Satellite imagery has moved far beyond the realm of government agencies and specialized scientific groups. Today, it’s a critical asset for enterprises across agriculture, insurance, energy, and logistics. If you need to monitor crop yields in Brazil, assess flood risk in Florida, or track shipping containers in Singapore, views from orbit provide the ground truth you need.
But the market is crowded. There are more constellations in orbit than ever before, and accessing their data is no longer straightforward. Enterprise leaders face a complex choice: do you negotiate directly with satellite operators, or do you buy through an aggregator platform?
This guide breaks down exactly how to evaluate your options so you can build a geospatial strategy that scales with your business.
The Two Paths: Direct Purchasing vs. Aggregation Platforms
Before diving into technical specs like resolution and revisit rates, you need to decide on your procurement model. This structural decision will impact your budget, your legal overhead, and your engineering resources.
Option 1: Direct Purchasing from Providers
This is the traditional route. You identify a satellite operator (like Vantor, Airbus, or Planet), negotiate a contract, and build a pipeline to ingest their specific data format.
Pros:
- Relationship leverage: Direct relationships can sometimes yield better pricing for massive, predictable volume.
- Tasking priority: If you need to “task” a satellite (tell it exactly where to look and when), direct contracts often give you higher priority than third-party users.
- Specific sensor access: If you need highly specialized data, going direct ensures you get exactly the raw data capabilities of that specific sensor without middleware interference.
Cons:
- Vendor lock-in: You are beholden to one provider’s coverage and uptime. If their satellite is over the wrong hemisphere or under maintenance, you have gaps.
- Integration headaches: Every provider has a different API, different metadata standards, and different delivery mechanisms. Using three different providers often means building three different integrations.
- Minimum commitments: Direct contracts often come with hefty annual minimum spends that can be hard to justify for pilot projects.
Option 2: Aggregation Platforms and Marketplaces
These platforms act as a single gateway to multiple satellite constellations. Think of them as the “Amazon” of earth observation.
Pros:
- Single API: You write code once and access data from multiple providers. This dramatically reduces engineering overhead.
- Better coverage: If Provider A has cloud cover over your target area, the platform can seamlessly check if Provider B has a clear shot.
- Flexible licensing: Many platforms offer transactional models, allowing you to buy small areas or single images without six-figure annual commitments.
Cons:
- Margin stacking: The platform takes a cut, which can make individual images slightly more expensive on a unit basis compared to bulk direct deals.
- Abstraction limits: Sometimes the platform “smooths over” the unique, raw capabilities of a sensor to make it fit a standard format. If you need extreme technical specificity, this abstraction layer can be a hindrance.
5 Key Factors for Evaluation
Regardless of which procurement path you choose, you must evaluate the underlying data and service against these five pillars.
1. Image Quality (Resolution and Spectral Bands)
“Resolution” describes the amount of detail captured in an image, typically measured in centimeters or meters per pixel. Satellite imagery can be categorized as follows:
- Very High Resolution (15cm – 30cm): Provides exceptional detail, allowing you to identify individual vehicles, small objects, and subtle infrastructure changes. This level is well-suited for urban planning, high-security monitoring, and precision mapping tasks.
- High Resolution (30cm – 1m): Enables recognition of cars in parking lots, detailed equipment, and building features. Commonly used in city management, utilities, and infrastructure analysis where fine details are important but the absolute highest clarity is not required.
- Medium Resolution (1m – 7.99m): Offers clear views of larger features like field patterns, forest boundaries, or sections of roadways. Ideal for applications such as agriculture monitoring, general land use assessment, and environmental studies where object-level distinction isn’t needed.
- Low Resolution (8m+): Primarily for observing broad changes at regional to global scale, such as shifts in vegetation, water bodies, or weather systems. Here, the focus is on patterns and trends rather than on identifying specific objects or features.
However, don’t stop at spatial resolution. Ask about spectral resolution. Does the provider offer Near-Infrared (NIR) or Short-Wave Infrared (SWIR) bands? For an agricultural company, a 3m image with NIR bands (to measure plant health) is infinitely more valuable than a sharp 30cm image that only has visual colors.
2. Revisit Rate (Temporal Resolution)
How often do you need a new picture?
- Financial trading: You might need daily updates on oil storage tanks.
- Real estate development: A quarterly update might suffice.
Check the provider’s revisit rate for your specific area of interest. A provider might claim “daily revisit,” but if that relies on a satellite that passes at 2:00 PM when your region is usually cloudy, that theoretical revisit rate is useless.
3. Licensing and Data Rights
This is the most common trap for enterprises. Satellite imagery is rarely “sold”; it is licensed.
- Derivative works: If you use an algorithm to count cars in an image and then sell that car-count data to a hedge fund, is that allowed? Some licenses forbid reselling insights derived from the data.
- Internal distribution: Can you share the image with your contractors, or just your full-time employees?
- Perpetuity: Do you keep the data forever, or do you lose access if you stop paying the subscription?
Always have your legal team review the “End User License Agreement” (EULA) specifically for derivative works clauses.
4. Ease of Access and Delivery Speed
How long does it take to go from “I want this image” to “I have this image”?
- Archive vs. Tasking: Pulling an image from the archive (past photos) should be instant. Tasking a satellite for a future shot can take days or weeks.
- Latency: Once the satellite snaps the photo, how fast does it downlink? For disaster response, a 24-hour delay is unacceptable. Look for low-latency providers who can deliver data within hours of capture.
- Cloud readiness: Does the provider deliver data directly to your AWS S3 bucket or Google Cloud Storage? Or do they force you to download massive FTP files manually? Modern enterprises need cloud-native delivery.
5. Scalability and Cost Predictability
Running a pilot project on a laptop is very different from scaling a solution globally.
- Minimum Order Quantity: Some providers require you to buy 100 sq km minimum, even if you only need to look at a single factory. This kills ROI for small targets.
- Pricing model: Is it per square kilometer? Per megabyte? Or a flat subscription?
- API Limits: Ensure the provider’s API can handle the volume of requests your automated systems will generate.
Making the Decision
For most enterprises starting their geospatial journey, aggregation platforms offer the safest entry point. They minimize technical debt, simplify licensing, and allow you to test value without massive capital expenditure.
However, as your usage matures, the economics may shift. Once you have massive, predictable volume in specific regions, negotiating access to data from a specific provider may make more sense.
The satellite market is launching new capabilities every month. The “best” provider today might be surpassed tomorrow. Therefore, the most important quality you can buy is flexibility. Avoid locking your entire enterprise into a rigid, multi-year single-source contract unless the price advantage is overwhelming. Build for agility, and the view from above will always be clear.
