Meet Your New Robo-Analyst

By Tal Sansani - October 1, 2018

The quarterly corporate earnings conference call is highly choreographed. Corporate executives frame financial results and expectations in a positive light—and with a carefully crafted statement. Investment analysts ask probing questions, attempting to read between the lines to glean new information to update their views of the company. A silent listener has recently joined these discussions and without asking a question, digests every word, takes meticulous notes and quickly generates a detailed report. Meet your new robo-analyst.

Thousands of earnings calls occur worldwide every quarter. It would take one person more than 700 days, working 24/7, to listen to a year's worth of calls. This is where the robo-analyst comes in. It doesn't get tired, get sick or go on vacation. It doesn't play favorites or have pet peeves. It provides intelligent, data-driven analysis in mere seconds.

The Intersection of Data, Technology and Investment Expertise

Artificial intelligence (AI) is changing the investment management industry and helping investors make better decisions. The emergence of the robo-analyst is just one example of this larger trend and signifies a natural evolution and convergence of data, technology and investment competition. American Century Investments® robo-analyst sifts through conference call transcripts, using natural language processing (NLP) and psycholinguistics to analyze the nuanced speaking patterns of management that may signal underlying challenges within the company.

An early iteration of our model sought mainly to determine if company management was optimistic or pessimistic. It captured what was being said, but not how. For example, what if corporate executives were exaggerating? The next step was to train the model to process information more like a discerning human, reading between the lines and identifying subtle forms of deception.

Decoding Corporate Speak

First, we had to identify the patterns of deception, which we did not find in traditional financial literature. Grounded in fields like psychology and computational linguistics, our research team identified four key signals that consistently stood out across variety of contexts. They can be seen in situations ranging from children lying to their parents to corporate executives trying to placate their boards of directors.

  • Omission: Failure to disclose key details. Management teams that avoid details or key data may be withholding relevant information from the investment public.
  • Spin: Exaggeration and overly scripted language. Conference calls are rampant with corporate enthusiasm. However, our model negatively views abnormal levels of corporate spin, especially in the face of analysts' uncertainty and questioning.
  • Obfuscation: Overly complicated storytelling. In the midst of underlying business challenges, management tends to avoid the simple story. "Hand-waving" and complex answers to simple questions often signal obfuscation.
  • Blame: Deflecting responsibility. Management takes credit for success but assigns failure to external forces. Management is either covering up real business issues or doesn't understand them. Either case should make investors uneasy.

Suspicious communication patterns from corporate executives have important investment implications. We have years of empirical evidence suggesting that management's candor reflects its leadership values and sends a meaningful signal about the company's underlying health. Communication quality also affects the way investors act upon information, impacting near-term price movements in meaningful and predictable ways. Longer term, poor communication patterns may also raise questions about corporate governance and cause investors to lose trust.

Putting the Robo-Analyst to Work

Big Data and AI enable investors to detect meaningful relationships across large volumes of increasingly complex data sources. The robo-analyst model benefits from an unbiased approach and immense scale. Human analysts may be more skilled at analyzing a small set of companies, but machines can systematically and objectively examine thousands at a time. We view these two stock-picking perspectives—the carefully discerning human analyst and the trained, data-driven robo-analyst—as complementary forces. Each provides crucial insight that helps us make better decisions that benefit our clients.

Tal Sansani
VP & Sr. Quantitative Analyst
Disciplined Equity Group
  • Related Articles
  • More From Author

9 Tax Tasks to Do Before Year-End

April's tax deadline may seem like far off, but a few year-end tasks can put things in order before your 2018 filing.

Meet Your New Robo-Analyst

It would take one person more than 700 days, working 24/7, to listen to a year's worth of earnings calls. This is where the robo-analysts provide value in the form of intelligent, data-driven analysis in mere seconds.

Is Your Portfolio Too Spread Out? There's a Simpler Way

You might have more money in more places than you'd planned. Take back control by streamlining your portfolio.

    Meet Your New Robo-Analyst

    It would take one person more than 700 days, working 24/7, to listen to a year's worth of earnings calls. This is where the robo-analysts provide value in the form of intelligent, data-driven analysis in mere seconds.

    Evidence of Gender Bias: An Analysis of Conference Call Dialogue

    Tal Sansani of the discipline equity group tackles gender bias with data from a recent analysis of corporate earnings call transcripts.

      Investment return and principal value of security investments will fluctuate. The value at the time of redemption may be more or less than the original cost. Past performance is no guarantee of future results.

      The opinions expressed are those of American Century Investments (or the portfolio manager) and are no guarantee of the future performance of any American Century Investments' portfolio. This material has been prepared for educational purposes only. It is not intended to provide, and should not be relied upon for, investment, accounting, legal or tax advice.