CEOs and CFOs are decidedly more nervous when fielding questions about China during earnings calls this year. What’s more, they are more likely to be deceptive with their answers. "Deception associated with questions on China has skyrocketed this quarter, up about 50% from last quarter and more than double a year ago," according to a study by text analytics provider Amenity Analytics.

Amenity Analytics is one of a handful of companies that are applying natural language processing (NLP), sentiment analysis and machine learning to the financial sector, evaluating earnings calls and other public meetings to unearth information of value to an investor. It is also rare technology that offers a clear path to ROI. Decoding language from quarterly calls and using that information for investment decisions can add an extra 6.3% to annualized returns, according to a study by Bank of America (BoA) Merrill Lynch that was reported in the Financial Times. BoA found that this kind of NLP outperformed a basket of 2,500 stocks over a 9-year period.

The Maturation of NLP

This level of financial analysis illustrates the next-gen use of natural language processing — namely when it has been married to artificial intelligence (AI) and sentiment analysis. But even bots without the bells and whistles afforded by machine learning and AI are providing value to analysis and investors. 

There are a variety of open source and inexpensive NLP commercial tools being used by investors today, said Doug Barbin, principal at Schellman & Co. Such bots are "programmed to look for keywords or performance indicators that could quickly alert a financial analyst to potentially adjust their recommendations on a particular stock. Additionally, key messages can be summarized and restated in the form of a news summary without the need for reporter intervention." 

This activity may come as a surprise for those who associate bots with the chatbots being deployed in customer service. "Bots aren't just ruled-based or knowledge-based responders, and their use cases extend far beyond Q&A and natural conversation bots," said Stephen Blum, co-founder and CTO of PubNub. He noted that chatbots are incredibly valuable in the world of transcription as automated stenographers who can quickly and accurately turn human speech into text, citing such available services as OBS, Watson Conversation, Amazon Transcribe and Temi. 

Related Article: Natural Language Processing Is Hitting Its Stride

Publicly-Traded Companies Take Note

To be sure, many of Blum’s aforementioned services cannot do what Amenity Analytics or any of its competitors — Prattle is one notable example — can do. But while they may be relatively few in number, the amount of such high-level offerings are growing and already making a mark on publicly-traded companies. Gartner, in a blog post from December 2018, highlighted the growing use of bots that are listening to earnings calls. "As investors look for an edge, they are turning to automated speech and text analysis for additional clues about performance," wrote contributor Tim Raiswell. The general idea is that "NLP and sentiment analysis can identify underlying and unspoken meaning about company performance and potential that isn’t explicit in direct language from senior leaders." For instance, according to the post, indicators of hidden sentiment include indirect answers to questions, exuberant words, a lack of fillers and qualifying statements.

Learning Opportunities

Related Article: Confused by AI Hype and Fear? You’re Not Alone

Going Mainstream

As this advanced technology matures and proliferates, Barbin expects its use will spread to all sectors of the business community, such as private or less public meetings restricted to investors or venture capital firms. "If an investor or stakeholder can't be in two places at once, record the call and catch-up with the AI generated cliff notes," he said. 

As it spreads, end users will have to take care that the information is handled appropriately, he added. From a security perspective, for example, a company will need to pay close attention to data storage, especially when a large portion of these technologies are offered as-a-service, such as a cloud service, he said. "While the audio may have been private or controlled, the transcription now lives on a server somewhere, with data protection you can't guarantee."

As for the CEOs and CFOs of these companies, they should make sure they are sending the right message to the bots that may be listening to their earnings calls, Raiswell wrote in his Gartner post. "Quarterly earnings calls are one of the most powerful tools a company uses to communicate with the investment community. As your investor relations team prepares for the next call, make sure they don’t ignore these emerging technologies."