Product Carbon Footprint: How to Calculate It Using Machine Data

To calculate a product’s carbon footprint, your company needs to move beyond facility-wide averages and analyze emissions at the process level. Data from machines, meters, production orders, quality systems, and material consumption can provide the greatest value.

This article explains what data to collect, how to connect it, and where to start so that the result supports manufacturing, ESG reporting, sales, and customer conversations.

This information can be especially useful when a customer asks about the carbon footprint of a specific product and an answer based on the facility’s average energy consumption is no longer sufficient.

A product carbon footprint represents the greenhouse gas emissions assigned to a specific product. ISO 14067 provides requirements and guidelines for calculating and reporting a product carbon footprint in connection with life cycle assessment.

Before you begin collecting data, you need to define the scope. Without a clearly defined scope, the result will be difficult to compare, explain, and defend.

Start by determining three things:

These decisions have a significant impact on the data. A batch produced after a long shutdown may consume more energy than a batch produced during a normal shift. A product that requires rework will also have a different result than one that moves through the process without corrections.

When these situations are combined into a single average, you may lose important information about where costs and emissions are actually generated.

A good starting point is one product or one product family. Ideally, choose one with high energy consumption, high sales volume, or increasing interest from customers asking for environmental data.

The GHG Protocol Product Standard recommends collecting primary data for processes under your company’s control. In a manufacturing facility, this usually includes shop-floor data such as energy consumption, operating time, fuel use, materials, waste, process parameters, and quality information.

A machine does not know which customer the production run is for. An energy meter records electricity consumption in kilowatt-hours but does not know the production order number. An ERP system contains product and batch information, but it typically has limited visibility into the actual operation of the production line.

Calculating a product carbon footprint therefore requires technical data to be connected with production data.

The most useful dataset typically includes the following:

Identification and traceability can be a major data gap. Machine data must be connected to time, production orders, products, and assets.

Without these connections, energy data remains assigned to the production line, product data remains in the ERP system, and the product carbon footprint is calculated by simply dividing the facility’s electricity bill by the number of units produced.

Energy is usually one of the first factors analyzed because it can be directly connected to both costs and emissions.

Eurostat reported that in 2024, industry in the European Union consumed 8,835 petajoules of energy. Electricity accounted for 33.3% of consumption, while natural gas accounted for 31.9%.

For a manufacturing facility, this context clearly shows why production energy consumption should be measured closer to the process rather than only at the utility-bill level.

An electricity bill provides total consumption for the entire facility. A department-level meter narrows the data to a specific area. A production-line submeter or machine data connected to a production order makes it possible to allocate consumption to a specific product.

A production line manufactured Product A for eight hours, during which the energy meter recorded 480 kWh. An additional 120 kWh for compressed air was allocated to the process according to an agreed methodology.

The line produced 9,000 good units and 1,000 rejected units.

Calculation based on good units:

600 kWh / 9,000 good units = 0.067 kWh per good unit

If the company divided the energy consumption by all manufactured units, the result would be 0.060 kWh per unit.

The difference depends on whether rejected units are allocated to the products that are ultimately sold. This rule must be documented because it will affect the results of future batches.

An appropriate emission factor is required to convert energy consumption into greenhouse gas emissions. In Poland, electricity emission factors are published by KOBiZE. For reporting purposes, the company should record the year of the factor, the unit used, and the person responsible for approving its application.

The quality of an ESG report depends on data collected earlier from meters, machines, production orders, waste records, material purchases, and utility invoices.

The Corporate Sustainability Reporting Directive introduced reporting in accordance with the European Sustainability Reporting Standards for companies within its scope. The European Commission states that the first companies applied the new rules to fiscal year 2024, with reports published in 2025.

The scope and implementation schedule were later subject to regulatory simplification. Each company should therefore verify its current status based on its size, industry, and role in the value chain.

For manufacturers supplying larger companies, ESG requirements often arrive before a direct legal obligation applies. Customers may request information about a product’s environmental performance, recycled material content, production energy consumption, or the sources of the emission factors used.

The Polish Agency for Enterprise Development, or PARP, notes that ESG requirements may extend to small and medium-sized enterprises through the value chains of larger companies.

Common customer questions include:

An average may be sufficient for an initial questionnaire. During a bid, audit, or discussion with a major customer, however, an average may not be detailed enough.

It is equally important to demonstrate where the number came from.

A Digital Product Passport, or DPP, is a digital collection of information about a product, component, or material.

The European Commission describes it as a tool designed to improve access to information about product sustainability, durability, repairability, material composition, and circularity.

This is no longer a distant regulatory topic. The DPP registry became operational on July 20, 2026, and a passport is expected to become mandatory for selected battery categories beginning February 18, 2027. Additional product groups will be covered by separate requirements.

For manufacturers, this increases the importance of data that is already necessary for calculating a product carbon footprint, including:

A company that organizes its data for product carbon footprint calculations is also building a foundation for future regulatory requirements and customer requests.

There is no need to create separate records for ESG, energy management, quality, and Digital Product Passports. A better approach is to create one shared data model that can be used by multiple departments.

Machine data analysis does not have to begin with a large, facility-wide system. First, determine whether you can establish which machine generated the consumption data, during what period, and for which production order.

When the answer is incomplete, the product carbon footprint will depend on assumptions. In some cases, that may be acceptable, but the calculation must clearly distinguish measured values from estimated values.

A practical approach includes five steps.

Data from the energy meter must be assigned to the time period during which a specific production order was processed.

Production, downtime, failure, warm-up, and changeover have different implications for allocating costs and emissions.

Rejected units, rework, and waste must be allocated according to a documented methodology. Without this rule, the kilowatt-hour-per-unit indicator may appear better than it actually is.

Compressed air, gas, steam, process cooling, and water can significantly affect the result, especially in energy-intensive processes.

For each value, document the source system, measurement frequency, scope, and the person responsible for the methodology.

This dataset supports more than emission calculations. It also shows which batches consumed more energy and helps explain why.

The data required to track production and its environmental impact is usually distributed across multiple systems.

ERP stores information about products, batches, and production routings. MES shows how production actually took place. PLCs, SCADA systems, and meters provide data directly from machines.

Spreadsheets often contain information that has not yet been automated, such as waste quantities, operator comments, allocation of shared utilities, or supplier data.

Problems arise when every department uses a different version of the data.

Production analyzes unit counts and downtime. The ESG team focuses on emissions. Controlling looks at costs. Quality focuses on defects. Sales needs information that can be provided to the customer.

When the data is not connected by a shared identifier, every analysis requires employees to manually review files and determine which version is correct.

A shared data model should tell you:

This is where collecting machine data and connecting it to production systems becomes particularly valuable.

The best approach is to begin with one machine or one section of a production line. The data can then be presented in reports and dashboards for production, quality, maintenance, and planning teams.

The best starting point has a limited scope and a measurable outcome. Select one production line, one product, or one process where energy consumption and production losses have a meaningful impact.

At the beginning, check whether:

When one of these elements is missing, do not stop the project. Record the data gap and decide whether a clearly documented estimation method can be used during the pilot.

The key is to avoid combining measurements and assumptions without labeling them.

A good pilot project should produce three results.

This is a metric that production, controlling, and ESG teams can all understand.

This reveals the impact of changeovers, production losses, operating speed, quality, and process settings.

After the pilot, the company will know which machine signals should be collected automatically and which information can remain a controlled operator entry.

Do not begin by mapping the entire facility. Begin with a process that can produce one useful and credible number.

A well-calculated product carbon footprint can support several departments and business areas.

Instead of offering a general statement, the company can provide a result, define the calculation boundaries, and identify the data sources.

Comparing batches of the same product often reveals the impact of startup, defects, changeovers, or the operation of supporting equipment.

ESG reporting in a manufacturing company becomes less dependent on manually collecting information from multiple spreadsheets.

The best starting points are processes where a high carbon footprint is connected to high energy costs, excessive waste, or low process consistency.

The company does not need to begin with a complete life cycle assessment for its entire product portfolio.

The first meaningful result from one production line can show whether the company has the data, methodology, and people needed to maintain the process.

A product carbon footprint stops being only a reporting topic when a customer asks about a specific product.

At that point, what matters is whether your company can present process-level data covering energy, materials, waste, operating time, quality, and the calculation methodology.

The best starting point is a simple data chain:

product → production order → machine → energy → material → waste → emission factor → result

This structure can help your company calculate product carbon footprints, reduce manufacturing energy consumption, answer customer questions, prepare ESG reports, and collect information required for Digital Product Passports.

The most practical first step is a focused data review for one production line. Determine which signals are already available, which can be collected directly from machines, and which are currently stored in spreadsheets.

This type of assessment quickly shows whether the company is ready to calculate product carbon footprints using process data or whether it first needs to improve energy measurement, quality data, and production-order identification.

A product carbon footprint is the total greenhouse gas emissions assigned to a product within defined calculation boundaries. These boundaries may cover manufacturing only, the process from raw material extraction to the factory gate, or the product’s entire life cycle.

The required data usually includes energy consumption, machine operating time, production order number, number of good and rejected units, waste, material consumption, supporting utilities, and process parameters.

The data should be connected to a specific product, batch, or production order.

Not for an accurate product-level result.

An electricity bill shows energy consumption for the entire facility. Calculating a product carbon footprint requires the energy to be assigned to a process, production line, machine, batch, or production order.

ESG reporting requires information about energy, emissions, materials, and waste.

In manufacturing, much of this information is generated on the shop floor through meters, machines, MES, ERP, quality reports, and warehouse data.

In many areas, yes.

A Digital Product Passport is intended to include information about the product, materials, durability, repairability, recycling, and environmental performance. Data about energy, raw materials, waste, and recycled content will therefore become increasingly important.

Not always.

A company can start with data from meters, machines, ERP, and a structured report for one production line.

However, an MES makes it easier to connect production orders, operating time, quality, energy consumption, and production output. At a larger scale, it can significantly reduce the effort required to obtain reliable data.