In recent years, the role of technologies in the transition toward sustainable development models has become central. The focus on circular economy, renewable energy, and responsible business practices is no longer just a strategic choice but a regulatory requirement, especially in light of European directives such as the CSRD (Corporate Sustainability Reporting Directive).
In this context, companies, institutions, and research organizations are called to innovate responsibly, aligning their efforts with the Sustainable Development Goals (SDGs) and ESG (Environmental, Social, Governance) criteria. But how can we truly understand which technologies are contributing to sustainability and complying with these standards?
The answer is closer than one might think: patents.
Patents are not only legal tools for protecting innovation; they also represent an enormous source of technical knowledge. They describe solutions to concrete problems—often feasible and already tested in the past, sometimes buried under decades of inactivity. Classifying patents according to their relevance to ESG criteria means making this wealth of information accessible, usable, and applicable to accelerate green, social, and governance transitions.
Why Classify Patents through ESG and SDG Lenses
The volume of patent data is immense: over 150 million documents worldwide. Manually searching for all relevant patents such as those related to climate change—is practically impossible without specialized expertise. A predefined ESG (or SDG) classification acts as a ready-made “map,” drastically reducing research time and enabling faster action.
The impact of this activity is significant: already invented technologies can be “rediscovered” and applied to current challenges, avoiding the need to start from scratch; companies and institutions can identify groups of technologies to collaborate with or promote for the common good, fostering collaboration and technology transfer; finally, classification allows for monitoring innovation trends, identifying where research efforts are concentrated, which countries or industrial players are emerging as leaders, and which areas remain overlooked.
As an example, consider CCUS (Carbon Capture, Utilization, and Storage) technologies: identifying the most relevant patents in this field makes it possible to understand which approaches are more mature, which countries and companies are investing the most, and where research gaps remain. This patent mapping is crucial both for policymakers designing decarbonization strategies and for companies investing in CO₂ reduction projects, as it supports incentive planning, market regulation, and targeted resource allocation—leading to more informed decisions that accelerate the sustainable transition.
Discover our AI software for the automatic extraction of information from technical documents.
From Greenwashing to Transparency
Another advantage lies in assessing the true sustainability level of a technology, helping fight against so-called “ESG washing.” Today, many technologies are presented as “green” or “sustainable” without truly being so. Accurate patent classification helps uncover these inconsistencies: if a supposedly sustainable innovation does not appear in any ESG category, it raises a red flag that warrants further investigation.
This transparency activity aligns perfectly with regulatory requirements such as the SFDR (Sustainable Finance Disclosure Regulation), which compels financial actors to disclose the environmental and social impact of their investments. Having a solid patent classification framework enables these disclosures to be backed by objective data, reducing the risk of misleading communication and greenwashing-related disputes while validating virtuous practices.
Why a New Approach Is Needed
At present, some ad hoc classification schemes exist for labeling patents relevant to sustainability. However, the task remains challenging due to its complexity and the need for constant updates. The approach described in this article aims to contribute to the field through accurate classification methods that deliver precise and reliable results, within a dynamic system able to evolve alongside emerging technologies.
Specifically, to address this complexity, ESG patent classification can be supported by a Natural Language Processing (NLP)-based pipeline. The procedure involves building a database of ESG concepts through Erre Quadro’s proprietary algorithms, on which ranking algorithms and queries are then applied to classify patents. In this way, each patent is characterized as relevant—or not—to each SDG objective, while also allowing for the possibility that a single technology may impact multiple SDG intervention areas. In parallel, another classification is performed in relation to broader categories connected to the three ESG pillars.
The methodology can be further enhanced through continual learning procedures, dynamically updating search and classification criteria in light of new emerging technologies and progressively improving the system’s accuracy.
Toward More Informed Policy Making
ESG classification of patent innovation is not an academic exercise but an operational tool. It can empower policymakers to identify technological priorities for investment, assess the real impact of innovations, promote international cooperation on key issues, and measure progress toward the SDGs. In a world where ecological and social transitions can no longer be delayed, equipping decision-makers with tools based on objective data is essential: patents, when properly analyzed and classified, become a true compass to guide more effective policies and more responsible industrial strategies.
Request a demo of our AI tools.
