In their never-ending quest for alpha, analysts and investors scrutinize the market-moving management language used in Earnings Calls for any evidence that would confirm or discredit their investment opinion about those companies. The problem with this traditionally instinctive approach to listening in on Earnings Calls and forming an opinion of management is that the analysis is entirely subjective, time-consuming and impossible for humans to do at scale.
Consequently, analysts and investors are increasingly using a new prism through which they view the information contained in management presentations that converts their instinct into real data. The technology of Natural Language Processing (NLP) enables them to consume and analyze millions of words and numbers for additional clues about management sentiment, confidence, clarity and credibility in minutes instead of days.
In our primer on NLP, we aggregated Earnings Call trends from individual companies to their geography and sector as well as the Covid- 19 theme. We then used this data to show the big picture in a Sentiment Index that was benchmarked against the MSCI GCC Index and net profit generated by GCC companies. Essentially, this is a 'nice-to-have' to get an alternative data point on market sentiment.
In this report, we move on from just exploring these big picture trends and start building the case for Quant Lens to become a 'must-have' as we now take a first look at how divergences between management and its audience can be quantified on a stock-specific level.
We let you be the judge on the result.
