The EU battery regulation contains three obligations that companies routinely merge into one project, and the third one behaves differently from the other two. The passport is a product record with a fixed date. Due diligence is a company policy with a fixed date. The carbon footprint declaration under Article 7 is a calculated figure whose timing depends on Commission acts that set the method and the format — which makes it the obligation where a confidently quoted date is the thing to be suspicious of.
Three obligations, one regulation
| Carbon footprint | Battery passport | Due diligence policy | |
|---|---|---|---|
| Article | 7 | 77 | 48 |
| What it is | A calculated figure with a declared methodology | A product data record | A company-level sourcing policy |
| Applies from | Applies 12 months (EV) or 18 months (industrial, LMT) after both the methodology delegated act and the format implementing act enter into force, or later where Art. 7(1) fixes a later date — neither act adopted as of 7 Oct 2026 | ||
| Verified by | Notified body: Module D1 (series) or Module G (non-series), Art. 17(2); checks the footprint data and the method's application | — | Independent third party |
| Where the data comes from | Your cell supplier's plant | Your own product records, plus suppliers | Four raw material chains |
Two of these three rows carry a date you can put in a plan. The first does not, and the reason is structural rather than administrative.
Why the date behaves differently
For the passport, the regulation names a day and that day arrives. For the carbon footprint, the regulation creates the obligation but leaves two things to be filled in: how the number is calculated, which comes in a delegated act on methodology, and what the declaration looks like, which comes in an implementing act on format.
That dependency is not a technicality. A carbon footprint figure without a prescribed methodology is not comparable to anyone else's, and a regulation that required declarations before defining the method would be requiring incomparable numbers. So the obligation waits for its method.
The practical consequence for a reader is unusual and worth stating plainly: the status of those acts, not the text of Article 7, determines when you are obliged. A date you saw in a slide deck, a vendor's timeline or a summary written a year ago may have been accurate when written and wrong now, in either direction. Before planning against any date for this obligation — including any you find on this page — check the current status of the methodology and format acts on the Commission's initiative pages. That is a two-minute check that a great deal of published guidance skips.
What the declaration actually asks of you
The scope of the calculation is what makes this expensive, and it is worth understanding even while the method is unsettled.
A carbon footprint under this regime is not a company average and not a product category estimate. It is per battery model and per manufacturing plant. Two identical packs from two plants are two numbers, because the plants draw different electricity.
For most companies reading this — pack assemblers, e-bike makers, equipment manufacturers who buy cells — the dominant share of that number is created before the battery reaches them. Cell manufacturing is energy-intensive and it happens at a supplier's site. Your footprint is largely your supplier's footprint, which means your ability to declare depends on your ability to obtain plant-level process and energy data from a company that has no direct obligation to you.
That is the same shape as the due diligence problem and the same shape as the passport problem: the obliged party is downstream of the data.
What the life-cycle boundary covers
Annex II, point 4, sets the boundary of the carbon footprint calculation: it runs from raw material acquisition to end of life and recycling, and the use phase is excluded. Four stages sit inside that boundary, and they are not equally hard to fill.
Raw material acquisition and pre-processing. Mining and other sourcing, pre-processing and transport of active materials, up to the manufacture of battery cells, battery components and electric or electronic components. This is the deepest part of the chain and the part where most companies fall back on database averages.
Main product production. Assembly of the battery cells, and assembly of the battery from those cells and its electric or electronic components. Cell assembly happens at the cell maker's plant; pack assembly is the part an assembler controls and measures directly.
Distribution. Transport of the battery to the point of sale.
End of life and recycling. Collection, dismantling and recycling.
The distribution of effort is the useful insight here. The stages you can measure best are not where Annex II expects the largest contributions, and the materials and components it singles out sit behind supplier relationships. That asymmetry is why the work is contractual: no amount of internal measurement rigour compensates for a cell supplier that will not disclose plant-level energy data.
It also explains why two credible-looking figures for similar batteries can differ widely. If one is built on plant-level primary data and the other on regional database averages, they are not comparable — and until a prescribed methodology fixes which sources are acceptable, both can be defensible on their own terms. That is precisely the problem a methodology delegated act exists to solve, and the reason the obligation waits for it.
Walkthrough: a pack assembler with two cell sources
A representative scenario. A company assembles battery packs for industrial equipment, buys cells from two manufacturers, and sells into the EU.
Step one — find the plants, not the suppliers. The contracts name two companies. The footprint depends on which plants those companies use for the specific cell types, and one of the two routinely allocates production across three sites. Until that is pinned down, no figure exists — not a wrong figure, none at all.
Step two — read the contract, not the regulation. Neither supply agreement obliges the cell manufacturer to disclose plant-level energy data. One includes a general sustainability cooperation clause that is not specific enough to compel anything. This is the actual blocker, and it moves at contract-renewal speed.
Step three — build the parts that survive a methodology change. A bill of materials mapped to plants, a data request format, and a named owner. None of that depends on how the delegated act ultimately treats electricity contracting instruments — the question that has made the methodology contentious. The calculation is method-dependent; the plumbing is not.
The counterfactual. Suppose the company waits for the delegated act before starting anything, on the reasonable-sounding grounds that it cannot calculate what has not been defined. When the act arrives, it starts a contract renegotiation with a cell supplier, which takes a renewal cycle — and the declaration deadline runs from the act, not from the renegotiation. The methodology was never the long pole. The contract was.
Edge cases
Batteries inside products. As with the rest of the regulation, placing a product with a battery on the market places the battery on the market. The obligation follows the battery.
Two plants, one model. Allocation across sites is a real practice and it splits one product into several declarations. Establish it early; it changes what you have to collect.
Recycled content and footprint are different questions. Recycled input affects the number. It does not answer a separate recycled-content requirement, and the two are documented differently.
A passport that carries a footprint field. The passport can display the value, which is why the two are conflated. Populating a field is not producing a verified declaration.
Estimates versus verifiable data. Early figures built on industry averages will not survive a methodology that requires plant-level data. Building on averages is fine as a rehearsal, provided nobody mistakes the rehearsal for the deliverable.
Zoom out: one data project, three obligations
The three battery obligations arrive from different articles, on different dates, in different formats. They ask for overlapping data about the same batteries from the same suppliers. Treating them as three projects produces three questionnaires to the same cell manufacturer and three separate contract conversations.
Treating them as one produces a single supplier data programme: which plants, which materials, which chains, with which contractual right to ask. Our battery passport guide covers the obligation with the nearest fixed date, and our battery due diligence guide covers the one six months after it. The carbon footprint is the third call on the same relationship — and the one where the useful preparation is contractual rather than analytical.
