Terrascope Cuts Clients' Emissions Data Processing Time by 10 Hours with AI-Powered Applications Built on Amazon Bedrock

As a provider of decarbonization solutions, Terrascope saw opportunities where AI technologies could enhance its platform to better support its enterprise customers in their sustainability journeys.

One key opportunity involved streamlining the tracking and measuring of Scope 3 emissions data. Scope 3 emissions, or indirect emissions from a company’s value chain, typically account for over 70 percent of an organization’s total carbon footprint, yet they remain challenging to measure due to their complexity and involvement of numerous external stakeholders. By developing tools to better handle diverse supplier data formats and standardize information across different activities—including purchased goods, transportation, and waste disposal—Terrascope could reduce the time required for emissions analysis by over 90% while also improving data consistency and reducing error rates by over 50%.

Terrascope also saw potential to identify high quality decarbonization initiatives that took into consideration not only decarbonization potential itself, but also financial ROI, client maturity, and the robustness of scientific literature supporting any given decarbonization measure. Previously, Terrascope’s experts had to manually research each customer’s circumstances against an internal library of over 20,000 potential initiatives to contextualize the recommended initiative for the customer, which could take up to several days to complete. By creating technology to automate and enhance this matching process, Terrascope could deliver customized recommendations more efficiently while providing actionable insights about implementation approaches and potential challenges.

By automating complex analytical processes related to Scope 3 emissions and generating personalized guidance, the company would be able to optimize costs, strengthen business resilience, and scale its impact in a rapidly evolving regulatory landscape.

Terrascope utilized Amazon Bedrock, a fully managed service that builds and scales generative AI applications with foundation models, to build two AI-powered applications: a Scope 3 carbon footprint analytical tool and an AI-based decarbonization advisor. The analytical tool reduced 10 hours of data processing time per customer and error rates by about 50 percent. Meanwhile, the AI-based decarbonization advisor helps Terrascope’s customers identify and prioritize decarbonization initiatives, which helps businesses to accelerate their net-zero journeys with tailored, evidence-based recommendations.

For Scope 3 emissions analysis, Terrascope’s AI automatically organizes a client’s products into clear categories that follow global standards. With Amazon Bedrock’s Retrieval Augmented Generation capability and Amazon OpenSearch Service, a service for running and scaling OpenSearch clusters, Terrascope translates complicated business activities into relevant emission factors so emissions can be measured accurately. In addition, the AI uses Amazon Textract, a machine learning (ML) service, to automatically extract text, handwriting, layout elements, and data from scanned documents provided by clients, such as invoices, purchase orders, and shipping receipts. Terrascope now accomplishes in hours what once took weeks of manual work, with improved consistency and accuracy. By accelerating the process, Terrascope can serve more clients and improve the precision of initial Scope 3 carbon footprint approximations.

Meanwhile, Terrascope’s new AI-based decarbonization advisor provides accurate recommendations to customers. The application uses Amazon Bedrock to analyze business needs, then uses Amazon Bedrock Knowledge Bases, a fully managed end-to-end RAG workflow that natively integrates Amazon OpenSearch Service, to recommend the most relevant decarbonization actions from a library of initiatives, such as energy efficiency upgrades and switching to renewable energy, all under one minute. It also pulls in the latest public research and industry updates in near real time, making sure that recommendations are always current, comprehensive, and backed by evidence.

AWS Professional Services also assisted Terrascope throughout the development process, from initial requirements gathering and architectural design to implementation and iterative feedback. AWS experts helped Terrascope deploy generative AI use cases by advising on the optimal architecture for handling unstructured and complex emissions data, leading to scalability and cost efficiency.

Terrascope is committed to further innovating its AI-powered platform, with ongoing investment in advanced data processing, automation, and responsible AI practices. The company will enhance the platform’s ability to handle unstructured data, improving the accuracy and speed of emissions measurement, and scaling personalized decarbonization recommendations. The company is also prioritizing data privacy, IP protection, and compliance with evolving global standards.

Terrascope will continue working with AWS to optimize its sustainable tech stack and platform architecture, making sure that its solutions remain scalable, secure, and energy efficient as customer demand grows. It aims to extend its market presence, particularly in Australia and North America and other key regions, building on recent expansions and partnerships with industry leaders.