The future of successful investing belongs to those who can merge the rigor of quantitative finance with the insights of qualitative assessment. AI models can provide an invaluable resource.

Editor’s note: This article was originally published on Dec. 12, 2025, on Forbes.com.​

I​n the relentless pursuit of alpha, investors have long turned to quantitative data—the hard numbers of earnings reports, balance sheets, and economic indicators. Just look at the myriad quant fund managers making a comeback in 2025. The industry has built complex models and algorithms to parse all this “big data” at lightning speed. But in a market where everyone has access to the same Bloomberg terminals and SEC filings, the quantitative edge is rapidly diminishing.

I believe the next frontier for outperformance won’t be found in a new number; it will be found in the words between the numbers. I encountered a few such service providers doing just that while attending Hong Kong’s 10th Annual FinTech Week earlier this month.

Specifically, I’m referring to the systematic analysis of qualitative data: the nuanced language in earnings call transcripts, the subtle shifts in regulatory filings, the geopolitical sentiment in newsfeeds, and the operational realities hinted at in supplier reports. This “unstructured data,” which constitutes over 80% of all enterprise data, according to IBM, has been largely ignored by traditional analysis. Until now.

The tools of artificial intelligence—specifically natural language processing (NLP) and sentiment analysis—are no longer science fiction. They are becoming an essential microscope for the modern investor, allowing us to quantify the unquantifiable and detect signals in the noise that others miss.

Reading between the lines of earnings calls

An earnings call is more than a scripted recitation of results; it’s high-stakes theater where management’s tone, confidence, and evasion can be more telling than the EPS figure itself. Here are two specific elements of these calls where AI analysis can make a difference:

1. Sentiment and certainty: AI models can now score management sentiment and measure the frequency of “uncertainty words” (e.g., “hopefully,” “we’ll try,” “challenging environment”). A sudden dip in confidence between quarters can be an early warning sign of underlying stress that hasn’t yet appeared on the income statement.
2. Competitive intelligence: By tracking how often a company mentions its competitors, and in what context, models can map competitive threats and strategic focus in real time. A sudden surge in mentions of a disruptive new entrant is a data point no investor could afford to ignore.

The ‘mosaic’ in the mundane

I’m not calling for qualitative data to replace fundamental analysis, but rather to augment it. The best investors have always been the ones who could build a “mosaic” of information from disparate sources. AI simply allows us to do this at a scale and speed previously unimaginable. My own firm’s investment committee is eager to benefit from this more over time, as today we do not regard it as a panacea but rather a further consideration.

Consider a few practical applications:

1. Analyzing 10-K and 10-Q filings: An AI model could track year-over-year changes in the “Risk Factors” section of annual reports. The introduction of a new risk or the subtle rewording of an existing one can signal a shift in the boardroom’s perception of legal, regulatory, or market threats.
2. Geopolitical and supply chain sensing: By analyzing global news data, an investor can gauge the potential impact of a port strike or a regional conflict on a company’s complex supply chain, moving from reactive headlines to proactive risk assessment.
3. B2B sentiment as a leading indicator: The sentiment expressed in earnings calls of a company’s key suppliers or customers can serve as a powerful leading indicator for demand.

The peril of ‘garbage in, gospel out’

The power of this approach comes with a significant caveat: The models are only as good as the data and the questions fueling them. Bias in the training data or poorly designed sentiment lexicons can produce dangerously misleading results. An investor must understand the why behind a sentiment score, not simply accept the output. The “quantitative fundamentalist” must now also be a “qualitative skeptic.”

The future of successful investing belongs to those who can merge the rigor of quantitative finance with the nuanced insight of qualitative assessment. It’s about training ourselves, and our models, to listen more carefully. The data is all there, waiting to be read. The question is, are you listening?

The opinions expressed in this article are those of the author and the sources cited and do not necessarily represent the views of Proactive Advisor Magazine. This material is presented for educational purposes only.

The information provided here is not investment, tax, or financial advice. You should consult with a licensed professional for advice concerning your specific situation.

Accredited Investment Fiduciary (AIF) is a registered certification mark of Fi360, a Broadridge company. Certified Fund Specialist (CFS) is a registered certification mark of The Institute of Business & Finance.