Alternative data: benefits and challenges
What benefits can the use of alternative data bring to financial institutions?
Quarterly earnings reports and industry whitepapers are often slow and infrequent and as these traditional sources of information are readily available, they tend to offer limited actionable intelligence or insight. No executive should go to their board and present a plan based on data from 6 months ago when they could be using data from yesterday. While traditional sources have served them well in the past, the rise of Big Data has brought about an exponential amount of new data that they could be overlooking.
Another benefit of alternative data over traditional data is its frequency. Instead of data being collected every 6 months or yearly, alternative data is accessible monthly, daily or even in real-time. Companies can conduct more information-dense and precise analysis, resulting in not just richer insights into the present performance but a forecast into the future.
Alternative data can help to give investors an edge. Datasets that are not commonly used can have less alpha decay associated with them. We are still at an early stage when it comes to the widespread adoption of alternative datasets within financial markets.
An information edge is the cornerstone of successful alpha generation. Whether an information edge is garnered through on-the-ground assessments of company fundamentals, supply chain dynamics, or structural themes—or, indeed through the systemisation of insights gathered from diverse alternative data—these inputs collectively serve to build a fuller picture of a company’s long-term prospects than any of these factors would if viewed in isolation. Therefore, alternative data serves as a novel input for investors to use alongside more traditional information sources, and its considerable expansion is creating a new dimension for information asymmetry to occur.
“Investors today don’t wait until a company publishes its financial results,” he continues. “They’re looking for additional datasets and indicators for how a company is doing during a quarter. If they see a company is [consistently] importing a certain level of goods and suddenly that goes up or down, it’s an indication of what’s happening with demand for its products.” (Ewout Steenbergen)
What are the key challenges faced when attempting to source and utilize alternative data?
The key challenges include structuring the data and being able to find the right alternative dataset for you. It isn’t possible to test every single alternative dataset out there. Here are the key challenges:
- The data is unstructured, hence requires NLP for text; or images require special processing through AI.
- The history of these datasets is limited, so the historical archive is not always large enough to make proper backtesting. We need to wait/accumulate until it’s testable.
- Content integrity, providers were not contemplating selling it and we need to normalize datasets and put it into a format that is useful.
- Data vendor risk—including risks around credibility, reputation, and continuity (i.e., the risk of the vendor ceasing to exist or provide a set of data)
- Data provenance—including ethical implications around how data is sourced, stored, shared, and used
- Material non-public information risk—the act of accessing data in itself does not necessarily constitute it as public information
- Data integrity—sparsity, completeness, accuracy, and bias assessments