Sustainability & ESG: Navigating IT Challenges
IT systems are no longer just about performance and cost. Businesses now face pressure to measure digital emissions, disclose sustainability risks and prevent technology expansion from undermining climate targets. These concerns affect data centres, cloud contracts, hardware procurement and the rapid growth of artificial intelligence infrastructure.
Environmental, social and governance (ESG) considerations have moved into technology management. Consequently, IT leaders must now address power consumption, carbon reporting, electronic waste and supply chain transparency alongside traditional operational priorities.
The shift reflects three parallel developments. First, regulators and investors expect more reliable sustainability disclosure. Second, AI workloads are increasing electricity demand at data centres. Third, organisations recognise that technology drives emissions but can also help reduce them elsewhere in the business.
For UK small and medium businesses, these trends create both obligations and choices. Some face reporting requirements through public sector contracts or supply chain requests. Others need to manage rising energy costs or prepare for stricter environmental standards. Many are simply trying to understand what credible action looks like.
IT infrastructure creates a measurable environmental footprint
Technology affects an organisation's carbon footprint through multiple channels. Data centres consume electricity for computing and cooling. Networks transmit data. End-user devices require manufacturing, power and eventual disposal. Cloud services shift those impacts to providers but do not eliminate them.
A credible sustainability programme starts with measurement. Businesses should establish a baseline covering energy use, emissions, electronic waste and resource allocation. That baseline supports decisions about server utilisation, data centre efficiency, renewable energy procurement and hardware lifecycle management.
IT teams also influence Scope 3 emissions through purchasing decisions. Supplier emissions, product durability, repairability and take-back programmes all matter. Procurement based solely on price and performance can transfer environmental costs elsewhere in the value chain.
Several frameworks help organisations set sustainability criteria. The Greenhouse Gas Protocol defines how to measure emissions across scopes. ISO 14001 provides an environmental management system standard. The Global Reporting Initiative offers disclosure guidelines. In addition, the UK government promotes standards like ENERGY STAR for equipment efficiency.
Cross-functional collaboration improves outcomes. IT leaders, procurement teams, finance, compliance and business units need shared objectives and accountability. Five to ten year action plans can connect sustainability targets with technology upgrades and contract renewals.
Artificial intelligence drives data centre electricity demand sharply upward
AI workloads have intensified the relationship between digital growth and environmental impact. These workloads depend on accelerated servers that consume more power than conventional infrastructure. They also require more cooling, grid capacity and physical space.
The International Energy Agency estimates that data centres consumed approximately 415 terawatt-hours of electricity in 2024. This represents around 1.5 per cent of global electricity consumption. However, the base-case outlook projects demand will more than double to roughly 945 TWh by 2030, approaching 3 per cent of global electricity use.
Accelerated servers, primarily driven by AI adoption, are expected to grow particularly quickly. The IEA projects their electricity consumption will increase by about 30 per cent annually in its base case. Conventional servers are growing more slowly by comparison.
These effects are concentrated geographically rather than spread evenly. Data centres cluster in specific regions, creating local competition for grid capacity. This can raise concerns about water availability for cooling, delay new technology projects and increase costs where electricity supply is constrained.
Efficiency gains alone will not solve the problem. Power consumption per AI task is falling as chips, models and infrastructure improve. Nevertheless, overall demand may continue rising because AI use is expanding and more energy-intensive applications are being deployed. Autonomous AI agents, for example, require sustained computing over longer periods than simple queries.
Carbon intensity depends on electricity generation, not just consumption
The climate impact of digital infrastructure depends on how electricity is generated, not only how much is used. The IEA estimates that data centre electricity-related emissions could rise from approximately 180 million metric tons of carbon dioxide today to around 300 million metric tons by 2035 in its base case.
Renewable energy is expected to supply nearly half of the additional electricity required by data centres over the next five years. However, natural gas, coal and nuclear power will also contribute to meeting demand. Therefore, claims about efficient infrastructure do not necessarily demonstrate low-carbon outcomes.
Businesses must examine both consumption and the carbon intensity of the electricity supplying their workloads. This distinction matters when choosing cloud providers, selecting data centre locations and reporting emissions. A facility powered by coal-fired electricity creates higher emissions than one using renewable energy, even if both run identical workloads.
Practical measures for reducing carbon intensity include locating workloads in regions with lower-carbon electricity grids. Flexible computing can be scheduled for periods of renewable availability. Server utilisation should be improved to reduce idle capacity. Cloud providers should be selected based on transparent energy and emissions data.
More efficient models, chips and cooling systems also help. Meanwhile, renewable energy claims must match credible procurement instruments like power purchase agreements or renewable energy certificates. Operational emissions should be reported separately from avoided emissions claims to prevent confusion.
Reporting standards are becoming more formal and investor-focused
Sustainability disclosure is shifting toward structured, investor-oriented reporting. The International Sustainability Standards Board has issued IFRS S1, which requires entities to disclose sustainability-related risks and opportunities that could reasonably affect cash flows, access to finance or cost of capital.
IFRS S1 covers four broad areas. First, governance processes, controls and oversight. Second, strategy for managing sustainability-related risks and opportunities. Third, processes for identifying, assessing and monitoring those issues. Fourth, performance including progress toward relevant targets.
The standard is effective for annual reporting periods beginning on or after 1 January 2024. However, adoption and implementation depend on the rules of individual jurisdictions. The IFRS Foundation has reported that more than 45 jurisdictions are on a path toward adopting or otherwise using the ISSB standards.
Differences between national regimes could create fragmentation and cross-border reporting challenges. For technology leaders, this shift means sustainability data can no longer be treated solely as a communications exercise. IT systems may need to provide auditable information about electricity use, emissions, assets, suppliers and cloud workloads.
Poor data quality weakens both regulatory reporting and investor confidence. Organisations that publish ambitious targets without robust measurement face accusations of greenwashing. Those that ignore infrastructure impacts may encounter higher energy costs, grid constraints and reputational damage.
Practical steps for building a credible IT sustainability strategy
A credible approach connects measurable environmental performance with operational and financial outcomes. It treats carbon and resource data as management information, measured consistently and subject to the same governance expected of other material business risks.
Businesses should start by establishing a baseline. Measure energy consumption, emissions, hardware inventories, electronic waste, data centre performance and major supplier impacts. The baseline should identify data limitations and distinguish measured figures from estimates. This creates a foundation for target-setting and investment decisions.
Not every technology activity has the same impact. Organisations should prioritise high-energy workloads, data centres, AI infrastructure, large procurement categories and cooling systems. Assets approaching end of life also deserve attention because replacement decisions affect emissions for years ahead.
Procurement processes should build sustainability into vendor assessments. Evaluate environmental policies, carbon footprints, renewable energy use, product durability and repairability. Examine take-back programmes and the transparency of sustainability reporting. Decisions based only on price and performance transfer environmental costs elsewhere in the value chain.
Cloud migration does not automatically reduce emissions. Benefits depend on workload efficiency, utilisation, geographic placement, service configuration and the provider's energy mix. Architecture reviews should consider storage retention, data movement, redundant workloads, server utilisation and software efficiency. These factors determine whether cloud adoption reduces carbon impact or simply relocates it.
Electronic waste represents both a risk and an opportunity. Longer device lifecycles, repair, refurbishment and responsible recycling reduce material extraction and waste. Asset management should track equipment from purchase through reuse or disposal, with controls against informal or unverifiable recycling channels. This approach lowers costs and environmental impact simultaneously.
Sustainability objectives should have owners, budgets, timelines and performance indicators. Examples include reducing data centre energy intensity, increasing device reuse, lowering cloud emissions per transaction or improving the percentage of suppliers providing verified emissions data. Accountability ensures that targets translate into action rather than remaining aspirational statements.
The commercial case for action is strengthening across multiple fronts
Three trends are converging to make sustainability a core technology management issue. Rapidly expanding AI infrastructure is increasing electricity demand. Pressure for reliable sustainability disclosure is intensifying from regulators, investors and customers. Recognition is growing that technology drives emissions but can also help reduce them elsewhere.
The opportunity is substantial. More efficient software, optimised logistics, smart grid systems, industrial monitoring and data-driven energy management could help reduce emissions in sectors beyond IT. AI could improve energy sector efficiency and support emissions reductions, although those benefits depend on widespread, well-governed deployment.
For UK businesses, sustainability increasingly affects procurement decisions. Public sector suppliers face carbon reduction requirements under Procurement Policy Note 06/21. Private sector buyers are asking similar questions about supply chain emissions. Businesses that demonstrate credible environmental management gain access to contracts and customers that require it.
Energy costs are also rising. Efficient infrastructure reduces operating expenses as well as emissions. Businesses that improve server utilisation, extend hardware lifecycles and optimise cloud workloads can lower costs while meeting environmental targets. The financial and environmental cases often align.
Risks are equally significant. Efficiency improvements can be outweighed by rising demand. Renewable energy claims can lack transparency. Fragmented reporting requirements increase compliance costs. Organisations that publish ambitious targets without robust measurement face accusations of greenwashing. Those that ignore infrastructure impacts may encounter higher energy costs, grid constraints and reputational damage.
For IT leaders, sustainability is now an operating requirement that affects architecture, procurement, risk management, financial planning and compliance. The strongest strategies treat environmental performance as a business discipline rather than a separate initiative. They measure impacts consistently, connect data to decisions and apply governance across the organisation.
Eight essential facts about IT sustainability and energy demand
- Data centres consumed approximately 415 terawatt-hours of electricity globally in 2024, representing around 1.5 per cent of total global electricity consumption according to International Energy Agency analysis.
- The IEA base-case outlook projects data centre electricity demand will more than double to roughly 945 TWh by 2030, approaching 3 per cent of global electricity use.
- Electricity consumption from accelerated servers, primarily driven by AI adoption, is projected to grow by about 30 per cent annually in the IEA base case.
- Data centre electricity-related carbon dioxide emissions could rise from approximately 180 million metric tons today to around 300 million metric tons by 2035 in the IEA base case.
- Renewable energy sources are expected to supply nearly half of the additional electricity required by data centres over the next five years, with natural gas, coal and nuclear power also contributing to meeting demand.
- IFRS S1 became effective for annual reporting periods beginning on or after 1 January 2024, requiring entities to disclose sustainability-related risks and opportunities that could affect cash flows, finance access or capital costs.
- More than 45 jurisdictions are on a path toward adopting or otherwise using the International Sustainability Standards Board standards, according to the IFRS Foundation.
- Efficiency improvements in AI infrastructure are reducing power consumption per task, but overall demand may continue rising as AI use expands and more energy-intensive applications are deployed.
Where businesses can find guidance and support
Organisations looking to develop credible IT sustainability strategies should start with authoritative frameworks and standards. The Greenhouse Gas Protocol provides the most widely used methodology for measuring and reporting emissions across Scope 1, Scope 2 and Scope 3 categories. It helps businesses understand which emissions they need to measure and how to categorise them correctly.
The International Energy Agency's analysis of energy and AI offers detailed projections about data centre electricity demand, efficiency trends and carbon intensity. This research helps businesses understand the scale of the challenge and the factors that will shape future infrastructure decisions.
For reporting requirements, the IFRS Foundation's IFRS S1 standard sets out general requirements for disclosure of sustainability-related financial information. While adoption varies by jurisdiction, understanding these standards helps businesses prepare for investor and regulatory expectations.
UK businesses working with the public sector should review Procurement Policy Note 06/21 guidance from the Cabinet Office. This explains carbon reduction plan requirements for suppliers bidding on central government contracts above certain thresholds.
SBS provides practical support for UK small and medium businesses navigating these requirements. Our net-zero programme for carbon reporting compliance helps organisations measure emissions, set credible targets and meet procurement requirements. We also offer ESG compliance support for carbon reporting services to businesses facing investor or regulatory disclosure obligations.
Technology sustainability is becoming a business discipline rather than a technical exercise. Organisations that treat it seriously, measure impacts consistently and connect environmental performance to commercial decisions will be better positioned for the regulatory, financial and competitive environment ahead.