What an Environmental Product Declaration actually is
An Environmental Product Declaration (EPD) is a Type III environmental declaration under ISO 14025: a standardized, independently verified, quantified report of a product’s environmental performance across its life cycle, produced against a set of Product Category Rules and published by a recognized programme operator.
Three parts of that definition do the work, and each is a place where declarations fail:
- Quantified, not narrative. An EPD reports impact category results in defined units, calculated from a Life Cycle Assessment conducted under ISO 14040 and ISO 14044. It does not rate, score, or rank the product, and it makes no claim that the product is environmentally preferable.
- Governed by Product Category Rules. The PCR fixes the functional or declared unit, the system boundary, the modules to be declared, cut-off rules, allocation, and data quality requirements. Two EPDs are only comparable if they follow the same PCR and declare the same modules.
- Independently verified and published. Verification by a qualified third party is not optional under ISO 14025. Without it, the document is not an EPD, whatever it is labelled.
Common misconceptions worth clearing early
An EPD is not a certification, and not an environmental label. ISO 14024 Type I labels (Blue Angel, Nordic Swan) certify that a product meets criteria. An EPD asserts nothing about performance thresholds; it reports numbers and lets the reader judge.
An EPD is not automatically comparable to another EPD. Comparability requires the same PCR, the same declared unit, the same declared modules, the same background database and version, and the same characterization method. Cross-comparing declarations from different programme operators without checking those five things produces conclusions that will not survive review.
A good EPD does not mean a low-impact product. The declaration reports what the product does. A high-impact product with a well-executed EPD is still a high-impact product — but its manufacturer now knows exactly where the impact sits.
The value of an EPD is not the PDF. It is the verified life cycle model behind it, which becomes reusable for Scope 3 accounting, product carbon footprints under ISO 14067, eco-design, and tender responses.
The standards that govern EPDs
An EPD sits inside a stack of standards rather than a single one. Knowing which document controls which decision is the difference between a declaration that passes verification first time and one that comes back with findings.
| Standard | What it controls |
|---|---|
| ISO 14025 | The governing framework for Type III environmental declarations. Defines the role of the programme operator, the requirement for Product Category Rules, the verification requirement, and the conditions for comparability. |
| ISO 14040 / ISO 14044 | The underlying LCA methodology: goal and scope, inventory analysis, impact assessment, and interpretation. Every EPD rests on an LCA conducted to these standards. |
| EN 15804+A2 | The core PCR for construction products in Europe. Fixes the declared modules, the mandatory impact categories and their characterization methods (EF 3.0), the biogenic carbon reporting rules, and the treatment of module D. The +A2 amendment substantially changed the indicator set from the original EN 15804+A1. |
| ISO 21930 | The international core rules for construction product EPDs, used where EN 15804 does not apply — predominantly North American programmes. |
| ISO 14067 | Product carbon footprint quantification. Not an EPD standard, but the same underlying model usually serves both, and PCF results are frequently extracted from EPD work. |
| Product-specific PCR / c-PCR | The rules for your specific product category, issued by the programme operator. Complementary PCRs (c-PCR) sit under EN 15804 and refine it for a product group. This is the document that governs your declared unit and your scenarios — read it before modelling anything. |
Choosing a programme operator
The programme operator administers the PCR system, maintains the verifier pool, registers and publishes declarations, and defines the general programme instructions your EPD must satisfy. The choice is commercial as much as technical — it determines which markets recognize the declaration.
Established operators include the International EPD System, IBU in Germany, EPD Norge, EPD Italy, and UL Solutions and NSF in North America. Select on the basis of where your buyers are, whether a suitable PCR already exists in that programme, mutual recognition arrangements with other operators, and verifier availability in your product category. Switching operator mid-project means re-reading the general programme instructions and often re-scoping.
How an EPD is actually developed, step by step
A first EPD typically runs three to six months from kickoff to publication. Modelling is rarely the constraint. Data collection is — specifically, getting verified supplier data and reconciling production records to the declared unit.
Select the programme operator and locate the PCR
Confirm a valid PCR exists for your product category and check its expiry date. If none exists, one must be developed or an existing one extended — that adds months and cost, and it is the single most common reason a first EPD slips.
Deliverable: PCR reference, programme general instructions, verifier shortlist.
Fix the declared unit and system boundary
The PCR dictates the declared or functional unit — 1 m² of installed product at a stated thickness, 1 tonne of material, 1 piece over a reference service life. Get this wrong and every downstream number is wrong. Decide at the same time whether the declaration is cradle-to-grave, or cradle-to-gate with options where the PCR permits it.
Deliverable: goal and scope document, declared unit, module list, reference service life.
Build the data collection framework before collecting anything
Define what is needed per module, per site, for which reference year, and who owns each input. Foreground data covers raw material inputs, energy carriers and quantities, water, process waste, packaging, and transport distances and modes. Ad-hoc collection produces gaps that only surface at verification.
Deliverable: structured collection template, data ownership map, reference year.
Collect and validate primary data
Manufacturing data must be primary and site-specific for modules A1–A3, covering a full representative production year. Reconcile mass and energy balances before modelling: inputs must account for outputs plus losses. Unreconciled balances are a standard verification finding.
Deliverable: validated inventory, mass and energy balance, documented data gaps and their treatment.
Model the system and select background data
Build the model in openLCA, SimaPro, or GaBi against a background database — typically ecoinvent, or the datasets mandated by the programme. Record the database name and exact version; a version mismatch between the model and the report is another routine finding. Apply the allocation rules the PCR requires, and document any deviation with justification.
Deliverable: LCA model, database and version record, allocation justification.
Run the impact assessment and test the result
Calculate the mandatory indicators using the characterization method the PCR specifies — EF 3.0 under EN 15804+A2. Then stress the result: contribution analysis to find the hotspots, sensitivity analysis on the assumptions that drive them, and a scenario check on end-of-life and transport, which are usually assumption-heavy.
Deliverable: indicator results by module, contribution and sensitivity analysis.
Write the background report and the declaration
Two documents, not one. The background LCA report carries the methodology, assumptions, data sources, data quality assessment, and calculations — it is what the verifier reads. The EPD itself is the public-facing summary in the programme’s required format. A thin background report is the fastest route to a failed verification.
Deliverable: background LCA report, draft EPD in programme format.
Verification, registration, and publication
An approved independent verifier checks conformity with the PCR, the underlying standards, and the programme instructions, and issues findings you must close. Once verified, the programme operator registers and publishes the declaration. Budget four to eight weeks for this stage, longer if findings are substantive.
Deliverable: verification statement, registered and published EPD, validity period start date.
What an EN 15804+A2 declaration contains
Results are reported by module, not as a single total. Which modules you declare determines what your EPD can be used for — a cradle-to-gate declaration cannot answer a whole-building assessment question.
| Modules | Stage | What it covers |
|---|---|---|
| A1–A3 | Product stage | Raw material supply, transport to the plant, and manufacturing. Mandatory in every declaration, and the modules requiring primary site data. |
| A4–A5 | Construction process | Transport to the construction site, and installation including losses and installation waste. |
| B1–B7 | Use stage | Use, maintenance, repair, replacement, refurbishment, operational energy use, and operational water use across the reference service life. |
| C1–C4 | End of life | Deconstruction or demolition, transport, waste processing for reuse or recycling, and final disposal. |
| D | Beyond the boundary | Benefits and loads outside the system boundary — recycling potential, energy recovery. Reported separately and never added into the cradle-to-grave total. |
Under EN 15804+A2, cradle-to-grave with module D is the default. Cradle-to-gate with options is permitted only under the conditions the standard sets out, and the declaration must state clearly which route was taken.
The core environmental indicators
EN 15804+A2 moved the indicator set to the EF 3.0 characterization methods and split global warming potential into four separately reported figures. Declarations built against the older +A1 indicator set are not directly comparable with +A2 declarations.
| Indicator | Abbreviation | Unit |
|---|---|---|
| Global warming potential, total | GWP-total | kg CO₂ eq. |
| Global warming potential, fossil | GWP-fossil | kg CO₂ eq. |
| Global warming potential, biogenic | GWP-biogenic | kg CO₂ eq. |
| Global warming potential, land use and land use change | GWP-luluc | kg CO₂ eq. |
| Ozone depletion potential | ODP | kg CFC-11 eq. |
| Acidification potential | AP | mol H⁺ eq. |
| Eutrophication, freshwater | EP-freshwater | kg P eq. |
| Eutrophication, marine | EP-marine | kg N eq. |
| Eutrophication, terrestrial | EP-terrestrial | mol N eq. |
| Photochemical ozone formation | POCP | kg NMVOC eq. |
| Abiotic depletion, minerals and metals | ADP-minerals&metals | kg Sb eq. |
| Abiotic depletion, fossil resources | ADP-fossil | MJ, net calorific value |
| Water deprivation potential | WDP | m³ world eq. deprived |
A further set of indicators — particulate matter, ionising radiation, ecotoxicity freshwater, human toxicity cancer and non-cancer, and land use related soil quality — is declared with a disclaimer on result reliability. Alongside the impact indicators, the declaration also reports resource use, waste categories, and output flows.
Why third-party verification is not optional
Verification is what makes the data unbiased, harmonized, and usable by anyone other than the manufacturer. ISO 14025 treats it as constitutive: a declaration that has not been independently verified is not a Type III declaration, and will not be recognized by certification schemes, procurement frameworks, or regulatory instruments that reference EPDs.
Two failure modes are worth naming. The first is documents marketed as EPDs that meet the format but carry no verification. The second is verification by a party without demonstrable independence from the study. Both leave the buyer holding a document that will be rejected the first time it is scrutinized.
What the verifier actually checks
- Conformity of the LCA with ISO 14040 and ISO 14044
- Conformity with the applicable PCR and the programme’s general instructions
- Plausibility, completeness, and traceability of the underlying data
- Correct application of allocation, cut-off, and system boundary rules
- Consistency between the background report, the model, and the published declaration
Validity, updates, and what invalidates a declaration
EPDs are typically valid for five years from publication. That period is not a guarantee — the declaration must remain representative of the product and process it describes. Material changes require the EPD to be updated and re-verified before the term expires.
Circumstances that commonly trigger an update:
- A change in formulation, supplier, or production site
- A significant change in the energy mix or process efficiency at the manufacturing site
- Expiry or revision of the governing PCR
- A revision of the underlying standard — the EN 15804+A1 to +A2 transition forced a wave of recalculations across the sector
Planning for the update at the outset is what keeps the second EPD cheap. A well-structured model with documented data sources can be re-run in weeks; an undocumented one is rebuilt from scratch.
What the declaration is worth once you have it
Green public procurement and tenders. Increasingly, EPDs are a qualification requirement rather than a differentiator — the absence of one removes you from the shortlist.
Building certification. LEED, BREEAM, and DGNB all award credits for products with verified declarations, and whole-building LCA under those schemes depends on product-level EPD data.
Scope 3 accounting. Purchased goods and services is the largest category in most corporate inventories, and supplier-specific EPD data replaces spend-based estimates with primary data. The same verified model feeds a product carbon footprint under ISO 14067 with limited additional work.
Eco-design and decarbonization. The contribution analysis behind the declaration identifies where the impact actually sits — which is where reduction effort belongs. Manufacturers routinely find that the hotspot is not where they assumed.
DEISO develops EPDs and the underlying LCA studies, and provides independent pre-submission technical review. DEISO is not a programme operator and does not issue verification statements or certifications; verification is carried out by verifiers approved under the relevant programme.
Environmental product declaration (EPD) development
End-to-end support for EPD development — turning product life cycle results into structured, credible, market-relevant environmental communication.
Explore service →EPD pre-submission review services
Independent pre-submission technical review of EPD-ready documentation — identifies methodological gaps, data issues, structural inconsistencies, and readiness risks before formal submission.
Explore service →Rapid 48–72 hour EPD data readiness review (modules A1–D)
Pre-submission data review and readiness assessment produced within 48–72 hours — identifies missing module data, inconsistent assumptions, weak traceability, and technical gaps across modules A1–D before verification and submission.
Explore service →EPD data collection framework — template A1–D for verification readiness
Structured EPD data collection framework covering modules A1 to D — supports internal organization, stronger data quality, and improved verification readiness across product declaration workflows.
Explore framework →Environmental product declaration (EPD) capacity building
Expert-led EPD capacity building programs built around practical competence in declaration structure, underlying LCA logic, data preparation, interpretation, and ISO-aligned EPD workflows.
Explore capacity building →Request a quotation
Request a tailored quotation for EPD support, pre-submission review, capacity building, or structured technical assistance aligned with your product scope, timeline, and business objectives.
Request a quotation →Why choose DEISO
See why enterprises, suppliers, and public sustainability organizations trust DEISO for EPD development, pre-submission review, capacity building, and verification-ready delivery.
Why choose DEISO →
