Solution Architect Dr. Michael Moore, Ph.D. outlines how graph tech supports financial institutions working to meet Scope 3 carbon reporting requirements.
Under the Greenhouse Gas (GHG) Protocol, financial services institutions will soon be tasked to estimate and report their carbon emissions, not just ‘Scope 1 and 2’ activities, but new Scope 3 activities that extend well beyond your company walls.
Scope 1 and 2 include emissions you make directly (for example, emissions from your facilities) and indirectly (such as emissions from power consumed by your data centers). That’s onerous enough, but on the heels of Scope 1 and 2 reporting requirements are the newer Scope 3 disclosures. Scope 3 encompasses indirect emissions that occur across a company’s entire value chain. Financial services companies typically have emissions associated with business travel, employee commuting, purchased goods and services, and customer-facing operations, among others. Financial services companies may also have significant Scope 3 emissions associated with their investment, underwriting and lending portfolios.
The Value Chain challenge
Scope 3 requirements are harder to monitor, as they involve factors outside of a company’s direct control or oversight—but as Oxford University has said, it is the set of environmental actions most visible to the consumer and therefore important to get right.
Scope 3 requirements are becoming a focus of attention for publicly traded companies in the US because the US Securities and Exchange Commission (SEC) has set a fiscal year 2024 due date for initial disclosures. Scope 3 emissions can be challenging to measure and report due to the complexity of value chains and data availability. Companies may need to collect data from various sources, estimate emissions when data is not readily available, and develop robust methodologies for accurate reporting. It’s no wonder that Scope 3 is being called the new Sarbanes-Oxley.
Financial services institutions are right to be concerned. Unifying this level of data from so many inputs is a huge undertaking and almost impossible to achieve using traditional SQL or NoSQL database management systems. That’s because traditional SQL or NoSQL approaches are so inefficient at modelling and querying hierarchical and deeply connected data. So is there a better approach?
Graph-based digital twins
There is, and it’s in the shape of graph technology using digital twins. Leading organisations are building digital twins of their value chain, backed by a modern native graph database that can scale easily. How does this help with compliance burdens? Digital twins are virtual representations of physical reality, and if done right can be as extensive and detailed as the reality they aim to capture.
They offer the only feasible way of modelling huge real-world and real-time systems with the appropriate level of fidelity to support the advanced analytics and decision-making needed for gaining visibility into carbon emissions across extended value chains. They provide a comprehensive, often real-time, digital view of the business processes they represent, and can serve as the diagnostic foundation for setting–and meeting–new lower emission targets.
The best data backbone for a digital twin is a graph database, which can elegantly and accurately represent the full complexity of the dense networks of connected elements commonly found in the value chain digital twin.
Graph technology provides a flexible, performant data foundation for analysing complex digital twin data. Graphs treat the relationships between data points as being just as important as the data points themselves. In a graph database, the relationships between data points are stored in-memory and on disk, creating a fabric of fully-connected data that can be queried across multiple hops with millisecond response times. Modern native graph databases can easily handle billions of relationships, which could make all the difference in meeting the Scope 3 challenge.
A graph-based digital twin can tie together networks of connected assets with emissions data from a variety of firm-wide and portfolio sources. Graph queries can quickly determine everything from non-compliance, critical dependencies, optimal routings, infrastructure alerts, poor data quality, and even the root causes of excess emissions. Data scientists can run powerful graph algorithms on digital twins to determine potential points of failure, node similarities, lowest carbon pathways, connected communities, and other hidden data patterns that would otherwise be difficult to identify.
Start with one use case–don’t boil the ocean!
Graph-based digital twins connect emissions data in a way that would be hard, or even impossible, to do in non-graph (i.e., relational or NoSQL) systems. With a graph-based digital twin you’ll have a fully contextualised view of every entity in the value chain and, as a result, its contribution to your organisation’s total carbon emissions. You can start with one simple use case and fold in additional data and adjacent use cases as your team’s graph expertise matures and your GHG estimation methodologies become more sophisticated. Developers who know SQL can quickly learn the graph data platform and its popular query language, Cypher.
Research has shown that the finance sector’s funded emissions are over 700 times greater than its own direct emissions. Building a digital twin of the value chain using a graph database is your best bet for tackling the complexities of Scope 3 reporting.
The author is Principal, Partner Solutions and Technology at graph database and analytics leader Neo4j
Subscribe to our newsletter
