Manufacturing differs from service sectors on a point that has direct consequences for sustainability data: the largest part of the environmental impact is not located within the company's own premises, but in the chain before and after it. Raw materials, semi-finished products, energy-intensive processing steps, transport between factories, waste streams on the shop floor — the organization's own operation is often only a small part of the total picture. This means that an organization in manufacturing depends, for a substantial part of its reporting obligation, on data that originates elsewhere: at a supplier, a carrier, an energy provider. That dependency is similar in pattern to what occurs in the agricultural sector, where the chain before and after the company itself also determines a large part of the impact.
Within the company's own walls, sustainability data is spread across systems that were not designed for this purpose. Energy consumption per production line is recorded in the building management system or with the energy provider, not linked to production volumes. Machine run times and downtime are recorded in the MES or in an Excel overview kept by the shift supervisor. Waste streams — scrap, residual material, chemical waste — are often registered per weight ticket by the waste processor, separate from the internal system. Water consumption is sometimes measured centrally for the entire site, without breakdown by process. For someone who needs an emissions figure per production line, this means a search across measurement systems that each have their own logic and frequency.
The ERP system contains purchasing volumes, material flows and sometimes supplier data, but rarely emission factors. The MES knows machine times and output, but does not automatically link that to energy or material consumption per unit of product. Between these two systems there is often a manual step: someone who pulls volumes from the ERP, combines machine data from the MES, and arrives at an emissions figure using a separate calculation sheet. That calculation sheet — and the assumptions in it — is precisely the point where traceability disappears if no one records which source, which factor and which assumption was used.
For scope 3, manufacturing is heavily dependent on what suppliers provide. One supplier sends an extensive sustainability report, another an estimate on an invoice, a third nothing at all. Some companies work with generic industry averages because supplier data is missing; others have primary data for a few large suppliers, but not for the rest of the chain. This mix of data sources — primary, secondary, estimated — must be recorded somewhere with a note on its origin, otherwise it is no longer possible to reconstruct afterwards which figure is based on which assumption.
Many manufacturing companies already have certifications such as ISO 14001 or energy audits in place. These processes generate valuable data — energy consumption, improvement plans, verified figures — but the outcomes often end up in a separate archive of the quality department, disconnected from sustainability reporting. The figure that the auditor has already verified is then collected again, separately, for the sustainability report, with the risk that the two figures no longer match.
Who is responsible for the energy figure of a production site: the plant manager, the facilities manager, or the sustainability coordinator at headquarters? Who manages the emission factors applied to transport data: purchasing, logistics, or finance? In many manufacturing companies, that responsibility has never been explicitly assigned. Data ownership arises implicitly, with whoever happens to have access to the system, not with whoever knows or can verify the content best.
The temptation is great to buy a software package that automatically connects with ERP and MES and produces neat dashboards. But a tool built on an unorganized foundation mainly produces faster and more convincing-looking versions of the same uncertain figures. Before a system can retrieve data, it must be established which data point comes from which source, who owns it and which quality rule applies to it. This is just as true for manufacturing as it is for sectors with a completely different chain structure, such as the transport sector or retail, where the data sources are different but the organizational problem is the same.
The Data Readiness Scan builds a data point register for manufacturing: per emission source, per production line, per chain link, it establishes which data point is needed, where it originates — machine, ERP, energy provider, waste processor, supplier — and via which route it reaches the final reporting point. This is accompanied by an owner per data point and a quality rule that determines when a figure is usable and when it is not. This is not a report and not a completed questionnaire, but the underlying structure that makes it possible to trace a figure back to its source at any moment. This structure is comparable to what is needed in professional services, where data is likewise spread across separate systems and spreadsheets, even though the chain looks different there.
The tool that performs this scan is under development. Those who want to prepare their organization for this can sign up for the waiting list.
Once it is clear which data points exist, where they originate and who manages them, it also becomes visible how much manual work goes into collecting, checking and retyping that data. That work — retrieving data from multiple systems, tracing figures back to a source, performing quality checks — consists of separate tasks that do not all require the same amount of human effort. FTE TO AI's work scan calculates, per task, what portion of it can be taken over by AI, making it clear where automation yields the most time savings and where human oversight remains necessary.
Vraag maar waar een datapunt vandaan komt. Dat is meestal de hele vraag.
Answers come from this site’s knowledge base. Not tailored advice, and not a scan of your company.