The Data Architecture Behind Digital Product Passports | Carbmee
Four Systems at the Core of DPP Infrastructure
One of the most common misconceptions about Digital Product Passports (DPPs) is that they are primarily a data collection challenge. In reality, the greater challenge is architectural: understanding where product data already lives, which systems own it, and how those systems can be connected to produce a reliable, dynamic product record.
Before collecting a single data point, organizations must map their data model and build the infrastructure that will make DPPs both compliant and scalable.
Most of the data required for a DPP already exists somewhere within an organization's systems. The challenge is that it is typically scattered across silos, owned by different departments, and stored in formats that cannot easily communicate with each other. A functioning DPP infrastructure connects four core systems.
The first is the Product Lifecycle Management (PLM) system. The PLM contains the bill of materials, material composition, and component descriptions for every product. For a simple garment, this may be a short list of inputs; for a complex product like a smartphone or automotive component, it may involve hundreds or thousands of entries. The PLM is the primary source of information about what a product is made of.
The second is the Enterprise Resource Planning (ERP) system. Where the PLM describes the product, the ERP records the business transactions surrounding it: who sourced which materials, when, from which supplier, and who verified the data. The ERP establishes the chain of custody that gives DPP data its traceability.
The third is the supplier portal. Unlike an EPD, which is updated once every several years, a DPP is meant to be a dynamic document. When a supplier switches to renewable energy, increases its use of recycled content, or changes a material input, that update should flow through to the DPP. A connected supplier portal makes this possible by giving suppliers a structured channel to submit and update their data continuously.
The fourth system is the LCA and PCF calculation engine. Environmental metrics, particularly carbon footprint at the SKU or batch level, are a core element of the DPP's environmental and circularity data layer. The carbon footprint engine produces these calculations and, when connected to the other three systems, allows them to update dynamically as upstream data changes.
Four Layers of DPP Data
The data that flows through these systems populates four distinct layers within the DPP itself. The identifier layer establishes which product or batch the passport belongs to and its relevant product category. The environmental and circularity metrics layer captures lifecycle performance indicators, with carbon footprint as the most certain requirement. The compliance layer documents certifications, hazardous substance declarations, and recycled content claims, each traceable to its verifying authority. The lifecycle and traceability layer records where the product came from, where it is going, and how its data has been updated over time.
Together, these four layers give different stakeholders − regulators, customers, recyclers, procurement teams − access to the information relevant to their role.
Data Quality: Dynamic, Not Static
A critical behavioral shift required by the DPP model is moving from periodic reporting to continuous data management. EPDs are typically generated once and uploaded to an EPD library, where they remain unchanged for up to five years. A DPP infrastructure, by contrast, is designed for a continuous workflow of data updates.
This distinction has real implications for product competitiveness. Using average emissions figures − as may sometimes be required under DPP guidance from the Joint Research Center − means organizations cannot differentiate themselves from competitors. Companies that invest in primary, supplier-specific data will be able to demonstrate genuine product performance rather than sector averages.
Connecting the Systems: Minimum Requirements
Organizations do not need a fully mature IT landscape to begin building DPP infrastructure. The minimum viable system architecture requires: a product identifier and definition in the ERP, a bill of materials with material composition in the PLM, a supplier portal connected to both, and an LCA engine that can calculate and update environmental metrics as data flows in.
Starting with this foundation allows organizations to build toward a scalable DPP program without waiting for perfect data or a perfect system before taking action.
