Despite its huge potential, the $6.8 billion Earth observation industry continues to struggle to establish a sustainable and scalable business model. New space companies are eager to democratize the vast quantities of Earth imagery data now available, but they struggle to monetize these assets effectively.
At its core, Earth observation (EO) remains dominated by defense and government contracts, with a disappointingly low adoption rate across other sectors. And that’s not acceptable when, strategically, the space industry can play a big role in solving the existential global problems we face.
While some value has been created, it’s been limited and serves only a niche audience. The business community, particularly the global companies that drive economic growth and play a critical role in tackling global challenges, should be served by EO in a way that matches their requirements. The EO industry must address this disconnect if it wants to build a sustainable business model that will allow it to make a meaningful impact on our planet’s most pressing issues.
So, how do we bridge this gap? Well, in simple terms, we must make EO affordable, accessible and, above all, usable.
Adopting a multi-purpose data model
At the moment, commercial adoption of EO is limited by its reliance on satellite tasking when capturing a designated area of interest, and by the additional analytics capabilities needed to process this data.
The single-use, single-client model — in which a satellite takes an image at a certain spot for a single, queueing customer — is economically inefficient and unsustainable because that image is not used for any other purpose. A more scalable approach would involve “always-on” monitoring, where satellites continuously capture data for multi-purpose applications. With an always-on model, the same data can serve multiple clients and applications, from environmental monitoring to agricultural analysis, without repetitive, single-purpose imaging. Some in the industry, such as Planet Labs and OroraTech, are in the early stages of pursuing this model, though OroraTech is not yet operational.
A shift from the question, “How many clients can a single satellite serve?” to “How many users need this information daily?” could dramatically transform the economics of EO. The tasking model approach may provide custom data to meet a specific purpose — whether for governmental situational awareness or disaster response — but it also locks that data into a single use case, often leaving it archived or discarded once the initial need is fulfilled.
For instance, an image of a single area could yield insights into biodiversity, soil health, crop conditions and urban development, among other things. Making this data universally available would allow EO companies to offer downstream analytics to a broader client base and make data acquisition more affordable for all involved. Distributing the costs across a broader range of clients improves the overall return on investment for each image, making EO services more financially accessible to industries beyond defense and government.
But how should the data actually be integrated into everyday life? An always-on approach with high global revisit provides the needed data to serve a multitude of global customers. But those customers would want fast, decision-making-ready information — not lots of global images or measurement data. With only a limited number of organizations that can handle raw images or data, it is hard to build a sustainable business by selling images alone.
On the other hand, dedicated analytics companies may find it difficult to obtain sufficient source data at a sensible cost, so the end customers are left with data they can’t process or service.
Value-added AI
Instead of paying per task, clients could access a suite of insights without an added markup on basic satellite services. This distributed cost model would be a critical shift towards democratizing EO and bringing its benefits to sectors that have traditionally been unable to afford or justify the expense of space-driven insights.
To provide these insights, companies will need robust artificial intelligence (AI) tools to crunch the vast amount of data. It’s just not possible for this process to be human-led due to the amount of data being acquired. This is the critical value-add needed to make EO more accessible and integrable into everyday life.
Luckily, there are plenty of options available to process global data more efficiently. For example, the industry should:
- Start utilizing higher radio speeds like Ka-band instead of X-band to increase the data download speed.
- Use wider ground station networks to get data down more often.
- Develop optical satellite-to-ground connections to have massive data transfer speed improvement.
- Develop satellite-to-satellite and satellite-to-IoT communication links to transmit data to the ground faster, bypassing the wait time for a satellite to reach a ground station.
- Develop in-orbit data rooms and cloud AI infrastructure to avoid raw data transfer to the ground altogether.
- Develop on-board processing for faster results and reduced need to transfer data in the first place.
- Develop more efficient AI algorithms to handle global data faster.
- Eventually start using quantum computing to manage massive amounts of data efficiently.
The current EO model often requires end-users to navigate between data providers, processing platforms, and analytics vendors, creating a complex web of services that can be costly and inefficient. Verticalization, however, aligns each layer of the EO stack — from satellite imagery and ground-station networks to AI-powered analytics and industry-specific reporting — into a cohesive, purpose-built solution. This alignment is crucial for maximizing the impact of EO data across industries, ensuring that clients from agriculture, insurance, urban planning, and beyond can access tailored insights without piecing together multiple service providers.
Breaking free from the status quo is essential for mass adoption
Scaling EO for mass adoption across diverse industries requires bold changes in how EO data is captured, marketed, sold and used. Embracing a multi-purpose model would allow for broader commercial adoption, benefiting industries that traditionally have had no access to — or use for — space-based insights. Ultimately, these shifts would reduce unit costs, improve accessibility and open the door for EO to become a powerful asset in solving global challenges.
The potential impact of EO on addressing issues like climate resilience, natural resource management and urban planning is profound. However, to make EO a valuable, ubiquitous resource, the industry must adjust to a model that prioritizes affordability, flexibility, and customer-centric insights. This change is not just beneficial; it is essential for EO to play a meaningful role in creating a sustainable, data-driven approach to some of the world’s greatest challenges.
Embracing these changes unlocks EO’s potential to impact industries and businesses, representing not only a competitive advantage for innovators but also a vital step for EO’s future relevance in shaping a better, more resilient world.
Jarkko Antila is the CEO of the Finnish hyperspectral microsatellite and AI-powered insights company Kuva Space. He has a master’s degree in space technology and a long background in deep-tech entrepreneurship, with a proven track record in product innovation and fundraising. Antila led technology teams at the Finnish Research Center and Inficon. He co-founded Spectral Engines, a successful spectral sensor company where he served as CEO for six years before its exit, and the deep-tech business consultancy firm Leaping Boulder.
