Satellite Imagery (Alternative Data)

How Satellite Imagery Analytics Works

The raw images captured by satellites are, in themselves, just massive files of visual data. The real value for an investor lies in the analytics—the process of extracting meaningful numbers from those images. This process typically begins with a specialized alternative data provider who manages a massive library of high-frequency satellite feeds. Because individual images can be affected by cloud cover, shadows, or poor lighting, these firms use sophisticated computer vision algorithms and machine learning models to normalize and analyze the data at scale. For instance, an algorithm might be trained to recognize the specific shape and size of a car or a delivery truck from an overhead perspective. The firm can then program this algorithm to scan images of thousands of retail parking lots—such as those of Walmart, Target, or Home Depot—across the entire country on a daily basis. By comparing these daily car counts to historical averages and seasonal trends, the firm can derive a highly accurate estimate of consumer foot traffic and, by extension, future revenue. This data is then sold as a "feed" to hedge funds and institutional traders, who use it to predict whether a company's upcoming earnings report will beat or miss Wall Street's expectations. Beyond simple counting, more advanced techniques are used to measure industrial activity. Infrared sensors can detect the heat signatures of factories, indicating whether they are running at full capacity or sitting idle. Synthetic Aperture Radar (SAR) can "see" through clouds and darkness, allowing for 24/7 monitoring of oil spills, shipping lanes, and construction sites regardless of weather conditions. These technical capabilities allow for a level of granular, real-time monitoring that traditional fundamental analysis simply cannot match, turning the entire planet into a searchable and quantifiable database for the savvy investor.

Important Considerations for Investors

While the promise of "looking from space" is compelling, there are several critical considerations that investors must understand before relying on satellite imagery. The first is the concept of "signal vs. noise." Just because a parking lot is full doesn't necessarily mean sales are up; customers might be spending less per visit, or the cars might belong to employees or neighboring businesses. Successful users of this data do not look at satellite imagery in isolation; instead, they use it as one piece of a larger "mosaic" that includes credit card transaction data, web scraping, and traditional fundamental research. Another major consideration is the rapid decay of the "information edge." As more funds subscribe to the same satellite data providers, the unique advantage of knowing a car count diminishes. If every major player in the market sees that a company's traffic is up, that information is quickly priced into the stock, leaving little room for profit. This leads to a constant "arms race" for more granular data, higher-resolution images, and more sophisticated algorithms. Furthermore, investors must be aware of the high cost of this data, which can range from tens of thousands to hundreds of thousands of dollars per year, often making it inaccessible to individual retail investors and smaller firms. Finally, there are ethical and regulatory considerations. While observing the outdoors from space is generally legal and does not constitute "insider trading" in the traditional sense, the increasing granularity of these images—some of which can now identify individual car models or even people—is sparking debates about privacy and surveillance. Investors must ensure that their use of this data complies with evolving privacy laws and that they are not inadvertently crossing the line into obtaining material non-public information through unauthorized means.

Key Applications Across Sectors

Satellite imagery is used across a wide variety of industries to gain unique insights:

• Retail Analysis: Counting cars in parking lots to estimate quarterly revenue for big-box retailers and restaurant chains.
• Energy and Oil: Measuring the shadows on the floating roofs of oil storage tanks to calculate global inventory levels in real-time.
• Agriculture: Using multispectral imaging to assess crop health, moisture levels, and predicted harvest yields for commodities like corn, wheat, and soy.
• Industrial and Construction: Monitoring the progress of large-scale infrastructure projects, such as new factories or housing developments, to verify management timelines.
• Shipping and Logistics: Tracking the movement of container ships and tankers at major ports to identify bottlenecks in the global supply chain.
• Mining: Estimating the volume of ore piles and the activity of heavy machinery at remote mine sites to predict production output.

Limitations and Technical Challenges

Despite its power, satellite imagery faces significant technical hurdles. Traditional optical satellites are useless when it's cloudy or dark, which can create multi-day gaps in data for regions like Western Europe or Southeast Asia during certain seasons. While Synthetic Aperture Radar (SAR) can solve this, SAR data is more complex and expensive to process. Additionally, the "revisit rate"—how often a satellite passes over the same spot—determines the frequency of the data. To get daily updates, an investor needs access to a large constellation of satellites, which further increases the cost. Finally, the resolution of commercial imagery is limited by government regulations to prevent the identification of sensitive military sites or individual people, which can sometimes make it difficult to distinguish between subtle changes on the ground.