Fifteen years into the cloud revolution, it’s clear that embracing technologies such as AI and machine learning (ML) has unlocked substantial value.1 These technologies, which can generally be run faster and more effectively through the cloud, have not only led to cost efficiencies and operational savings but also optimized resource use.2
Until recently, however, the tremendous potential of cloud-powered technologies to aid companies in achieving their decarbonization goals has received much less attention. The cloud offers organizations and individuals virtually limitless computing, storage, networking capabilities, and advanced software applications, with generative AI (gen AI) solutions becoming increasingly prevalent. Running these tools on the cloud enables companies to acquire new decarbonization-related capabilities that might previously have been too expensive or time-consuming to build on premise.
Our research suggests that using cloud-powered technologies can accelerate the implementation of 101—or 47 percent—of the 217 representative decarbonization initiatives that are required to achieve the global 1.5° pathway by 2050 (see sidebar “About the research”).3 Given previous estimates that achieving net-zero emissions by 2050 would require spending $9.2 trillion on decarbonization per year,4 the cloud’s potential contribution through these technologies could be worth hundreds of billions of dollars annually.
According to our estimates, the climate benefits could also be significant. In addition to accelerating decarbonization initiatives, cloud-powered technologies can play a role in abating up to 32 metric gigatons of CO2 equivalent (GtCO2e)—nearly half of the total 65 GtCO2e that we estimate is required to reach net-zero emissions by 2050. For the subset of initiatives in which the cloud can play a significant role, we calculate that each use of the cloud to power key technologies can reduce the cost of implementing a decarbonization initiative by 2 to 10 percent. On aggregate, we estimate that the total benefit of using the cloud to accelerate decarbonization could be up to 1.5 GtCO2e per year by 2050. The manufacturing and transportation sectors stand to benefit the most from making full use of these key technologies.
In this article, we’ll discuss the considerable potential the cloud has to enhance the sustainability transition by expediting the adoption of decarbonization strategies. We’ll explore the impact of three predominantly cloud-powered technologies—AI, ML, and the Internet of Things (IoT)—in accelerating decarbonization initiatives. Additionally, we’ll share why manufacturing and transportation stand to benefit most from these technologies, looking more closely at automotive manufacturing. We’ll also examine the ways cloud-powered technologies can assist with regulatory and compliance rules and lay out some initial steps for companies to get started using these technologies to achieve their sustainability goals.
Accelerating the most critical decarbonization initiatives using cloud-powered technologies
Many companies already understand the cloud’s potential to decrease carbon emissions in IT. Migrating applications to the cloud and shutting down data centers can significantly reduce IT carbon emissions because cloud service providers tend to run ultraefficient data centers on renewable energy.5 Ensuring ultraefficient data centers are used in the infrastructure mix can represent a significant reduction in IT emissions.
However, the cloud’s role in enabling the full set of actions required to get to net zero is less understood. We selected 217 initiatives from the 1,200 levers currently available in the McKinsey Decarbonization Lever Library and found that using the cloud to enable key technologies can play a significant role in accelerating 101 initiatives by reducing lever-implementation costs by 2 to 10 percent (Exhibit 1). Using the cloud to enable key technologies can also provide a lesser benefit (reducing cost by less than 2 percent) on an additional 82 initiatives.
In addition, with increasingly stringent climate-related regulations coming into effect each year, cloud-powered technologies can also play an important role in facilitating, accelerating, and decreasing the cost of target setting, reporting, and compliance. At the root of this benefit is the potential to enhance data observability—an organization’s understanding of the location and quality of data within and outside their systems—and enable data exchanges between competitors and across industries, which will become increasingly necessary to address Scope 3 emissions.
The potential of AI, the IoT, and ML in decarbonization
Across initiatives, the following three cloud-powered technologies could, when combined, play an important role in decarbonization:
AI and data exchange for Scope 3 transparency. Compiling and consuming large data sets is faster and cheaper with cloud-based data lake technology, including third-party APIs, queries, documents, and databases. For example, decentralized data exchanges throughout the supply chain can enhance a company’s understanding and transparency of its Scope 3 emissions data. This allows the company to uncover new and more cost-effective pathways for decarbonization. Notably, in the logistics sector—in which, according to our analysis, Scope 3 emissions contribute up to 80 percent of total emissions—we estimate that cloud-enabled data observability and exchanges across intricate supply chains could reduce the time to enable an actionable decarbonization strategy from six to eight weeks to as little as one week. Additionally, these data exchanges can help identify emission-reduction opportunities worth as much as 40 percent of a company’s emissions through detailed Scope 3 visibility. According to our analysis, this refined data exchange procedure has the potential to save companies approximately 80 percent of the time typically needed for data collection, cleansing, and estimation.
Physical-asset transition based on IoT. Digital-twin technology can make transitioning a business’s physical assets (such as machinery, equipment, and facilities) significantly more efficient by allowing companies to upgrade and optimize—or, in some cases, transition to—a digital or automated environment. By employing IoT sensors to capture data from physical objects (goods and raw materials) and convert it into digital information (bits and bytes) as well as share data between assets, a digital twin can serve as a powerful tool for transforming assets with lower emissions. With $4.5 trillion of spending expected by 2050 for adopting or retrofitting to low-emission physical assets,6 the cost and time required for the asset transition will be critical metrics of success. Historically, our research has found that with digital twin–based diagnoses, companies can typically expect a reduction of more than 10 percent in transition lead time for low-emission assets. In addition, a digital twin can bring a more comprehensive, real-time understanding of how companies’ physical assets are used and how they perform, such as their energy and material consumption. With digital-twin technology, companies can reduce their energy consumption by about 10 percent on average.
Leveraging high-performance computing for ML resource optimization, including material, energy, and labor. ML models with virtually unlimited on-demand access to high-performance computing of the cloud can support the creation and execution of complex simulation models. These models play a crucial role in finding a balance between cost and carbon emissions. Critical initiatives with this dual focus include redesigning products, optimizing delivery and charging routes for electric vehicle (EV)–based logistics (where to go and when to charge), and planning for the energy transition (including, for example, suggesting where infrastructure such as EV chargers should be built). McKinsey has found that redesigning products, for example, can reduce costs by 5 to 15 percent and Scope 3 emissions for purchased goods and services by more than 25 percent.7
Manufacturing and transportation benefit most from cloud-powered technologies
The impact of cloud-powered technologies on decarbonization efforts will vary by industry, depending on the size of the industry’s carbon footprint and the potential for reducing it. Our research suggests that manufacturing and transportation are currently the industries that can benefit most from cloud-powered technologies for decarbonization. This is in part because of their size: the transportation industry, for example, emits nearly a quarter of total global greenhouse-gas (GHG) emissions, according to McKinsey analysis, which means that the rate of decarbonization in this sector has a significant impact on overall decarbonization efforts.8
Manufacturing. More than 210 of the 455 cloud use cases we have identified are relevant to manufacturing (see sidebar “Zooming in on automotive manufacturing”). Notable examples include using digital-twin technologies on the manufacturing line, employing real-time analytics for predictive maintenance, and creating a simulator that considers both the financial costs and the carbon emissions associated with product design (Exhibit 2). Companies could also achieve cost savings from increased productivity and reduced energy consumption, which will help accelerate the industry’s sustainability transition. For example, according to our analysis, using cloud-powered technologies to adjust biomass-related processes based on the composition of raw material can improve yields by up to 5 percent.
Transportation. There are more than 50 decarbonization initiatives relevant to the transportation sector from among the 455 use cases identified, and they could play a role in abating up to 2.8 GtCO2e by 2050. Top initiatives include optimizing loads and routes, charging networks, and real-time warehouses. The acceleration will mostly come from lowering costs, reducing inventory, and reacting more quickly to changing demands in inventory management. Scope 3 visibility, one of the most crucial problems in transportation and logistics decarbonization, can also be significantly enhanced with cloud-powered technologies. For example, the Smart Freight Centre Exchange Network demonstrates how these technologies can enable decentralized, trust-based data exchanges in commercially sensitive situations, enabling all participants to get granular visibility into their Scope 3 emissions.9
Facilitating compliance and promoting accountability with cloud-powered technologies
In addition to playing a direct role in accelerating the implementation of decarbonization initiatives, cloud-powered technologies can help companies navigate the evolving environment of sustainability targets and regulations.
Using cloud-powered technologies can enable the use of multiple services for retrieving internal data (such as customer-relationship-management and enterprise-resource-planning tools) and external sources (such as press releases) that can facilitate compliance and reporting. Where primary data sources are not currently available, AI and ML technologies can suggest secondary sources and generate estimations. Gen AI could do even more, with modules that can generate responses to regulators’ questions by crawling through data points and creating a narrative from them. Companies can also use the modules to do advanced scenario planning and test decarbonization strategies.
In addition to assisting with reporting and compliance, cloud-powered technologies can establish realistic sustainability targets that are grounded in important internal and external data, as we have already seen in the automotive manufacturing sidebar. Reliable target setting will help companies avoid overshooting and greenwashing claims.
For these reasons, cloud-powered technologies will be an important accelerator of reporting and compliance in relation to regulatory developments such as the European Union’s Corporate Sustainability Reporting Directive (CSRD). CSRD requirements are complex and cut across multiple business units (see sidebar “What the European Union’s CSRD means for ESG reporting”). Our preliminary internal analysis indicates that once a CSRD tool is in place, the end-to-end process for data connection and output review can be reduced to one to four weeks from several months, potentially saving up to 70 percent in costs and time.
Starting the decarbonization planning and implementation process
Companies should begin by identifying critical decarbonization initiatives, which will depend on factors such as industry and geography. For example, country-by-country differences in the price and availability of renewable energy will be an important determinant of the impact of switching energy sources. With decarbonization initiatives identified, companies can take the following high-level steps:
- Develop a technology-enabled decarbonization plan. The plan, embedded within companies’ systems and based on operational data rather than estimates, can provide the basis for a model that can be frequently updated as new solutions become available.
- Understand the potential of cloud-powered technologies for each initiative. Companies will need to identify, within their decarbonization plans, the initiatives where cloud-powered technology can have the biggest carbon and cost impact to accelerate their transition.
- Develop an investment and implementation plan. By developing a carefully sequenced plan, companies can maximize decarbonization through judicious investment in critical technologies. Crucially, businesses can consider synergies between the use of cloud-powered technologies for decarbonization and their broader business-backed technology strategy.
Companies aiming to take a leading role in decarbonization cannot afford to overlook the adoption of cloud-powered technologies. In everything from assessment and reporting to major transformations and business enablement, these technologies serve as a crucial tool for achieving decarbonization goals swiftly and efficiently. As companies gear up to comply with increasingly complex sustainability regulations, strategic use of cloud-powered technologies can be a crucial differentiator.