Using primary data often reduces module A1 (raw material supply) of an EPD. For products like concrete or windows, this is ususally the largest impact driver.
Generic data is conservative by design. LCA datasets carry safety margins between 10 and 30 % in databases like the ÖKOBAUDAT and the Dutch NMD.
Building caps, corporate targets (SBTi) and building certifications (like BREEAM) drive manufacturers to improve their EPD values through primary data.
Many manufacturers create Environmental Product Declarations (EPDs) because their customers ask for them. For a growing number of those customers, the EPD is not a checkbox. The results are read, compared and used in purchasing decisions. When your EPD sits next to a competitor's, the numbers in it should describe your product, not an industry average.
In most EPDs, a large share of the impact sits in the raw materials. That part of the result depends largely on your suppliers, and on which data you use to model them. This article explains the two data types used in a life cycle assessment (ISO 14040/14044), how primary data from suppliers improves accuracy compared to secondary data, and why that matters commercially.
1. Primary vs. Secondary Data in an LCA
An LCA needs input data for the whole life cycle: part of it you measure in your own plant (electricity mix, product composition, waste, transport), part of it comes from your suppliers. There are two types of input data that manufacturers can work with: primary and secondary data. Each of them are explained below, with a summary in this table:
Primary data
Secondary data
Source
Measured at the facility where the process takes place: your plant or your supplier’s plant
Third-party life cycle inventory database, e.g. ecoinvent
Represents
The material you actually buy and the process you actually run
A typical industry average for a material type and region
Verification
Supplier figures usually need a third-party verified EPD
Datasets from established LCI databases are accepted
1.1 Primary Data (Specific Data)
Primary or specific data is directly measured at the facility where the activity takes place. For your own processes, that is data you collect internally. For raw materials, it is data collected by your suppliers at their facility.
For a verified Product Carbon Footprint (ISO 14067), LCA or EPD, that supplier data needs to hold up in verification. Verifiers will question unverified supplier figures and often require a third-party verified EPD (ISO 14025).
Most verified EPDs are publicly accessible. Emidat Atlas, our free public database for searching and comparing verified EPDs, and ECO Platform's ECO Portal are good places to start (read our blog post on where to find an EPD). Many raw material producers have already published EPDs you can use directly, the cement producer Schwenk Zement for example (as seen below).
Not every public supplier EPD can be directly linked into your own calculation. It needs to follow EN 15804+A2, declare a comparable declared unit and still be valid (EPDs expire after five years). In the Emidat platform, this eligibility check runs automatically and you can link all valid EPDs directly. If a supplier has not published an EPD yet, you need to request the data.
1.2 Secondary Data (Generic Data)
Secondary or generic data comes from a third-party life cycle inventory database. It represents a typical industry average for a material type, not the specific material delivered to your factory. Database providers build them from industry-wide data and update them regularly.
One of the most widely used databases for building and construction LCAs is ecoinvent. Other options include Sphera's Managed LCA Content (formerly GaBi) and material-specific databases such as Carbon Minds or the Plastics Europe Eco-profiles.
Because LCA results become part of the Digital Product Passport (DPP) under the new Construction Products Regulation (CPR), the EU Commission is also working on generic data: it provides the characterisation factors for the calculation (Regulation (EU) 2024/3110, Recital 7) and has thus started with the development of its own background database for construction products.
2. The Impact of Primary Data on an LCA
An EPD reports results per life cycle module. The production stage covers modules A1 to A3: A1 is the supply of raw materials, A2 their transport to your plant, and A3 your own manufacturing. Using primary data from suppliers mainly has an impact on module A1, since EN15804+A2 requires you to use your specific data for the calculation of module A3. How much A1 weighs in your final EPD result depends on the product. For manufacturers that process purchased materials, such as concrete, mortar or windows, A1 is typically the largest module and using supplier data can improve their final EPD results.
There are two reasons why the specific result is often lower. First, generic data (such as ecoinvent) is built conservatively. ÖKOBAUDAT, the German database for environmental data of building products, adds a safety margin between 10 % and 30 % to generic datasets. The Dutch database Nationale Milieudatabase (NMD) applies a 30 % markup to generic data. A supplier that merely matches the industry average in its own EPD therefore already beats the generic value, and an own EPD built on generic data carries higher values than it needs to.
EPD visual: generic vs. supplier data (en)
Second, generic data hides your supplier's own sustainability efforts. Similar products from different suppliers carry the same impact, no matter what those suppliers have done. The actual impacts can vary a lot, depending on recipe and process. For cement, for example, clinker content, fuel mix and electricity mix make a large difference. Primary data reflects your supplier's actual production, for a single plant or as an average across its plants.
A supplier can also be worse than average, and primary data will show that too. That is still worth knowing: it tells you what your supply chain actually contributes, and it gives you a basis for improving it, whether by working with the supplier or by changing it.
3. Three Market Drivers That Reward Lower EPD Results
Three market drivers reward lower EPD results. Which one matters most depends on where you sell: national emission caps in Denmark, France and the Netherlands today and in the rest of the Nordics next, scope 3 requirements wherever you supply large corporates, and building certifications in almost every market.
Market Drivers That Reward Lower EPD Results
Building emission caps
Corporate scope 3 targets
Building certifications
3.1 Building Emission Caps
Building emission caps limit the CO2 emissions of the whole building, including its materials.
National emission caps
Caps are in force or being introduced mainly in the Nordics (Denmark, Sweden, Finland, Norway), in France and in the Netherlands. National building regulation requires buildings to stay below a defined limit per square metre, depending on the building type. Denmark and France cap CO2e; the Netherlands sets a shadow-price limit in euros across several environmental indicators.
Project developers in these markets already run whole-building LCAs (EN 15978). As the caps tighten, they select materials based on verified product data. Every kilogram of CO2 a product saves in its EPD is headroom for the developer. A product modelled with generic data does not show this advantage.
Energy Performance of Buildings Directive (EPBD)
The EPBD requires the calculation and disclosure of whole-life carbon for new buildings above 1,000 m² from 2028 and for all new buildings from 2030, and obliges every member state to set national limit values that tighten over time. To assess the contribution of each material, the calculation is expected to draw on the environmental data in Digital Product Passports required under the CPR.
3.2 Corporate Scope 3 Targets
More than 10,000 companies have validated emission targets through the Science Based Targets initiative. For most of them, meeting those targets means reducing scope 3 emissions, which means requirements for purchased materials and for the construction of their own buildings. If you supply such a company or one of its construction projects, they will ask for accurate emission data on your products and assess your values.
Data centre projects show this already. Amazon has written EPD-based embodied carbon tracking into its construction contracts and states that manufacturers are rewarded for using supply-chain-specific primary data instead of averages. Microsoft uses third-party verified EPDs to steer material choices in its data centres.
Accurate data lets your customer calculate their scope 3 emissions correctly. Once they enforce CO2 limits on purchased materials, the supplier who can show a lower result with verified supplier data wins.
3.3 Building Certifications (BREEAM, LEED, DGNB)
Certification schemes such as BREEAM, LEED and DGNB require a building LCA compared against embodied carbon benchmarks to reach the higher ratings. BREEAM has tightened these requirements in its latest version, BREEAM V7, and awards points for product-specific EPDs.
The more accurate the product data, the better a developer's building performs against those benchmarks and the higher the rating it can achieve. Here, too, only supplier-specific data lets your product beat the category average.
4. How to Bring Supplier Data into Your EPD
If you already have EPDs, start at step 2.
1. Model the LCA with the data that is available
Use primary data from suppliers who already have an EPD, and secondary data where none exists. In the Emidat platform, you can search all available supplier EPDs and link them directly to your calculation.
2. Identify hotspots and request supplier data
With a first LCA in place, you see where most of the impact comes from. For every material that contributes a large share and has no EPD yet, request primary data from the supplier. A handful of materials usually accounts for most of module A1, so focus there. In the Emidat platform, you can automate these requests. If you are still collecting your own plant data, our data collection guide covers that part.
3. Replace secondary data with supplier data
Once the supplier EPDs are linked, replace the generic datasets in A1 with them and submit the updated results for verification. With the Emidat platform, verification takes one to two weeks.
The result is an EPD that shows what your product actually is, not what the industry average happens to be. In a market that is starting to compare, that is the difference that wins deals.
FAQ
What is the difference between primary and secondary data in an LCA?
Primary (specific) data is measured at the facility where the process takes place: your own plant for production, your suppliers' plants for raw materials. Secondary (generic) data comes from a life cycle inventory database such as ecoinvent and represents an industry average for a product category in a region. Primary data describes the material you actually buy; secondary data describes a typical one.
What is module A1 in an EPD?
EPDs report results per life cycle module. A1 covers the supply of raw materials, A2 their transport to your plant, A3 your own manufacturing. Together they form the production stage (cradle to gate). For products made from purchased materials, such as concrete or insulation boards, A1 is usually the largest module, so the data used for A1 has the biggest effect on the final result.
Can I use my supplier's EPD in my own EPD?
Yes, if it is eligible. The supplier EPD must follow EN 15804+A2, use the same characterisation factors as your LCA, declare modules A1 to A3 for a comparable declared unit, and still be valid. EPDs expire after five years, and EPDs under the older +A1 version cannot be linked. The Emidat platform checks these criteria automatically.
Does supplier data always lower my EPD result?
No. Primary data makes the result accurate, not automatically lower. If a supplier performs better than the industry average, your result drops. If it performs worse, your result rises. In both cases you learn where your hotspots are before a customer does.
