There is a problem at the heart of institutional natural capital investment that almost nobody talks about directly. It is not the policy environment, though that is complex. It is not the absence of institutional appetite, though that has taken time to build. It is not even the heterogeneity of the assets themselves, though that matters enormously.
The problem is intelligence. Or rather, the absence of it.
Most institutional investors are still underwriting natural capital the same way they underwrote infrastructure in 2005 — on site visits, broker relationships, and point-estimate financial models that cannot adequately capture the ecological, climatic, and regulatory dimensions of assets that will be held for decades. Diligence cycles run to 12 or 18 months not because the assets are intractable, but because the information infrastructure to underwrite them with confidence does not yet exist at institutional scale.
That is the problem we set out to solve. Not by reducing the complexity of natural capital — it is complex for good reasons — but by building the intelligence infrastructure that makes institutional-grade investment in it possible at speed and at scale.
Data as a Compounding Resource
I have been thinking a great deal, since reading Brad Smith and Carol Ann Browne’s Tools and Weapons last year, about a deceptively simple observation: data is probably the world’s most renewable resource. Unlike physical assets, it does not depreciate with use. Its value compounds through reuse, integration, and interpretation. The same dataset can be used again and again, by dozens of participants, without any loss of utility to any one of them.
In most asset classes, that observation is interesting. In natural capital, it is strategically critical.
Natural capital assets are heterogeneous, long-dated, and exposed to compounding layers of ecological, climatic, and policy risk that standard financial models are not built to handle. The variables that determine whether a restoration estate in Scotland or a peatland restoration project in Cumbria will perform over a 25-year investment horizon are not static — they evolve with rainfall patterns, regulatory frameworks, carbon market structures, and the ecological succession of the landscape itself. A point-estimate financial model cannot capture this. Only a dynamic intelligence infrastructure, continuously updated and integrated across multiple data domains, can.
The competitive advantage in natural capital will not be built on who owns the most land. It will be built on who understands it best.
This is not a theoretical observation. It is something we have been building and testing, at institutional scale, across 50,000 acres and a £400 million portfolio, for several years.
The 3-Layer Intelligence Stack
What we call Oxygen Intelligence draws on more than 100 integrated datasets — ecological, climatic, spatial, economic, and regulatory. Some of this data is publicly available. Some is proprietary. Some is generated directly through our operations on the ground. The real value, however, does not come from any individual dataset. It emerges from how these variables interact — from the connections between soil carbon sequestration rates, local hydrological systems, biodiversity indicators, policy eligibility criteria, and long-term financial projections.
The architecture that processes these connections has three layers, each solving a distinct part of the capital deployment problem.
Layer 1 — Opportunity Screening
The first layer uses AI-driven analysis across open, public, private, and proprietary datasets to identify and rank acquisition opportunities. In a market where most deal origination still depends on personal relationships and broker networks, this matters enormously. It reduces time-to-shortlist from months to weeks, removes the geography and relationship biases that distort most origination pipelines, and enables a systematic, repeatable approach to market scanning that scales independently of headcount.
Layer 2 — Risk Assessment and Diligence
The second layer is where intelligence shortens the diligence cycle most directly. Rather than underwriting against a single projected outcome, it integrates ecological, climatic, policy, and financial modelling to stress-test assets across scenario ranges. Investment committees receive a bounded picture of uncertainty rather than a false precision — which is both more honest about the nature of the asset class and more useful for making high-conviction allocation decisions under genuine uncertainty. This is the layer that allows us to move from first contact to investment committee in a timeframe that institutional allocators recognise as credible.
Layer 3 — Forward Landscape Modelling
The third layer addresses the defining challenge of long-duration natural capital investment: the future is not static. Landscapes evolve. Regulatory frameworks shift. Climate scenarios materialise differently from projections. The forward modelling layer simulates how managed estates will develop under different management interventions, climate trajectories, and policy environments — stress-testing not just the financial return but the ecological integrity and impact durability of the investment over 10 to 25-year horizons.
This layer is linked to our MRV — monitoring, reporting, and verification — infrastructure, which means that simulations are not theoretical constructs. They are grounded in, and continuously updated by, the actual ecological data flowing from our managed estates. That auditability is what converts long-duration exposure from a risk to be managed into a feature to be valued.
Speed and Scale Are Intelligence Problems
The strategic implication of all of this is straightforward, even if the operational execution is not.
In complex, long-duration markets, intelligence reduces friction. It shortens diligence cycles by giving investment committees the analytical confidence to act earlier. It strengthens risk assessment by replacing point estimates with scenario ranges. It enables earlier, higher-conviction decisions in a market where the cost of hesitation — measured in foregone returns, missed regulatory windows, and eroding pipeline quality — is real and rising.
Those with robust intelligence infrastructure will deploy capital faster, at greater scale, and with the discipline required to deliver durable returns. Those without will wait. In a market moving as quickly as this one, waiting has a cost.
This is not simply a matter of operational efficiency, though efficiency matters. It is a question of structural position. The institutions — whether asset managers, allocators, or corporate offtakers — that build or access robust intelligence infrastructure in the next 24 to 36 months will have a material information advantage over those that do not. That advantage will compound over time, in the same way that data itself compounds: through reuse, integration, and interpretation.
The analogy I keep returning to is fixed income in the 1980s, when Bloomberg built the terminal that made institutional bond markets legible at scale. Or equity risk modelling in the 1990s, when MSCI’s frameworks became the analytical standard that every allocator used. Natural capital is at an equivalent inflection point. The intelligence infrastructure layer does not yet exist as a standalone commercial product available to the market. But the foundations are being built — and the institutions that get in early will define the category.
Sharing Intelligence Without Giving Away the Edge
Oxygen Intelligence was built, first and foremost, as an internal tool — the analytical backbone of our own investment process. The PhD work in data analytics and environmental science that I brought to the business has been materially strengthened, stress-tested, and institutionalised by Chris Winter and the team he has built. What exists today is a platform designed not just for insight, but for repeatability, auditability, impact, and scale.
But we are increasingly interested in the question of how elements of that platform — tools, techniques, and potentially data products — could be made available to external partners. Not in a way that compromises the proprietary methods that underpin our own competitive position, but in a way that supports broader market development and improved decision-making across the sector.
The logic is not philanthropic. A better-functioning natural capital market is a larger market. More institutional capital deploying with higher conviction means more demand for high-integrity assets, more depth in the credit markets that underpin our revenue streams, and more regulatory support for the nature recovery activities that are central to our purpose. Intelligence, in this sense, is genuinely a renewable and shareable resource — one whose value increases, rather than diminishes, as it is more widely used.
During the course of this year, we plan to share more detail on how data informs our investment process end-to-end — from opportunity assessment and acquisition selection through to landscape management, impact monitoring, and credit verification. We are also exploring how the first external institutional partnerships around Oxygen Intelligence might be structured.
A Question for the Reader
I am conscious that the intelligence infrastructure challenge looks different depending on where you sit in the ecosystem. For an allocator building a natural capital exposure for the first time, the problem may be origination — finding credible, institutional-quality assets to underwrite. For an existing fund manager, it may be diligence speed and investment committee confidence. For a corporate sustainability team navigating BNG obligations or TNFD disclosures, it may be the verification and auditability of the credits they are buying.
What I am genuinely curious about — and what I would welcome your response to, either in the comments below or directly — is this: where do you see the biggest intelligence gaps in natural capital today? What is the specific friction in your process that better data infrastructure would most directly address?
The answer shapes what we build, and what we open up. It also shapes how this market develops. I think the conversation is worth having openly.