I referred to 2025 as the year EO stopped pretending – dual-use went mainstream and defense became the default growth engine, just as civilian EO programs came under question. We can't deny that the center of gravity of EO has shifted, and no, I don't think it is shifting back again. At least for a while.
So, what does 2026 hold? Here are my 15 predictions that could shape the year ahead. And, yes, I went bolder than usual!
To make it easier for you to scroll through, I have somewhat loosely organized them by themes: Geopolitics & sovereignty (1–5) | Market structure & procurement (6–10) | Technology evolution (11–12) | EO adoption (13-14) | People (15)
1. 2026 will see Earth observation used more routinely as a geopolitical lever
2025 gave us a preview of this: the US suspended Ukraine's access to government-purchased commercial imagery, using it as diplomatic leverage. That was just the start. In 2026, I expect more of this, whether that is through restricting access to imagery, shutter control on commercial providers, or tightening export controls on satellites, sensors, and intelligence solutions.
EO is no longer just a tool for observation, it is becoming a lever of power. The sector spent years proving it was commercially viable, now it has to reckon with the fact it has actually become strategically valuable.
2. The tension between EU-integrated EO architecture and national sovereign capabilities will intensify
Europe wants both collective efficiency and national control of EO. ESA's role as integrator and NATO's growing interest in space-based ISR create overlapping coordination attempts, but I believe the European member states will continue hedging with national programs.
With EU's upcoming budget (MFF for 2028-2035) and the implementation of the new European Resilience from Space program amid growing uncertainty over the transatlantic alliance, conditions seem perfect for more togetherness on EO strategy. However, when EO becomes security-critical, cost-sharing logic loses to sovereignty, but the problem is no one country can actually afford to go it alone. I don’t expect this tension to resolve in 2026.
3. China will take on a bigger role in the global EO sector as both a capacity-building partner and an institutional data provider
As more countries invest in sovereign EO, China will emerge as a potential partner of choice with no ITAR friction, competitive pricing, willingness to transfer capability, not just sell data. Egypt showed what this looks like: China helped build local capacity, deepening trade and dependency.
Meanwhile, China is making its imaging and weather data more accessible globally. With increasing launches of Chinese imaging and weather satellites, growing calls to open up China's EO archives, and remote sensing research peaking in the country, expect China to expand its footprint in the global EO sector, by making more data available to global users.
4. Isaacman will push efficiency-focused reforms at NASA's Earth Science division, but whether this shakes the foundations of civilian EO or modernizes it remains uncertain
NASA's new administrator, Jared Isaacman's "science-as-a-service" framing faced pushback, but the underlying diagnosis isn't wrong. Civilian EO, especially in the US, has efficiency problems: cost overruns, long development cycles and duplication across agencies.
So, I'd bet Isaacman moves forward regardless – whether through more commercial data acquisition, tighter cost/schedule discipline or better mission formulation. The question isn't whether these reforms happen, but I am curious whether industry will be ready to deliver without long-term funding commitments and clearer data licensing frameworks. On NASA's side, we will probably see better risk management and anticipation as it depends more on commercial providers.
5. Earth science worldwide could become marginalized as civilian EO is absorbed into dual-use architectures
EO is now viewed primarily through a national security and disaster management lens. According to our analysis, government EO spending from 2021-2025 hit historic highs, but it turns out that actually less than 15% of those investments went to pure EO science missions. As civilian programs continue folding into dual-use frameworks, science becomes an afterthought.
I believe EO science budgets will remain stable, but deprioritized. The risk isn't that science missions disappear, but they become dependent on defense priorities for funding, thus losing the autonomy that makes long-term research possible.
6. EO constellation economics will remain under pressure as SpaceX's dominance tightens, with hopes pinned on Starship and Neutron to eventually break the bottleneck
EO constellation economics depend on launch costs. SpaceX has played a major role in the growth of the sector, with its rapid launch rate and relatively lower prices, but its market dominance means that access to launch is increasingly constrained and alternatives are limited. Many EO business models were built assuming rapidly falling launch costs, but that assumption is running into a reality of limited alternatives and constrained slot availability.
If SpaceX's Starship and Rocket Lab's Neutron progress as hoped, it opens a path to better economics. If they don't, the squeeze continues, and some constellation plans may need to be rescoped or abandoned.
7. At least one EO satellite company will fold or be acquired under distress in 2026, while at least two EO companies will go public
The sector is bifurcating. Established players are expanding internationally, outspending smaller competitors on R&D, and widening the capability gap. Companies without scale or differentiation won't survive, even if governments favour local players in the short-term. So, I expect at least one distressed exit.
On the other hand, it is not all bleak: I think at least two companies (Iceye and Vantor) could go public in 2026 – the former with a fast-growing (valued at $2.4B) SAR business and the latter as a cash out by the private equity firm that acquired Maxar and restructured it. The sector is maturing, and maturation means more consolidation and more IPOs.
8. Consolidation will accelerate in 2026, both horizontally across the value chain and vertically by defense primes
Horizontal consolidation means companies expanding across the value chain, from acquisition to platforms to intelligence, to offer more complete solutions to their end-users. Vertical consolidation refers to vertical market leaders acquiring EO capabilities. Watch for some defense primes that build integrated ISR offerings through acquisitions of EO businesses.
Both trends reflect the same reality: standalone EO is a tough business. Watch out for deals in both directions. I also think that the trend of strategic acquisitions of EO-based solutions by enterprises in infrastructure, insurance and finance verticals will continue – this has been a recurring trend over the past couple of years (see #12 and #13 for why).
9. Sovereign procurement will shift from standalone capabilities to integrated, turnkey-based EO solutions
The era of piecemeal EO procurement is fading. Sovereign buyers want turnkey solutions in which sensors, satellites, data processing pipelines, and AI-enabled analytics, all in one package. Companies that can offer integrated ISR architectures will win, while those selling single capabilities will find themselves competing for subcontractor roles or left out entirely.
This shift favors large primes and well-capitalized integrators over specialized startups. For smaller EO companies, the path to government contracts increasingly runs through partnerships, not direct sales.
10. NOAA and EUMETSAT will expand commercial weather data acquisition, as global open data exchange frameworks feel the strain
Commercial weather data buys are mature: radio occultation moved from pilots to operational procurement in 2025. The next frontier is microwave sounders, currently in pilot phase. This will happen in parallel to continuing deployment of institutional weather satellites from agencies around the world.
However, as the public sector starts to rely more on commercial providers for core capabilities, the WMO's open data sharing framework (based on Resolution 40 and now the Unified Data Policy) will come under increasing pressure, potentially forcing agencies to explore alternative models to sustain the commercial weather sector.
11. Geospatial foundation models and embeddings will bring more users to EO than ever before, but operational adoption will be hard
As foundation models improve and embeddings abstract imagery away from users, the barrier to experimenting with EO drops. Expect more pilots, more demos, more proofs-of-concept. Organizations will be able to spin up an EO-based pilot for a given use case in hours/days, not weeks/months.
But operational use will be hard to follow as EO adoption is not just a technology problem, it is also a workflow integration, procurement, and trust problem. Geospatial foundation models and embeddings definitely lower the barrier to EO experimentation, but I am not fully convinced that they have reduced the barriers to EO adoption – at least not yet!
12. The tension between faster optical downlinks and on-orbit processing via space-based data centers will force EO operators to make architectural bets
Optical laser downlinks are maturing, moving data to Earth faster than ever. Meanwhile, edge computing continues to gain market traction as space-based data centers (one of the buzzwords of 2025) are emerging. I think we will see their feasibility and economics start becoming clearer in 2026.
Whether these two approaches - a) downlink faster and cheaper and b) store and process in orbit - are complementary or competing or hybrid remains to be seen, but EO operators and those that procure EO will increasingly need to make architectural bets. The wrong bet could mean stranded investment or locked-in dependencies.
13. Critical infrastructure monitoring and insurance will become the verticals with highest revenue traction for EO
Most commercial EO verticals remain stuck in the pilot trap – or worse, the project trap where one-off successes never scale. Critical infrastructure monitoring and insurance are the exceptions. These verticals have found product-market fit: recurring revenue models, quantifiable ROI, and buyers who don't need to be convinced that EO works.
In 2026, while others chase pilots, I think these will be the verticals actually growing. If you are looking for where commercial EO actually works, it is not in the flashy press releases, it is where EO is quietly running in the background, or in other words, invisible.
14. EO-based climate solutions will remain on the backburner, as the focus shifts to translating them into economic value, not climate compliance
Regulations and funding are shifting away from climate and sustainability. But the underlying capabilities won't disappear, they will just find new buyers with different motivations. Emissions monitoring will grow not because of climate mandates but because energy firms want to capture lost value. Climate risk analytics will grow not because of regulation but because financial institutions need to assess investments and price risk. Same for biodiversity. Same for forests.
The conversation has moved from compliance to commercial value. Companies still pitching EO as a climate solution might not get a lot of traction, while those reframing the same capabilities as cost savings or risk management will find buyers.
15. As EO's center of gravity shifts to dual-use, frustrated talent will scatter – some to adjacent domains, some to applied science, others out of the sector entirely
EO has historically attracted people who are both technology-loving and impact-driven. The appeal was work in organizations that focused on advanced technologies that make real-world difference. But, as dual-use becomes dominant and science (especially climate and sustainability) starts to get somewhat reduced in priority, that value proposition for EO could erode, especially for those who entered the sector to work on long-horizon science and impact.
I expect, unfortunately, some talent in EO to drift away from the sector. Some might still be in adjacent domains such as digital twin modelling, spatial intelligence and AI or get pulled into applied roles in insurance, energy, and agriculture where the impact story is clearer. But, in general, I fear this year will start to see the sector risk losing some people who built it, not to competitors, but to disillusionment.
Closing Note
These are just some bets on where the industry is heading, viewed from an independent perspective. Some will age well, some won't. I will probably revisit these at the end of 2026 and see how they held up.
The picture is not all bleak. The EO sector is maturing, finding its footing in defense and select commercial verticals, and the technology continues to advance. But 2026 will be a year of hard choices: for companies deciding where to compete, for governments balancing sovereignty and cooperation, and for professionals deciding where their skills and values fit. The geopolitical landscape remains volatile, and EO sits squarely in the middle of it.
If you think I missed something important, or see it differently, I'd like to hear it.
Until next time,
Aravind
