In asset management, we often talk about the “plumbing” of finance. As an Enterprise Architect, my role is to design, connect, and fortify that plumbing. For decades, investment firms have invested heavily in systems of record for transactions, portfolio accounting, market data, performance, and financial risk. Yet the transition toward a net-zero economy is exposing a new architectural challenge: climate data does not naturally fit into the traditional investment technology landscape.
Climate data is fragmented, estimated, frequently revised, and highly dependent on methodology. It spans corporate disclosures, external data providers, satellite observations, physical climate models, transition scenarios, regulatory datasets, and proprietary research. More importantly, climate risk is not confined to one investment function. It affects portfolio construction, risk management, research, stewardship, compliance, reporting, and ultimately the valuation of assets.
When an asset manager commits to frameworks such as the Net Zero Investment Framework or aligns portfolios with Paris-aligned investment objectives, the commitment is therefore much more than an investment-policy statement. It becomes an enterprise architecture problem. The organisation must translate climate ambition into data, decisions, workflows, controls, and measurable outcomes.
The first architectural priority is to establish climate data as an enterprise capability rather than an ESG data silo. A unified Climate Data Engine should ingest, normalise, enrich, govern, and distribute climate information across the investment platform.
This is challenging because climate metrics lack the consistency traditionally associated with financial data. A security has an ISIN, price, currency, issuer, and accounting attributes that fit established data models. Climate information is different. Emissions may be reported, estimated, modelled, restated, or unavailable. Different providers may apply different methodologies to the same issuer. Scope 3 emissions can vary dramatically depending on boundaries and estimation techniques. Even apparently straightforward measures such as carbon intensity require multiple underlying datasets.
The architecture therefore needs to support both absolute and intensity-based measures. Absolute emissions help investors understand the scale of an issuer’s contribution to greenhouse-gas emissions, while intensity metrics such as Weighted Average Carbon Intensity help compare portfolios and companies across different economic scales.
The data model should also support financed-emissions methodologies, including the use of Enterprise Value Including Cash where appropriate, so that emissions can be attributed to investors consistently across asset classes. This becomes particularly important when the objective is to understand the financed emissions associated with an entire portfolio rather than simply ranking securities by emissions intensity.
However, historical emissions alone are insufficient for investment decision-making. Climate risk is inherently forward-looking. An enterprise climate architecture therefore needs to connect historical carbon data with transition plans, technology assumptions, policy scenarios, physical climate projections, and financial models.
This creates an important distinction between measuring a portfolio’s carbon footprint and understanding its climate risk. A portfolio can have a relatively low current carbon intensity while remaining highly exposed to future transition risk. Conversely, a company with significant current emissions may have a credible transition strategy, substantial investment in low-carbon technologies, and a business model that could benefit from the transition.
The architecture must therefore accommodate forward-looking measures such as transition-risk indicators, Climate Value-at-Risk, scenario analysis, and implied temperature or alignment metrics. The objective is not to create another dashboard. It is to create a common analytical foundation that allows portfolio managers, risk teams, and stewards to reason about climate risk using consistent and traceable information.
The principle is simple: climate data should become a first-class citizen of the investment data architecture. Its lineage, provenance, quality, methodology, versioning, and auditability should be governed with the same seriousness applied to prices, positions, ratings, and other investment-critical data.
The second architectural challenge sits much closer to the front office. Once climate data becomes available, how should it influence portfolio decisions?
This is where asset managers encounter what I would call the “paper decarbonisation” paradox. Portfolio carbon intensity can often be reduced rapidly by selling or underweighting the highest-emitting companies. From a portfolio measurement perspective, the result can look impressive. But the underlying economic activity may not have changed at all. The emissions have simply moved from one owner to another.
This distinction is critical. Portfolio decarbonisation and real-world decarbonisation are related, but they are not the same thing.
A sophisticated investment architecture should therefore distinguish between changes caused by portfolio positioning and changes caused by underlying issuer behaviour. Emissions attribution becomes essential. When portfolio carbon intensity falls, the investment platform should be able to answer a fundamental question: did the reduction occur because companies actually reduced their emissions, or because the portfolio changed its holdings?
That capability changes the conversation from “How much did we decarbonise the portfolio?” to “How much real-world transition did our capital contribute to?”
The front-office architecture must also integrate climate constraints with traditional portfolio construction. Carbon targets do not exist in isolation from tracking error, liquidity, factor exposures, sector allocations, investment guidelines, risk budgets, and expected returns. A carbon optimisation engine operating independently from the portfolio risk model can produce theoretically attractive but commercially impractical portfolios.
The better architecture is therefore one in which climate objectives become additional dimensions within the portfolio decision framework. The portfolio manager should be able to understand the trade-offs between emissions reduction, financial risk, valuation, liquidity, sector exposure, and portfolio objectives within a common optimisation environment.
This is where enterprise architecture can make a significant difference. Instead of creating a separate “green portfolio” technology stack, climate considerations should become part of the same decision infrastructure used for mainstream investment management.
But portfolio construction is only one side of the equation. If the objective is to influence real-world transition, ownership creates another powerful mechanism: stewardship.
Stewardship should not be treated as a collection of meetings, letters, voting decisions, and manually maintained spreadsheets. It should be treated as a structured investment workflow.
I would architect stewardship as an Engagement CRM: a system of record for the relationship between an asset manager and the companies in which it invests. The starting point is the identification of companies where engagement could materially influence climate outcomes or protect long-term shareholder value.
The workflow then moves through objective setting, counterparty mapping, engagement, milestone tracking, escalation, and ultimately an assessment of whether the engagement has achieved its intended outcome.
This creates an important shift in mindset. Engagement is no longer an activity that produces meeting notes. It becomes a measurable investment process.
For example, an engagement objective might require a company to establish credible emissions targets, improve climate governance, disclose material Scope 3 emissions, align capital expenditure with its transition strategy, or demonstrate progress against defined transition milestones. Each objective can be represented as a structured data object with an owner, target date, evidence requirements, status, and escalation criteria.
The resulting architecture creates a feedback loop between investment research, portfolio management, and stewardship. A portfolio manager can see that a holding is subject to an active engagement programme. The stewardship team can understand the portfolio significance of the company. Risk teams can incorporate transition indicators into their assessments. Voting decisions can be linked to previous engagement commitments. Compliance and sustainability teams can trace the evidence supporting external disclosures.
This is particularly important as stewardship reporting becomes increasingly demanding. An asset manager should be able to reconstruct the complete chain of reasoning behind an engagement or voting decision: why the company was selected, what objective was established, what conversations occurred, what progress was made, what evidence was received, and why escalation occurred.
The architecture should therefore create an auditable relationship between engagement milestones and voting actions. A vote against management should not appear as an isolated event in a proxy-voting system. Where appropriate, it should be connected to the history of engagement and the company’s progress against previously communicated expectations.
The same principle applies to investment research. Climate stewardship becomes much more powerful when the engagement workflow can connect with financial information. Annual reports, financial statements, capital expenditure plans, transition strategies, governance structures, and relevant audit disclosures can provide evidence for assessing whether climate commitments are translating into economically credible action.
Ultimately, this creates something more valuable than an ESG reporting platform. It creates a climate-aware investment operating model.
The architectural pattern is therefore broader than a Climate Data Engine or an Engagement CRM individually. It is an integrated climate investment architecture connecting four capabilities: climate data, portfolio decision-making, stewardship, and enterprise governance.
At the centre is a governed climate data layer. Above it sit analytical services that transform raw climate information into portfolio-level metrics, scenarios, attribution, and risk measures. These services feed portfolio construction and risk systems. In parallel, the same information identifies stewardship candidates and informs engagement objectives. Engagement outcomes then flow back into investment research, risk assessment, voting decisions, and portfolio construction.
This feedback loop is crucial because climate risk is dynamic. A company’s transition trajectory changes. Regulation changes. Technology costs change. Physical risks evolve. Capital expenditure changes. Management commitments can strengthen or deteriorate. A static annual ESG score cannot capture this complexity.
The enterprise architecture therefore needs to support continuous learning. Climate data should be refreshed. Models should be recalibrated. Engagement outcomes should update investment views. Portfolio decisions should generate measurable attribution. Those results should feed back into the next investment decision.
This is ultimately the difference between climate reporting and climate stewardship.
Reporting asks whether the organisation can measure and disclose its climate exposure. Stewardship asks whether the organisation is using its capital, influence, information, and ownership rights to manage that exposure and influence the transition.
For asset managers, the strategic opportunity is significant. The firms that treat climate as a standalone sustainability problem will continue to struggle with fragmented data, inconsistent metrics, manual processes, and disconnected decision-making. The firms that treat climate as an enterprise architecture problem can embed it directly into the investment operating model.
The goal should not be to build another ESG silo. It should be to remove the silo altogether.
A modern asset manager should be able to trace a climate signal from its original source, through data governance and analytics, into a portfolio decision, through an engagement programme, into a voting action, and ultimately back to an assessment of investment and real-world outcomes.
That is what it means to de-silo the carbon ledger.
Climate risk is systemic, and systemic problems require systemic architecture. The competitive advantage will belong not simply to firms with better climate data, but to firms capable of turning that data into better investment decisions, more effective stewardship, stronger governance, and measurable real-world outcomes.
For the Enterprise Architect, the mandate is therefore clear: build the infrastructure that makes climate stewardship part of how the investment organisation works—not an additional process that sits beside it.