Product Carbon Footprints: from reporting artefacts to collaborative infrastructure for capital allocation and decarbonisation

Why Carbon Reporting Is Failing the Food System and What Comes Next

The global food system sits at the centre of the climate challenge. It accounts for roughly a third of global greenhouse gas emissions and close to 80% of global land-use conversion, including deforestation.

Up until now, efficiently allocating finance to demonstrable decarbonisation projects has been challenging for businesses looking to support supply chain emissions reduction.

Overall, the understanding of actual emissions from food remains one of the most opaque, uncoordinated, and under financed systems when it comes to effective climate action.

Despite rapid progress in climate reporting, targets, and frameworks, we are still failing to answer the most important question:

Where, exactly, should capital go to reduce emissions and who should act?

The uncomfortable truth is that today’s carbon accounting infrastructure was never designed to answer this question. As a result, much of what passes for climate action in food systems is built on averages, proxies, and narratives rather than on real, system-specific data. This failure is felt most acutely in Scope 3 Category 1: Purchased Goods & Services, the largest emissions source for almost every food and beverage company that have verified science-based targets as of 2026.

The Limits of Today’s Carbon Footprinting

Most carbon footprinting platforms are designed for scale, not accuracy. Until the mid-2020s, this was necessary and valuable: secondary databases and average emission factors enabled businesses to gain an initial view of climate risk and opportunity. These datasets have been fundamental to the development of carbon accounting in food, and that first step should not be undervalued.

However, this approach falls short when applied to complex, heterogeneous supply chains and specific products. Meta-analysed averages flatten reality. They obscure differences between producers, practices, and processes, precisely where emissions are created and where reductions are possible.

Distributed across hundreds or thousands of suppliers, Scope 3 is multi-layered, difficult to verify, and often internally contested. Corporate-level averages hide more than they reveal, while product-level impacts are reduced to generic assumptions that offer little direction on where change should occur over the long term.

The result is a system that can report emissions but cannot reduce them. Companies are left guessing. Suppliers are inundated with extractive questionnaires that are rarely repeatable or editable. Capital is deployed into offsets or broad initiatives disconnected from real hotspots. And everyone, to some extent, hides behind the narrative that “the data isn’t good enough.”

Why Product-Level Reality Matters

Decarbonisation does not happen at the corporate level. It happens in fields, factories, logistics routes, energy systems, and material choices. It happens at the product and process level.

An average emission factor may indicate that an ingredient or product, let’s say onions, from a given region are emissions-intensive, but it cannot tell you how to reduce these intensive practices. Is the impact driven by synthetic fertiliser use? Mechanical fuel use for on farm equipment? Soil carbon loss from a lack of cover cropping? Irrigation energy? Land management practices? Without that clarity, capital cannot be allocated efficiently, and effective collaboration across the value chain becomes impossible.

This is why Scope 3 reductions remain elusive - not because businesses lack intent, but because the system is structurally incapable of showing them where to act with confidence.

This limitation is further exposed by the widespread reliance on a single emissions intensity (CO₂e/kg) figure for a product carbon footprint. While useful for high-level reporting and comparison, a single number offers almost no practical guidance to the actors expected to reduce emissions within a product’s lifecycle.

A useful analogy is consumer health. Someone trying to build muscle often does not rely on a single “health score” printed on a product. They depend on ingredient lists and nutritional breakdowns - protein content, sugar levels, fat types, portion size - to make informed, goal-oriented decisions. Without that detail, progress is impossible. It also isn’t assumed that every product in a given category has the same nutritional value, many products are fortified or processed in different ways.

Carbon works the same way. A single emissions intensity number tells you a product has impact, but not how to reduce it.

Sustainable finance faces the same problem: without visibility into the underlying components of emissions, capital cannot flow efficiently to the interventions that deliver the fastest and most durable reductions.

The Competitive Edge: Data Sovereignty

At the same time, a deeper structural shift is underway, one that extends beyond carbon accounting.

As generalist AI systems rapidly commoditise undifferentiated services and information, data ownership and sovereignty are becoming core sources of competitive advantage. Businesses are realising that their most valuable asset is not generic insights, but primary, system-specific information: how they operate, how their supply chains function, and how their products perform in the real world against climate risks and opportunities.

Carbon data sits squarely within this category.

Yet many sustainability platforms still treat supplier data as something to be extracted, standardised, and absorbed into opaque systems solely for reporting purposes. This undermines trust, discourages participation, and strips suppliers of control and value.

In a future shaped by AI, this approach is not just flawed, it is obsolete.

Reframing Product Carbon Footprints as Collaboration Infrastructure

What if Product Carbon Footprints (PCFs) were not compliance artefacts, but collaborative, representative systems that enabled multiple users to coordinate and contribute?

Instead of buyers extracting data from suppliers, imagine a product-centric platform where each actor in the value chain contributes, owns, and controls their part of the emissions data. PCFs are built progressively as data is added over time, evolving from estimates into genuinely supplier-specific representations of reality.

In this model, the PCF becomes a shared reference point: a living map of the value chain that highlights real emissions hotspots and turns them into actionable opportunities.

Hotspots are no longer abstract numbers or educated guesses. They become signals, invitations for collaboration. The goal is no longer about reducing impact from an average starting point, but getting genuine access to the specific opportunities and risks associated with products and product portfolios that the business proudly owns.

The food sector needs to reach a point of data accuracy and specificity on products before it can make any material claims on reductions in corporate reports. So long as emission reductions are driven by reporting for the sake of reporting, it will remain opaque and untrusted. Rather, it needs to be treated as part of the product design itself.

From Reporting to Capital Allocation

Once hotspots are visible and trusted, everything changes. Solution providers, whether in energy, agriculture or logistics can efficiently engage directly with the value chain, offering quantified reduction options and cost estimates tied to real data. Businesses can see where emissions sit, and what it would cost to reduce them, revealing the long-term impact those investments would deliver.

Crucially, this enables targeted insetting: capital flowing directly into supply-chain interventions that reduce emissions at source, rather than into offsets disconnected from operations. Manufacturers can fund upstream changes. Retailers can co-invest with suppliers. Progress becomes visible, auditable, and shared through enhanced product and supply-chain value.

In this world, reporting is no longer the goal of sustainability or ESG. It is a by-product of action.

Why This Matters Now

The food system is approaching a breaking point. Climate volatility, input price shocks, regulatory pressure, and scrutiny of greenwashing have converged over the past several years. At the same time, AI is compressing margins in undifferentiated businesses and rewarding those that enable coordination, trust, and system-level intelligence.

The platforms that will matter over the next decade are not those that generate more dashboards from extractive approaches to generic data. They are those that help actors work together to solve problems that cannot be solved alone and positively expose product-level action plans.

The Opportunity Ahead

A product-centric, sovereign, collaborative PCF system does exactly that. It aligns incentives across the value chain. It protects sensitive data while enabling shared understanding. It directs capital to where it can do the most good, whether you are an investor allocating funding, a farmer managing risk and opportunity, or a manufacturer articulating product-level value and long-term resilience.

Fundamentally, it transforms climate action from a reporting exercise into a system-level operating model.

The food sector can no longer rely on marginal improvements. It needs a new form of infrastructure, one that connects data, decision-making, and capital in a way today’s systems cannot.

In today’s information abundant world, trust remains scarce. AI doesn’t solve for this specific risk. It can replicate almost anything except genuine collaboration, the ability to coordinate climate action across supply chains for specific products and brands may become one of the most defensible sources of value a food business can have.