Deep dive: Embodied carbon measurement and reduction — data quality gaps and how to close them

Why It Matters

Buildings and construction account for roughly 37 percent of global energy-related CO₂ emissions, and embodied carbon from materials, transport and on-site construction processes represents nearly 11 percent of that total according to the United Nations Environment Programme (UNEP, 2024). As operational energy efficiency improves through better insulation, heat pumps and smart controls, embodied carbon's relative share is growing. A 2025 analysis by the World Green Building Council (WorldGBC, 2025) estimated that for a typical new-build meeting near-zero operational standards, upfront embodied carbon can exceed 50 percent of total lifecycle emissions. Despite this significance, the data underpinning embodied carbon assessments remains fragmented, inconsistent and difficult to compare across regions. Environmental Product Declarations (EPDs) vary widely in scope and methodology, regional databases use different background datasets, and Scope 3 boundaries are drawn inconsistently from one project to another. Closing these data quality gaps is not merely an academic exercise. It determines whether architects, developers and policymakers can set credible benchmarks, specify lower-carbon materials with confidence and verify that reduction targets are actually being met.

Key Concepts

Whole-life carbon assessment (WLCA). WLCA captures emissions across all lifecycle stages of a building, from raw material extraction (module A1) through manufacturing (A3), transport (A4), construction (A5), use-phase replacements (B1 to B5), operational energy (B6), and end-of-life demolition and disposal (C1 to C4), plus any benefits from reuse or recycling (module D). The European standard EN 15978 provides the framework, while ISO 21930 governs product-level declarations. EN 15978 was updated in 2024 to strengthen requirements for biogenic carbon accounting and module D reporting (CEN, 2024).

Environmental Product Declarations (EPDs). EPDs are third-party verified documents that quantify a product's environmental impacts using life cycle assessment (LCA). They are meant to enable apples-to-apples comparisons, but in practice significant variability exists. A 2025 study by the Carbon Leadership Forum (CLF, 2025) found that EPDs for structurally equivalent ready-mix concrete products showed embodied carbon values ranging from 180 to 450 kgCO₂e per cubic metre depending on which Product Category Rules (PCRs) were applied, which background database was used and whether transport and end-of-life modules were included.

Carbon databases and generic values. When project-specific EPDs are unavailable, practitioners rely on generic databases such as the Inventory of Carbon and Energy (ICE) maintained by the University of Bath, Ökobaudat in Germany, KBOB in Switzerland, or the US-based EPD libraries from the National Ready Mixed Concrete Association. These databases use different system boundaries, electricity grid mixes and allocation methods, meaning a generic value for structural steel in Germany can differ by 30 to 60 percent from one in the United States (One Click LCA, 2025).

Scope 3 boundary challenges. Embodied carbon sits largely within the Scope 3 emissions of developers, contractors and investors. The GHG Protocol's Scope 3 Category 1 (purchased goods) and Category 2 (capital goods) apply, but guidance on system boundaries for construction materials remains limited. The Science Based Targets initiative (SBTi) published updated sector guidance in late 2025 requiring real-estate companies to cover at least 67 percent of Scope 3 by spend or mass, yet many firms still rely on spend-based estimates that can be inaccurate by a factor of two or more (SBTi, 2025).

Benchmarking and targets. Credible reduction requires credible baselines. The Royal Institution of Chartered Surveyors (RICS) published its updated Professional Statement on whole-life carbon assessment in 2024, establishing benchmark ranges by building typology. The London Energy Transformation Initiative (LETI) recommends targets of 300 to 500 kgCO₂e per square metre (modules A1 to A5) for residential buildings, with a 2030 aspiration of under 300 kgCO₂e per square metre (LETI, 2024). These benchmarks are increasingly referenced in planning policy, but their reliability depends entirely on the quality and comparability of the underlying data.

What's Working and What Isn't

Progress on EPD availability. The number of construction-product EPDs registered globally exceeded 120,000 by early 2026, up from approximately 80,000 in 2023 (EPD International, 2026). Major manufacturers of cement, steel and insulation now routinely publish facility-specific EPDs, and several jurisdictions including France, the Netherlands and parts of Canada require them for public procurement. The EC3 tool developed by Building Transparency allows designers to query thousands of EPDs in real time and compare embodied carbon by product category, region and performance quartile.

Digital tools accelerating uptake. Platforms such as One Click LCA, Tally and the open-source EPIC database have made early-stage carbon estimation accessible to architects who previously lacked LCA expertise. One Click LCA reported that its user base grew 45 percent year-on-year in 2025, with over 50,000 projects assessed globally (One Click LCA, 2025). Integration with BIM workflows through plugins for Revit and ArchiCAD means that carbon data can be generated alongside structural and cost models, enabling iterative design optimisation.

Regulatory momentum. France's RE2020 regulation, in force since 2022, set binding whole-life carbon thresholds that tighten every three years. Denmark introduced a 12 kgCO₂e per square metre per year limit for new buildings over 1,000 square metres in 2023, with the threshold dropping in 2025. The EU's revised Energy Performance of Buildings Directive (EPBD recast, 2024) requires member states to calculate and disclose whole-life Global Warming Potential for all new buildings from 2030, creating a harmonised demand signal for reliable embodied carbon data across 27 countries.

Persistent data inconsistency. Despite progress, comparability remains elusive. The Carbon Leadership Forum's 2025 benchmarking study found that identical buildings modelled using different regional databases produced whole-building embodied carbon estimates that varied by up to 40 percent (CLF, 2025). Background electricity grid mixes, allocation rules for co-products such as blast-furnace slag, and inconsistent treatment of biogenic carbon in timber products are the largest sources of divergence. Until PCR harmonisation efforts led by the European Committee for Standardization (CEN) and the International EPD System reach completion, cross-border benchmarking will remain unreliable.

Module D ambiguity. Module D captures potential benefits from material reuse and recycling beyond the building's lifecycle. Its inclusion is optional in many frameworks and its calculation methodology is contested. Some EPDs for steel claim large module D credits based on high recycling rates, which can reduce the reported A1-to-A3 footprint by 20 to 40 percent. Critics argue this creates perverse incentives and masks real production emissions. The updated EN 15978 requires module D to be reported separately, but adoption is uneven.

Scope 3 estimation gaps. Many developers still rely on spend-based emission factors rather than product-specific data, particularly for fit-out materials, mechanical systems and electrical equipment. A 2024 survey by the UK Green Building Council (UKGBC, 2024) found that only 18 percent of commercial real-estate developers could quantify embodied carbon for more than 80 percent of their material specifications using product-specific data. The remainder used industry averages or extrapolations that are insufficiently granular to drive procurement decisions.

Key Players

Established Leaders

• One Click LCA — Market-leading whole-life carbon software with 50,000+ projects assessed and integrations across major BIM platforms.
• Building Transparency (EC3) — Open-access Embodied Carbon in Construction Calculator with the largest free EPD database in North America.
• EPD International (The International EPD System) — Global programme operator for EPDs, hosting over 120,000 declarations.
• RICS — Professional body that sets whole-life carbon assessment standards adopted globally.

Emerging Startups

• Tangible Materials — AI-powered platform that benchmarks embodied carbon across supply chains and recommends lower-carbon material substitutions.
• 2050 Materials — Database and analytics engine that aggregates and normalises product-level carbon data for early-stage design.
• Cerclos — Circular-economy platform connecting demolition sites with new-build projects to facilitate material reuse and track avoided embodied carbon.

Key Investors/Funders

• Breakthrough Energy Ventures — Investing in low-carbon cement, steel and timber innovations that directly reduce embodied carbon at source.
• LETI (London Energy Transformation Initiative) — Industry-funded network publishing open-source benchmarks and target-setting guidance.
• Laudes Foundation — Funding systemic initiatives to decarbonise the built environment value chain, including data infrastructure and standards development.

Examples

Skanska's EPD-first procurement (Sweden/UK). Skanska adopted a policy in 2024 requiring product-specific EPDs for all structural concrete and steel on projects exceeding €10 million. On its Malmö residential development, the contractor used EC3 and One Click LCA to compare EPDs from 14 concrete suppliers, ultimately selecting a supplier whose Portland Limestone Cement mix delivered 32 percent lower embodied carbon than the regional generic value. Skanska reported that the EPD-first approach added less than 0.5 percent to material costs while reducing modules A1-to-A3 emissions by approximately 1,200 tCO₂e across the project (Skanska, 2025).

France's RE2020 in practice. Bouygues Construction documented its experience meeting RE2020 whole-life carbon thresholds on a 15,000 square metre office building in Lyon. The project team ran more than 40 design iterations using One Click LCA to test combinations of structural timber, low-clinker concrete and recycled steel. The final design achieved 580 kgCO₂e per square metre (A1 to C4), 18 percent below the regulatory threshold, while maintaining cost parity with a conventional design. Bouygues credited the regulation's prescriptive calculation method with creating a level playing field, though the team noted that discrepancies between the French INIES database and manufacturer EPDs required manual reconciliation on at least six product categories (Bouygues, 2025).

Microsoft's campus expansion (Redmond, USA). Microsoft's 2024 Puget Sound campus expansion set internal embodied carbon budgets per building and used Building Transparency's EC3 tool to track procurement decisions against those budgets in real time. The project achieved a 30 percent reduction in upfront embodied carbon relative to a conventional baseline by specifying low-carbon concrete with supplementary cementite materials, electric-arc-furnace steel and mass timber hybrid structures. Microsoft published its methodology as an open-source playbook to encourage replication across the technology sector (Microsoft, 2025).

City of Amsterdam material passport requirement. Amsterdam's 2025 circular construction policy requires all new municipal buildings to include a material passport documenting the embodied carbon and reuse potential of every structural and facade element. The Buiksloterham neighbourhood pilot used Madaster, a digital platform for material passports, to register over 8,000 building components. Early results show that buildings with passports achieved an average 22 percent lower embodied carbon through design-for-disassembly strategies and reuse of reclaimed steel and concrete (City of Amsterdam, 2025).

Action Checklist

• Require product-specific EPDs for all structural materials in procurement specifications, prioritising facility-specific over industry-average declarations.
• Standardise your LCA methodology by selecting a single background database and PCR framework for all projects and documenting any deviations transparently.
• Set project-level embodied carbon budgets early in design using published benchmarks such as LETI or RICS targets, and track performance against budgets at each design stage.
• Integrate carbon assessment into BIM using tools like One Click LCA or Tally to enable real-time iteration and avoid late-stage value engineering that ignores carbon impacts.
• Report module D separately and avoid netting it against upfront emissions to maintain transparency about actual production impacts versus future recycling potential.
• Invest in Scope 3 data quality by transitioning from spend-based estimates to activity-based and product-specific emission factors for at least 80 percent of material specifications.
• Engage your supply chain by communicating carbon expectations to suppliers, incentivising EPD publication and rewarding lower-carbon products in tender evaluation criteria.

FAQ

Why do embodied carbon estimates for the same building vary so much? Variations arise from differences in background databases, system boundary choices (which lifecycle modules are included), allocation rules for co-products such as slag or fly ash, and whether biogenic carbon in timber is counted. A building modelled with the German Ökobaudat database may show substantially different results than the same building modelled with ICE or North American EPD averages. Until PCR and database harmonisation efforts are complete, practitioners should document their assumptions transparently and use sensitivity analysis to test the impact of methodological choices.

Are EPDs reliable enough to drive procurement decisions? EPDs vary in quality and scope, but they remain the best available tool for product-level carbon comparison. The key is to compare EPDs that use the same PCR and cover the same lifecycle modules. Facility-specific EPDs are generally more reliable than industry-average ones. Third-party verification by an accredited body provides an additional layer of assurance. Designers should use EPD data directionally to identify lower-carbon options within a product category, rather than treating decimal-point differences as definitive.

What role does regulation play in improving data quality? Regulation is the single most powerful driver. France's RE2020 forced the development of a comprehensive national database (INIES) and standardised calculation methods. Denmark's carbon limits created demand for reliable EPDs across the Nordic market. The EU EPBD recast will extend this effect across all 27 member states by 2030. Regulatory requirements create the demand signal that motivates manufacturers to invest in facility-specific EPDs and database operators to harmonise methodologies.

How should firms handle the transition from generic to product-specific data? Start by identifying the materials that dominate your embodied carbon profile, typically concrete, steel, aluminium and insulation. Require product-specific EPDs for these categories first, using generic data only for lower-impact items. Track the percentage of your material specifications covered by product-specific data and set annual improvement targets. Many design tools now flag where generic data is being used, making it straightforward to identify and close gaps progressively.

What is module D and should it be included in benchmarks? Module D accounts for potential environmental benefits beyond the building's lifecycle, such as the avoided emissions from recycling steel or reusing timber at end of life. It should be calculated and reported for transparency, but most leading frameworks recommend excluding it from benchmark comparisons because it represents future potential rather than actual current emissions. Including module D in targets can create misleading comparisons between buildings designed for disassembly and those using conventional construction.

Sources

• United Nations Environment Programme. (2024). 2024 Global Status Report for Buildings and Construction. UNEP.
• World Green Building Council. (2025). Bringing Embodied Carbon Upfront: Coordinated Action for the Building and Construction Sector. WorldGBC.
• Carbon Leadership Forum. (2025). Embodied Carbon Benchmark Study: Cross-Database Variability in Whole-Building Assessments. University of Washington.
• One Click LCA. (2025). Annual Impact Report: 50,000 Projects and the State of Embodied Carbon Data. One Click LCA.
• EPD International. (2026). Global EPD Registry Statistics: Q1 2026 Update. The International EPD System.
• LETI. (2024). Embodied Carbon Target Alignment: Residential and Commercial Benchmarks. London Energy Transformation Initiative.
• European Committee for Standardization. (2024). EN 15978:2024 Sustainability of Construction Works — Assessment of Environmental Performance of Buildings. CEN.
• Science Based Targets initiative. (2025). Buildings Sector Guidance: Scope 3 Boundary Requirements for Real Estate Companies. SBTi.
• UK Green Building Council. (2024). Embodied Carbon Data Maturity Survey: Commercial Real Estate Sector. UKGBC.
• Skanska. (2025). EPD-First Procurement: Lessons from the Malmö Residential Development. Skanska Sustainability Report.
• Bouygues Construction. (2025). Meeting RE2020: Whole-Life Carbon Optimisation on a Lyon Office Project. Bouygues Technical Report.
• Microsoft. (2025). Puget Sound Campus Expansion: Embodied Carbon Reduction Playbook. Microsoft Green Building Initiative.
• City of Amsterdam. (2025). Circular Construction Policy: Material Passport Pilot Results from Buiksloterham. Municipality of Amsterdam.
• RICS. (2024). Professional Statement: Whole Life Carbon Assessment for the Built Environment. Royal Institution of Chartered Surveyors.