Search every question across sub-FAQs

Data Act Article 25 requires a minimum retrieval period of at least 30 calendar days after the agreed transitional period ends. The contract also needs a clause guaranteeing full erasure of exportable data and digital assets generated directly by, or directly relating to, the customer after the retrieval period expires or after a later agreed period, provided switching has completed successfully.

Retrieval and erasure should be operationally distinct. Retrieval terms should say how the customer accesses the data and digital assets after transition, while erasure terms should say what gets erased, when erasure starts, which evidence is generated, and which legally retained or non-exportable material is outside the erasure commitment.

Data Act Article 25 requires the contract to state when it is considered terminated and when the customer is notified. Termination occurs, where applicable, upon successful completion of the switching process, or at the end of the maximum notice period where the customer does not switch but instead wants exportable data and digital assets erased on service termination.

The termination clause should be linked to the switching workflow. A provider should not treat termination, retrieval, and erasure as a single event, because Article 25 gives them different functions and clocks.

Data Act Article 26 requires providers to give customers information on switching and porting procedures, methods, formats, restrictions, and technical limitations. It also requires a reference to an up-to-date online register with the data structures, data formats, relevant standards, and open interoperability specifications in which Article 25 exportable data is available.

For PaaS and SaaS, Article 30 requires free open interfaces for customers and destination providers to support switching, with enough service information to enable software communication for data portability and interoperability. Compatibility with published common specifications or harmonised standards applies after the Data Act repository trigger and the Article 30 compliance clock.

Data Act Article 31 creates a specific regime for services whose main features are custom-built for one customer or whose components were developed for one customer and are not offered at broad commercial scale through the provider's service catalogue. It also excludes non-production testing and evaluation services provided for a limited period from Chapter VI obligations.

The Commission FAQ cautions that custom-built services are not fully outside Chapter VI. Some obligations can still apply, including open interfaces and structured, commonly used, machine-readable export where relevant. Before relying on Article 31, providers must inform the prospective customer which Chapter VI obligations do not apply.

The Commission has published non-binding standard contractual clauses for cloud computing contracts to help parties implement the Data Act. The Commission material says the clauses are voluntary and can be adapted, so they are a drafting aid rather than a substitute for checking the provider's actual services, export inventory, charges, interfaces, and transition process.

The Commission describes cloud SCC coverage for switching and exit, termination, security and business continuity, non-dispersion, non-amendment, and liability. Those themes are useful checkpoints for a Data Act contract review, especially where a legacy cloud agreement spreads switching obligations across order forms, service descriptions, online policies, support terms, and pricing pages.

Data Act evidence for cloud switching should prove both contract coverage and operational performance. Keep the signed contract version, Article 25 clause matrix, export inventory, online-register snapshot, switching notice, transition calendar, assistance log, customer and destination-provider communications, charge calculation, retrieval confirmation, erasure record, and any technical-unfeasibility justification.

The record should also show why exceptions or limits were used. For example, keep the basis for exempting internal-functioning data, the reason a service was treated as custom-built or testing-only, the cost evidence for reduced charges before 12 January 2027, and the standards or interface status used for interoperability decisions.

The Data Act switching rules apply to providers of data processing services that provide those services to customers in the Union. The definition covers digital services giving on-demand network access to configurable, scalable, and elastic computing resources, which is why the Commission describes Chapter VI as covering cloud and edge services.

The fee review should therefore start with the service catalogue and customer contract, not with a generic cloud label. Identify the source provider, the customer, the service type, any destination provider or on-premises target, and the exportable data and digital assets involved in the switching process.

A switching charge is a charge, other than standard service fees or early termination penalties, imposed by a provider on a customer for the actions required by the Data Act to switch to another provider or to on-premises ICT infrastructure. The definition expressly includes data egress charges.

This means the billing inventory should not stop at line items called exit fees. It should include data egress, migration assistance fees, export tooling charges, charges for required switching actions, and any renamed or bundled fee that is triggered by the customer's switch.

Article 29 describes a reduced-charge period from 11 January 2024 to 12 January 2027. Because the Regulation generally applies from 12 September 2025, the operational compliance period for the Article 29 cost cap begins on that application date. During that period, reduced switching charges must not exceed the provider's costs directly linked to the switching process concerned.

A provider should therefore keep cost support for each reduced charge. A generic fee schedule is weak evidence unless it shows the charge is tied to actual direct switching costs and not to lock-in, margin recovery, ordinary service revenue, or a disguised deterrent.

From 12 January 2027, providers of data processing services must not impose switching charges on customers for the switching process. This prohibition covers data egress charges incurred for switching. It does not erase standard service fees or early termination penalties, which the Regulation excludes from the definition of switching charges.

Enforce the 12 January 2027 sunset in billing, contract templates, support scripts, renewal packs, procurement responses, and public pricing materials. After that date, a charge triggered by the switching process needs a documented reason why it falls outside the switching-charge definition.

The contract must set out the customer's switching rights and the provider's switching obligations in writing and be available before signature in a way the customer can store and reproduce. The contract must include a maximum notice period for starting the switching process that does not exceed two months.

After that notice period, the contract must allow switching or porting of all exportable data and digital assets without undue delay and not later than the mandatory maximum transitional period of 30 calendar days. During that period, the provider must give reasonable assistance, maintain business continuity, provide clear information about known continuity risks on the provider side, and maintain a high level of security.

Yes, but the Data Act gives specific conditions. If the mandatory 30-calendar-day transitional period is technically unfeasible, the provider must notify the customer within 14 working days of the switching request, justify the technical unfeasibility, and indicate an alternative transitional period that does not exceed seven months.

Separately, the contract must give the customer the right to extend the transitional period once for a period the customer considers more appropriate for its own purposes. Providers should distinguish provider-justified technical unfeasibility from a customer-requested extension.

The contract must provide a minimum data retrieval period of at least 30 calendar days. That retrieval period starts after the end of the transitional period agreed between the customer and the provider.

The contract must also guarantee full erasure of exportable data and digital assets generated directly by, or relating directly to, the customer after the retrieval period expires, or after a later agreed period, provided the switching process has completed successfully.

Before entering into a contract, providers must give prospective customers clear information on standard service fees, early termination penalties, and any reduced switching charges that may be imposed during the transition period. Where relevant, providers must also give information on services involving highly complex or costly switching, or where switching is impossible without significant interference in data, digital assets, or service architecture.

Providers must also give customers information on available switching and porting procedures, methods, formats, restrictions, and known technical limitations. They must provide a reference to an up-to-date online register with details of data structures, data formats, and relevant standards and open interoperability specifications for exportable data.

Keep evidence that connects the contract term, billing item, customer request, technical assessment, and outcome to the Data Act rule being applied. Maintain a switching-fee and deadline register that can be reviewed without reconstructing decisions from emails or support tickets.

For each customer or service, the record should show the charge classification, direct-cost support where reduced charges are used before 12 January 2027, notice-period start and end dates, transition-period dates, any 14-working-day technical-unfeasibility notice, retrieval-window dates, erasure evidence, and the owner who approved the decision.