be vastly improved. NLP will automate those tasks and bring new efficiencies in the same way electronic trading did.”

One use case, particularly for the buy side, is around compliance. Hedge funds, for example, are keen to have a handle on trade reconstruction in order to comply with Mifid’s best-execution laws. As a result, Tassone says a common request from buy-side firms is for GreenKey to transcribe and analyze their audio and chat data and unify it into one big audit trail.

But the biggest mover in NLP adoption is not compliance at all, Tassone says. The major question GreenKey tries to answer when analyzing firms’ audio files is determining why a quote doesn’t turn into a trade. Often those reasons are that they didn’t offer the best price or size, or that their response time was too slow. And at the same time, while reconstruction and surveillance are pragmatic uses for NLP, OTC traders are just trying to keep their heads above water in a sea of data.

Sales traders are being bombarded with unstructured data, Tassone says. Not only are there more messages, calls and quotes than ever before, but the state of trading is in flux. While a younger generation emerges into the workforce, veterans age out, and buy- and sell-side players devote larger parts their budgets to tech projects. It’s a major challenge, he adds, and it’s one that will have to be sorted out by younger and fewer human beings.