Alternative Data
Alternative data refers to information used in investment analysis that comes from sources outside the traditional set of prices, financial statements, and economic releases. Examples include satellite imagery, credit card transaction records, web traffic, app usage, and text from news and filings.
The appeal of alternative data is timeliness and breadth. Traditional data describes what a company reported after the fact, while alternative data can offer earlier or more granular signals about activity as it happens. In machine learning for finance, these datasets serve as raw material that is converted into predictive features.
Categories of Alternative Data
Alternative data is often grouped by how it is generated. Each category carries its own strengths and its own collection and quality challenges.
| Category | Examples | What It Can Signal |
|---|---|---|
| Individual activity | Credit card spending, app usage, search trends | Consumer demand and revenue trends ahead of reporting |
| Business activity | Shipping records, job postings, supplier data | Operational momentum and supply-chain conditions |
| Sensor and imagery | Satellite photos, geolocation, weather | Physical activity such as foot traffic or production |
| Text and media | News, filings, transcripts, social posts | Sentiment and disclosure changes, often via natural language processing |
Most of these sources arrive in messy, unstructured form. Turning a stream of satellite images or transaction records into a usable signal requires substantial feature engineering, and text sources in particular depend on natural language processing (NLP) to become numerical features.
Why It Matters
The theoretical case for alternative data rests on information advantage. If a dataset reveals a company's sales trajectory before the quarterly report, an investor who can process that data may form a more accurate view than the market consensus. The efficient market hypothesis suggests such advantages are hard to sustain, because once a signal is widely used, prices adjust and the edge fades.
Key Principle
The value of an alternative dataset lies not in its novelty but in the durable, non-public information it contains. A dataset that many investors already use offers little advantage, because its information is already reflected in prices. Edge comes from data that is genuinely informative and not yet widely exploited, which is precisely the kind of edge that tends to erode over time.
Known Limitations
Limitations to Keep in Mind
- Short and shifting history. Many alternative datasets have only a few years of history, which makes it hard to know whether a pattern is real or a coincidence. Short histories raise the risk of overfitting to a brief and unrepresentative period.
- Survivorship and coverage gaps. Data providers may cover only certain companies, time periods, or regions, and may drop entities that fail. This can introduce survivorship bias that flatters apparent results.
- Look-ahead and point-in-time problems. Alternative data is often revised or backfilled, so a value available today may differ from what was knowable in the past. Without careful point-in-time handling, look-ahead bias creeps in, an issue that out-of-sample testing helps surface.
- Crowding and decay. As more investors license the same dataset, any edge it offered tends to shrink. Signals that worked in early testing may weaken once the data becomes widely adopted.
- Cost, legal, and privacy concerns. Alternative data can be expensive, and some sources raise questions about privacy, consent, or compliance. These considerations can outweigh the analytical benefit and must be assessed before use.
Applications
Alternative data feeds into quantitative models in the same way traditional data does: it is cleaned, converted into point-in-time features, and validated. The resulting features may complement characteristics from factor investing, adding a dimension that prices and accounting statements do not capture. Common uses include estimating company sales ahead of reports, gauging consumer sentiment, and monitoring supply-chain activity.
The discipline applied to alternative data matters more than the data's novelty. Because these datasets are noisy, short, and prone to revision, honest cross-validation and a clear economic rationale separate a genuine signal from a statistical artifact. Alternative data broadens the information set available for analysis; it does not remove the need for careful judgment about what the information means.
Further Reading
- Monk, A., Prins, M. and Rook, D. (2019). "Rethinking Alternative Data in Institutional Investment." The Journal of Financial Data Science, 1(1), 14–31.
- Katona, Z., Painter, M., Patatoukas, P.N. and Zeng, J. (2018). "On the Capital Market Consequences of Alternative Data: Evidence from Outer Space." Working paper.
- de Prado, M.L. (2018). Advances in Financial Machine Learning. Wiley.
Related Terms
Foxholm Financial is a fee-only registered investment adviser serving Georgia. We bring quantitative rigor to every client engagement. Explore our services or get in touch to discuss how we can help. To see how this kind of analysis informs real client work, explore a Strategic Portfolio Review.
Are you an institution or FinTech firm? Learn about our Quantitative Consulting Services.
Foxholm Financial trains the next generation of quantitative analysts. Students and early-career researchers can explore our quantitative investment fellowships.
This content is for educational and informational purposes only and does not constitute an offer to sell or a solicitation of an offer to buy any securities. Nothing herein constitutes investment advice or recommendations tailored to your individual situation. All investments involve risk, including the potential loss of principal. Past performance is no guarantee of future results. Information presented is believed to be factual and up-to-date, but Foxholm Financial does not guarantee its accuracy and it should not be regarded as a complete analysis of the subjects discussed. Before making investment decisions, consult with a qualified financial advisor who can evaluate your specific circumstances.