Data governance in the battery passport: Annex XIII and access control explained

When an end-user scans the QR code on an EV battery in 2027, they see one subset of data. When an independent recycler scans the same battery, they see another. When a market surveillance authority does it, they see a third. And when a competing manufacturer tries to scan it, they should not see very much at all.

That is not a bug. It is the whole point of Annex XIII of the Battery Regulation, the architectural core of the battery passport's data governance. Annex XIII defines three access layers, and it is the manufacturer's responsibility to ensure that each data field is exposed to exactly the right actors, no more and no less.

For manufacturers, this is where all the compliance work comes together. BMS data from Article 14, disassembly instructions from Article 11, recycled content declarations from Article 8 and the carbon footprint from Article 7 all end up in the same place: in one battery passport, where access control makes the difference between lawful transparency and unintended IP leakage.

Why is access control needed at all?

If the battery passport were just a public PDF, the whole Battery Regulation would have been dead on arrival. EV manufacturers would not accept publishing detailed material compositions that reveal their cathode formulations. Industrial battery manufacturers would not accept publishing individual supplier relationships. And end-users would drown in technical data they cannot use for anything.

At the same time, the circular economy, the whole purpose of the regulation, would collapse if recyclers, repairers and second-life operators could not access the technical data they need to do their work safely and efficiently.

Annex XIII solves the dilemma by making transparency differentiated. Different actors have different legitimate interests in different data fields, and access is granted precisely by role. It is not a compromise, but the core operating model of the whole system.

The three access layers

Annex XIII divides the data fields into three layers, each referenced in points 1, 2, 3 and 4 of the annex:

The unique battery ID is the entry point to the whole system

Access control only works if each battery can be identified unambiguously. The regulation therefore requires each in-scope battery (LMT, EV, industrial >2 kWh) to have a unique battery identifier encoded in a QR code that complies with the ISO/IEC 15459:2015 standard or an equivalent.

The unique ID acts as the entry point to two layers of data:

• Model data (static): Information that applies to the whole battery model: carbon footprint, recycled content, material composition, performance parameters, certifications. This data only changes if the model is revised or a delegated act updates the methodology.
• Unit data (dynamic): Information specific to the individual battery: production batch, date, the quantity of mass balance cobalt allocated, and over time: SoH, cycle count, temperature events, status (in service, repurposed, end of life).

For a recycler scanning an 8-year-old EV battery in 2035, it is the unit data that determines whether the battery can be repurposed for stationary storage or should go straight to material recovery. For an end-user scanning the same battery, only the publicly available model data is relevant.

The architectural consequence is that the battery passport is not one document, but a data service that dynamically composes the right answer based on who is asking and which layer they have access to.

Role verification is the under-described obligation

For Layer 3 to be meaningful, someone must be able to verify that an actor actually has a "legitimate interest". The regulation does not say exactly who, but it does say that access must be controllable, and that the manufacturer is responsible for the passport being correct and up to date.

In practice, this leaves the manufacturer with an architectural challenge. Three approaches are emerging in the industry:

• Centralised role verification by the manufacturer. The manufacturer runs (itself or via a compliance platform) a mechanism that validates the user's role, for example by requiring documentation such as certification as an electrician, authorised repairer or approved recycler. This gives the manufacturer control, but requires ongoing administration.
• Third-party verification through industry certifications. The manufacturer grants access based on established certificates, such as EMAS-registered recyclers, IATF-certified automotive repairers or WEEELABEX-certified WEEE handlers. This reduces administrative overhead, but moves responsibility for quality control elsewhere.
• Federated identity via a national authority. The cleanest long-term solution, but it assumes that the Member States build the necessary infrastructure. This is expected to develop gradually as the implementing act on "legitimate interest" takes shape.

Until the Commission's implementing act takes effect, manufacturers must take responsibility for role verification themselves. This means the compliance platform must not only control what is exposed, but also who is authorised to see it, and how this is documented.

In Syntakk's current model, centralised role verification by the manufacturer is combined with flexible recognition of third-party certifications. In practice, the manufacturer can define which certificates or accreditations automatically give access to Layer 3 (such as EMAS registration for recyclers or national authorisation for electricians), while unclear cases can be handled manually with a full audit trail. When the Commission's implementing act on "legitimate interest" is available, the model can be adapted to the new definition without reconfiguring the whole platform. That is exactly what a flexible access architecture should be able to deliver.

Business-critical information is protected

One of the most frequent objections from manufacturers is: "We cannot publish our material composition." The objection is often based on a misunderstanding of what the battery passport actually requires.

The regulation explicitly distinguishes between the information that must be publicly available (Layer 1) and the information only available to actors with a legitimate interest (Layer 3). The publicly available composition is at a general level, typically the main materials and the chemistry category. The detailed composition down to the level of the cathode formulation is reserved for Layer 3, where it is critical for safe disassembly and recycling.

In addition, there is explicit protection: commercially sensitive information and IP-protected formulations are exempt from disclosure, even within Layer 3, if disclosure would harm the manufacturer's legitimate business interests without serving the purpose of the circular economy.

For manufacturers, this means the compliance platform must be able to handle three dimensions at once:

• Which layer does the recipient have access to (public, authority, legitimate interest)?
• Which data fields fall under which layer?
• Are there specific fields that are additionally IP-protected and only available under special conditions?

This is three-dimensional access control, not a simple public/private distinction.

Where are the compliance traps?

Three mistakes recur among manufacturers that have just started addressing data governance in the battery passport:

Trap 1: Treating the battery passport as one public document. If the whole data set is exposed to every QR scan, the manufacturer leaks both IP and business data. If the opposite happens, and only public data is exposed, the manufacturer fails to support the circular economy as the regulation requires. Both extremes are compliance failures.

Trap 2: Underestimating role verification. A "legitimate interest" is not a declaration of trust from the user themselves. It is a verifiable professional role, and the manufacturer must be able to document how the verification was done, especially if doubts later arise about whether a specific actor really had a legitimate interest.

Trap 3: Separating static and dynamic data into silos. Model data and unit data have different update frequencies, but they are delivered in the same place: through the battery passport's data service. An architecture that keeps them physically separate without a common access layer will struggle to deliver consistent answers over the battery's 10-15 year life.

Why data governance is the architectural core

Access control is not a separate compliance burden. It is the layer that makes all the other burdens sustainable.

Take the recycled content declaration: the figure "16% recycled cobalt" is public (Layer 1), but the underlying mass balance accounting that documents the claim is only for notified bodies (Layer 2), and the supplier relationships are protected by the IP exemption.

Take the disassembly instructions: they are summarised publicly in Layer 1 (category-level information), but the detailed sequences with exploded views are in Layer 3, because an unauthorised person should not be able to disassemble a high-voltage battery.

Take BMS data and State of Health: the SoH value for an individual battery is in Layer 3 and only for actors with a legitimate interest, because it reveals individual user behaviour. Aggregated model data on typical SoH development can be in Layer 1 or Layer 2 depending on sensitivity.

Take the carbon footprint: the declared value and performance class are public (Layer 1), the calculation and the CFF parameters are for notified bodies (Layer 2), and the underlying supplier emissions data is IP-protected even in Layer 3.

Each of the four compliance areas has its own layering. It is data governance that makes it possible to bring them together in one coherent battery passport without leaking anything that should not get out.

What you can do now

For manufacturers of in-scope batteries, four steps are worth taking already in 2026:

• Map your data points to the Annex XIII layers. For each data field in your compliance portfolio: which layer does it belong to? Is it public, restricted to authorities or for legitimate interest? Is there specific IP protection?
• Set up a role verification model for Layer 3 access. Which certifications or documentation do you accept as a "legitimate interest"? How are access requests logged and audited?
• Do not separate static and dynamic data into silos. Model data and unit data must be delivered through the same access layer, even if they technically live in different systems.
• Prepare for the Commission's implementing act. The definition of "legitimate interest" was scheduled to be clarified by 18 August 2026 at the latest. Manufacturers with a flexible access model that can be adapted will be able to update without major architectural changes.

Data governance is the scaffolding that carries all the other compliance burdens. The difference between a battery passport that works and one that leaks IP or fails an audit does not lie in the data fields themselves, but in the layer that decides who sees what, and when.

One platform or five silos

When you look at the whole compliance picture of the Battery Regulation together (due diligence, BMS, removability, recycled content, carbon footprint and the access control that ties them together), one thing becomes clear: the five areas cannot be handled in silos.

Due diligence data from the supply chain is input to recycled content verification. Recycled content data is input to the carbon footprint via the Circular Footprint Formula. The carbon footprint and recycled content are public in Layer 1 but have private layers underneath. Disassembly instructions are in Layer 3 but must be linked to the publicly available model identifiers. BMS data flows continuously and must fit into the same access structure.

A central compliance platform, a single source of truth, is in practice the prerequisite for making this feasible. Not because it is tidy to have data in one place, but because access control only works if it is consistent across all data points. A battery passport where the recycled content declaration lives in one place and the disassembly instructions in another, with separate access control in each, will sooner or later fail on one of the layers.

The manufacturer that builds compliance as five separate projects will discover that integration between them is the most expensive part. The manufacturer that starts with data governance and lets the compliance areas plug into the same architecture has a model that is both cheaper to maintain and easier to verify.