Emissions factor database: Why it matters more than method

Why your emissions factor database matters more than your calculation methodology

Your carbon footprint is only as accurate as the emissions factors behind it. Most sustainability managers invest significant time choosing between calculation methodologies: activity-based versus spend-based, location-based versus market-based, GHG Protocol versus ISO 14064. These are legitimate decisions. But the single greatest source of inaccuracy in most corporate carbon footprints is not the methodology. It is the emissions factor database used to convert activity data into CO2e.

Specifically: using the wrong regional factor, an outdated database vintage, or a globally averaged figure where a jurisdiction-specific value should have been applied.

This guide explains what emissions factor databases are, why regional accuracy matters so much, how the major databases differ, and what to look for when choosing a carbon accounting platform.

What is an emissions factor database?

An emissions factor is a coefficient that converts a unit of activity into a quantity of greenhouse gas emissions. One kilowatt-hour of electricity consumed, one litre of diesel burned, one kilogram of refrigerant released: each requires a factor to translate it into kilograms or tonnes of CO2e.

An emissions factor database is a structured collection of these factors, organised by activity type, fuel, geography, and year. They are the foundation of every carbon footprint calculation. As the GHG Protocol makes clear, the accuracy of any inventory is limited by the quality of the factors applied to it: the right methodology with the wrong factor produces the wrong answer.

The major authoritative sources include:

• UK: DESNZ (formerly DEFRA) GHG Conversion Factors, published annually by the UK Government
• Australia: National Greenhouse Accounts (NGA) Factors, published annually by DCCEEW
• EU: IEA Emissions from Fuel Combustion dataset and country-level grid factors under ESRS E1
• US: EPA eGRID for electricity, EPA Emission Factors Hub for broader categories
• Global: IEA, Ember Yearly Electricity Data, IPCC Emission Factor Database

Each database has its own update cycle, geographic scope, and methodological basis. Using the wrong one for your operating region is one of the most common and least visible errors in corporate carbon accounting.

Why regional factors matter: the electricity grid example

Electricity is the clearest illustration of why regional accuracy is not optional. Grid carbon intensity varies dramatically by country, and within countries by state or region, based entirely on the generation mix.

Under the 2026 DESNZ factors published in June 2026, the UK’s location-based electricity factor fell to 0.131 kg CO2e per kWh, a 26% reduction from the 2025 figure of 0.177, driven by a mix of genuine grid decarbonisation and a DESNZ methodology change (more on this in Database vintage, below). Zero-carbon sources now generate approximately 63% of UK electricity. Applying the 2025 factor to 2026 activity data would overstate your Scope 2 electricity emissions by more than a quarter.

The variation across countries is even more striking. The figures below are drawn from a single consistent source: the Ember Yearly Electricity Data 2025 (covering calendar year 2024), the most widely used open dataset for international grid factors. The UK figure uses DESNZ 2026, the authoritative UK-specific source, as Ember does not publish a separate UK row.

Note: factors reflect the generation mix for the data year shown. Update annually — Ember publishes a new release each April.

The range from Norway to India is more than 25:1. A multinational applying a single global average to electricity consumption across all its sites is not producing a carbon footprint. It is producing a rough approximation with systematic error built in.

The UK’s grid factor reflects a generation mix that is now more than 63% zero-carbon. Applying it to a coal-heavy grid in another country would understate Scope 2 emissions by more than half. The DESNZ guidance is explicit that its electricity factor applies to UK consumption only.

The Australian state-level problem

The regional variation challenge is especially acute in Australia, where the National Greenhouse Accounts Factors publish separate electricity factors for each state and territory reflecting their very different generation mixes.

Under the NGA Factors 2025 workbook published by DCCEEW:

• Victoria’s Scope 2 electricity factor is 0.78 kg CO2e/kWh
• Tasmania’s is 0.20 kg CO2e/kWh

That is almost a four-fold difference between two states in the same country under the same reporting framework. Apply the national average factor of 0.62 to a multi-site operation with sites in both states and every figure is wrong: Victorian sites are understated; Tasmanian sites are significantly overstated.

This matters for NGER reporting, where the Clean Energy Regulator requires state-specific factors for electricity consumption under Method A1 in the National Electricity Market. It also matters for AASB S2 climate disclosure, where Scope 2 figures are subject to phased assurance requirements. An auditor reviewing your AASB S2 disclosure will check whether your Scope 2 calculation uses state-level or national average factors, and the distinction is material.

DCCEEW switched from a three-year averaging approach to single-year data from AEMO in September 2024. This means factors are more current but also more volatile year-on-year. A site in Tasmania, which relies heavily on hydro, can see its factor shift significantly in a low-rainfall year when imports via Basslink increase. Understanding that variation is essential for accurately explaining emissions movements in your annual disclosure. See our guide to why regional emission factors matter for multi-site operations for a deeper look at this.

Database vintage: why the year matters as much as the source

Using the right geographic factor is half the problem. Using the right vintage year is the other half.

Emissions factor databases are updated annually because the underlying reality they represent changes every year. Grid mixes shift as renewables are added and coal plants retire. Fuel compositions change. Supply chain emissions for materials evolve as manufacturing processes improve.

The 2026 DESNZ update illustrates the stakes. The UK location-based electricity factor fell 26% in a single year, from 0.177 to 0.131 kg CO2e/kWh, because of a combination of a genuine grid decarbonisation and a methodology change that reduced the data lag from two years to one, alongside minor corrections to auto generator and import/export treatment… For an organisation consuming 10,000 MWh of UK grid electricity annually, this difference is approximately 460 tonnes CO2e. That is not a rounding error. It is material to disclosure, to target-setting, and to year-on-year comparisons, and organisations should disclose both drivers rather than treat the fall as a like-for-like improvement.

The same applies to Australia. The NGA Factors are published each September and should be applied to the following year’s reporting period. Organisations using the previous year’s factors for current reporting are introducing a systematic lag error.

The DESNZ factors are published each June and the UK Government does not retrospectively revise previously published factors, meaning which version you used for which reporting period is a permanent record. Documenting your factor source and vintage year is not just good practice. Under CSRD assurance requirements and AASB S2 limited assurance, it is an auditable requirement.

How the major databases differ

Understanding the differences between the principal databases helps sustainability managers choose the right source for each activity category.

DESNZ / DEFRA (UK)

Published annually in June by the Department for Energy Security and Net Zero and the Department for Environment, Food and Rural Affairs. The most widely used emissions factor dataset in the UK, accepted by CDP, ISO 14064-1, and CSRD ESRS E1 as a credible methodology basis. Covers electricity, fuel combustion, transport, freight, business travel, waste, and water.

DEFRA fuel combustion factors are broadly applicable globally because combustion chemistry is universal and UK fuel composition aligns closely with OECD averages. The electricity factor, however, is UK-specific and must not be applied to non-UK operations. All DEFRA factors use IPCC AR5 GWP-100 values.

The 2026 update introduced a significant methodology change: a new approach uses more up-to-date electricity generation data, reducing the data lag from two years to one. Alongside genuine grid decarbonisation, this is behind the 26% reduction in the UK location-based electricity factor (see Database vintage, above, for the full breakdown). Organisations comparing year-on-year Scope 2 figures should disclose this methodology change.

NGA Factors (Australia)

Published annually by the Department of Climate Change, Energy, the Environment and Water (DCCEEW), the NGA Factors are the primary source for Australian carbon accounting under both NGER and AASB S2. They cover electricity by state and territory, fuel combustion, transport, and refrigerants.

The official DCCEEW NGA Factors 2025 publication is the authoritative source for NGER reporting. Note: the NGA Factors should not be used directly for NGER Act reporting, which has its own NGER Measurement Determination. For voluntary reporting and AASB S2 purposes, NGA Factors are the appropriate source.

The Australian Government’s electricity generation fuel mix data for calendar year 2025 shows clearly why state variation is so significant: coal continues to dominate in NSW, Victoria, and Queensland, while renewables accounted for 97.5% of Tasmanian generation and 77% in South Australia in 2025.

EU grid factors (ESRS E1 / CSRD)

For companies reporting under CSRD and ESRS E1, country-level grid emission factors for EU member states are required for Scope 2 location-based disclosure. These are drawn from the IEA Emissions from Fuel Combustion dataset and updated annually.

The variation within the EU mirrors the global picture: a manufacturer in Poland operates on a grid with approximately 26 times the carbon intensity of a Swedish competitor. Applying a single EU average to a multi-country operation introduces systematic error that is directly visible to assurance practitioners reviewing ESRS E1 disclosures.

Under CSRD, ESRS E1 requires both location-based and market-based Scope 2 reporting. Location-based uses the grid emission factor for the region of consumption. Market-based uses supplier-specific factors or residual-mix factors where renewable energy certificates apply. Both require the correct regional factor vintage.

Material-specific emission factors: beyond electricity

The database vintage and regional accuracy problem extends beyond electricity to material emission factors for Scope 3 reporting.

For construction and manufacturing companies, the embodied carbon in purchased materials, Category 1 under the GHG Protocol Scope 3 Standard, depends on material-specific emission factors that vary by production method, country of manufacture, and the energy mix of the producing facility. A tonne of steel produced in Sweden has a fundamentally different carbon footprint to the same tonne produced in China, and a globally averaged steel emission factor will not capture that difference.

For freight categories (Category 4 upstream transportation, Category 9 downstream transportation), the emission factor varies by mode, fuel type, and vehicle class. Air freight has a dramatically higher factor per tonne-kilometre than sea freight. Using a blended average across modes when your actual freight split is known introduces unnecessary inaccuracy.

For refrigerants and fugitive emissions, the GWP values themselves vary between IPCC AR5 and AR6 assessment reports. AASB S2 requires AR6 GWP values. Some legacy databases still carry AR5 values. For operations with significant methane emissions, the difference is not trivial: IPCC AR6 revised the 100-year GWP for fossil methane (CH4) to 29.8, compared to 28 in AR5. Note that AR6 distinguishes between fossil and biogenic methane, with biogenic methane carrying a lower GWP of approximately 27.0. If your operations involve both, confirm which value applies to each emission source before applying a single factor. See our AASB S2 Scope 3 guide for how this flows through to supplier data collection.

What this means for your 2026 disclosure

For organisations preparing disclosures under AASB S2 (Australia), CSRD (EU), or UK SRS, the practical implications are clear.

Your emissions factor database is not a technical detail. It is a fundamental input to the figures you are disclosing to investors, assurers, and regulators. Factors that are out of date, geographically wrong, or drawn from an inappropriate source will produce a footprint that cannot survive limited assurance review.

Three questions worth confirming for your current reporting:

Which database are you using for each emission category? DEFRA for UK electricity, NGA for Australian electricity by state, IEA for EU grid factors, EPA eGRID for US operations. Using a single global database across all regions introduces the systematic regional error described above.

What vintage year are your factors? For 2026 disclosures, you should be using DESNZ 2026 (published June 2026), NGA Factors 2025 (published September 2025, applied to 2025-26 reporting), and IEA 2025 data for EU and international grid factors.

Are your GWP values consistent with your framework requirements? AASB S2 requires IPCC AR6. DESNZ uses AR5. If you are combining factors from multiple databases, consistency in GWP basis is an auditable requirement.

These are not questions an assurance provider will overlook. They are the first tier of methodology review in any limited assurance engagement.

How eco-shaper handles emission factor accuracy

Manually tracking, updating, and applying the correct regional emission factor across every emission category, every site, every year is one of the primary sources of error and administrative burden in corporate carbon accounting.

eco-shaper’s platform applies emission factors from a database of 120,000 regionally specific factors updated annually, across electricity grids, fuels, materials, freight, and refrigerants. The correct factor is applied automatically based on where the activity occurs: a facility in Victoria gets the Victorian NGA electricity factor, not the national average. A UK office gets the current DESNZ factor, not the previous year’s figure. A supplier in Germany gets the German grid factor, not an EU average.

This matters not just for accuracy but for audit readiness. Under AASB S2, CSRD, and UK SRS, every factor used in your disclosure needs to be traceable to its source. eco-shaper’s platform maintains that provenance automatically, so your assurance provider can verify the basis of every figure in your report.

For a practical overview of how this works across multi-site and multi-region operations, see our Supply Chain Engagement feature and our AASB S2 Group 2 guide.

The bottom line

The best calculation methodology in the world cannot compensate for an outdated or geographically wrong emissions factor. The two work together: methodology determines what you measure, and the factor determines how accurately you measure it.

For sustainability managers and ESG directors preparing disclosures in 2026, the emissions factor database your platform uses deserves the same scrutiny as the methodology it applies. Ask your platform provider which databases they use, how frequently they update them, and how they handle regional variation. The answers will tell you a great deal about the reliability of the figures you are signing off on.

The method is only as good as the factor behind it.

eco-shaper is a carbon accounting and sustainability reporting platform used by organisations across Australia, the UK, EU and beyond. Our platform applies 120,000 regionally specific, annually updated emission factors automatically across Scope 1, 2, and 3 calculations. Speak to our team to learn more.