Google Enables Free Multicloud Data Transfers in EU/UK Under Data Act

Google has moved to eliminate a key friction point for organisations running workloads across multiple cloud providers in Europe and the UK, announcing that its Data Transfer Essentials service will be available at no cost for customers processing workloads “in parallel” across two or more clouds — a step that goes beyond the EU Data Act’s minimum “at cost” requirement and arrives just as the new rules take effect. (reuters.com)

Background​

What changed and why it matters​

The European Union’s new Data Act is designed to lower switching costs and make it easier for organisations to move data and workloads between cloud providers. One of the Act’s practical effects is to constrain the way hyperscale providers can charge for data transfers (commonly known as egress fees), essentially forcing those charges to reflect cost rather than being used as a commercial barrier. Google’s announcement — to offer free multicloud transfers via a branded “Data Transfer Essentials” option in the EU and UK — was explicitly timed to coincide with the Act’s coming into force. (reuters.com)
This move sits inside a wider regulatory push. UK and EU authorities have been scrutinising hyperscalers — especially Amazon Web Services (AWS) and Microsoft Azure — for licensing practices, data egress pricing, and interoperability issues that can lock customers into a single vendor. Recent regulatory reports and investigations have highlighted egress fees and licensing complexity as core competitive bottlenecks in the cloud market.

---

What Google announced — the details​

Data Transfer Essentials: free for parallel multicloud​

Google Cloud has introduced Data Transfer Essentials, positioning it as a simple solution to move data between Google Cloud and other providers while running workloads in parallel across clouds. The company stated the service would be available at no cost to customers in the EU and the UK as the Data Act takes effect. Google framed the move as going beyond mere compliance: while the law allows passing through costs, Google elected to waive those fees for relevant multicloud scenarios. The announcement included a quote from Jeanette Manfra, senior director of global risk and compliance at Google Cloud, underscoring the “available today at no cost” position. (reuters.com)

How “parallel” is likely to be interpreted​

Google’s public messaging emphasises scenarios where organisations process workloads in parallel across providers. That phrase typically means architectures where multiple clouds collaborate on the same workload or where parallel pipelines are used for resilience, burst scaling, or specialised processing (for example, data ingest on one cloud and analytics on another). The precise eligibility criteria for the free transfers and any technical or contractual conditions were described at a high level in the announcement; enterprises should expect provider-specific technical requirements and contractual definitions to appear in product documentation or service terms.

---

How Microsoft and AWS have responded so far​

Microsoft: “at-cost” policy already in place​

Microsoft updated its Azure data transfer guidance in late August, outlining a process for EU customers to obtain at-cost data transfers between Azure and other providers. The Azure documentation explains that customers in the EU can request at-cost transfer pricing for internet transfers related to interoperable, parallel use of multiple providers, and details steps to create a support request for this purpose. That page was last updated on August 26 and describes the administrative workflow customers must follow to signal intent and obtain the accommodation. (learn.microsoft.com)

AWS: reduced rates for eligible cases​

AWS has also signalled accommodation for EU customers, noting in guidance that customers may request reduced data transfer rates for eligible use cases under the Data Act. AWS’ public network and global infrastructure FAQ references the ability for EU customers to request reduced rates and suggests contacting AWS support for specifics. Historically, AWS has emphasised that most customers pay little or nothing for common data transfers, while also offering mechanisms for reduced or waived fees in particular migration or research scenarios. (aws.amazon.com)

Historical context: earlier egress fee changes​

This is not the first time the industry has shifted on egress. In 2024 some providers announced moves to remove or ease egress charges to address regulatory and customer pressure. Those earlier actions set the tone for the Data Act and the September announcements now being rolled out across providers and regions. (reuters.com, cnbc.com)

---

Why this is strategically important​

1. Lowers a major switching cost​

Egress fees have been a blunt instrument for discouraging migration. Removing or reducing those fees for multicloud parallel workloads materially reduces the financial friction of moving data and increases the viability of multi-provider strategies.

2. Strengthens multicloud adoption and resilience​

Organisations that use multiple clouds for resilience, regulatory distribution, or best-of-breed capabilities gain more flexible options to balance workloads by cost, performance, or compliance needs. That flexibility supports architectures such as active-active deployments, geographic failover, and specialised processing chains.

3. Shifts the battleground to licensing and interoperability​

Regulators and customers have consistently flagged licensing as a remaining problem: even with cheaper transfer pricing, complex or punitive software licensing can keep workloads bound to one provider. The UK’s Competition and Markets Authority (CMA) and other bodies have singled out licensing as a priority since licensing can be used to preserve market power even if transfer fees fall.

---

Technical and commercial caveats​

Not all transfers are identical​

Practitioners should expect that “free” or “at-cost” transfer offers will come with technical and procedural conditions:

• Eligibility definitions (e.g., parallel processing, selected services or APIs).
• Required support requests or approvals (Microsoft’s documentation, for example, requires a support ticket and specific metadata such as subscription ID and ASN information). (learn.microsoft.com)
• Time-limited windows or process windows for migrations or transfers.
• Potential volume thresholds or usage patterns that qualify for reduced pricing (AWS’s guidance points customers to request reduced rates for eligible use cases). (aws.amazon.com)

Network and performance realities remain​

Waiving fees does not change physical network constraints. Large data movements still consume bandwidth and require engineering to manage throughput, time-to-migrate, and consistency. Organisations moving petabytes will still need careful planning around transfer windows, parallelisation, and staging.

Providers can shift the economics elsewhere​

Providers may respond commercially by adjusting other elements of pricing or contract terms (for example, support tiers, storage or compute pricing, licensing bundles, or specialised network services). The absence of an egress fee does not by itself guarantee a cheaper overall bill.

---

What IT teams should do now (practical checklist)​

• Audit data transfer patterns
• Identify where egress volumes are highest and which workloads are candidates for multicloud or migration.
• Map licensing dependencies and contracts
• Inventory all software licences that might behave differently on other clouds, including Microsoft server and database licences.
• Test portability in small, controlled pilots
• Validate the workflow: initiate a parallel workload across clouds and verify the transfer and processing steps, monitoring latency and cost signals.
• Open formal support cases if needed
• Follow vendor-prescribed procedures (e.g., Azure’s support request steps) to declare intent and obtain at-cost or waived transfer accommodation. (learn.microsoft.com)
• Negotiate contract clauses at renewal
• Use the new regulatory baseline as leverage to secure better exit terms, migration assistance, or explicit egress allowances.
• Align security and compliance workstreams
• Ensure data portability actions meet data protection, residency, and audit requirements across jurisdictions.

These steps give organisations a pragmatic path to exploit the new commercial openings while reducing the risk of surprises or hidden costs.

---

Competitive and regulatory implications​

The market dynamic: competition through openness​

Google’s decision to offer free multicloud transfers in the EU and UK is a competitive gambit as much as regulatory compliance. By waiving fees, Google positions itself as the most permissive vendor on transfer economics — a persuasive marketing point for customers considering multicloud architectures. This could nudge enterprises to experiment more with Google Cloud for components of their stack that previously stayed bound to a single provider due to egress economics. (reuters.com)

Regulators still have work to do on licensing and interoperability​

Regulators have repeatedly identified licensing and technical incompatibility as the next barriers to true portability. The UK CMA’s work highlights how licensing, proprietary APIs, and non-standard data formats can be used to entrench market positions even where pricing is constrained. That means industry-level change requires both commercial concessions and technical standards work.

Risk of regulatory fragmentation​

Different jurisdictions may interpret or implement the Data Act and related obligations differently. The UK has pursued its own cloud market scrutiny via the DMCCA and CMA, and coordination between Brussels and London is not automatic. Enterprises operating globally should be mindful that the practical availability and terms of “free” transfers may vary between EU member states, the UK, and other regions.

---

Strengths of Google’s move — and potential downsides​

Notable strengths​

• Simple, customer-friendly headline: “Free transfers” removes a clear barrier for multicloud testing and migration. (reuters.com)
• Regulatory signalling: By going beyond the Data Act’s minimum, Google demonstrates proactive compliance and positions itself favourably with regulators and customers.
• Competitive differentiation: If Google’s free option is broad and operationally simple, it could draw workloads previously locked to other clouds by transfer economics.

Potential risks and limitations​

• Eligibility and fine print: The phrase in parallel and product-specific terms may narrow applicability in practice; customers should verify the exact scope.
• Commercial rebound: Providers can reallocate costs to other line items (support, compute, storage, or licensing) — so the “free” headline may not produce equivalent total cost savings in every case.
• Operational complexity: Removing fees doesn’t remove the technical work required for safe, performant migrations across clouds.
• Short-term marketing vs long-term policy: Providers could change policies after market conditions shift; enterprises must avoid assuming permanence without contractual commitments.

Any claims about universal savings or instant portability should be treated cautiously until organisations validate outcomes in their own contexts.

---

What remains uncertain or requires verification​

• Precise technical eligibility: the operational definition of “in parallel” and the list of eligible services for Data Transfer Essentials.
• Long-term permanence: whether “no cost” is a stable, evergreen policy or tied to a specific period or set of conditions.
• Interaction with licensing terms: how Microsoft, Oracle, and third-party software licensing will interact with eased transfer fees; licensing complexity remains a primary switching cost.

These points are likely to be clarified in vendor product pages and legal terms, and organisations should request written contract language when negotiating significant migrations.

---

Long-term outlook: competition, architecture and standards​

The new environment reduces an obvious economic barrier to multicloud, but the bigger structural questions remain: will the industry move toward technical standards and BYOL parity that truly enable seamless workload mobility? Or will competition shift to other levers — licensing, integrated platform features, or enterprise services — that continue to divide the market?
Regulators have signalled they will not stop at pricing. The CMA and EU bodies are focused on creating conditions where interoperability and fair licensing enable choice, not just cheaper exits. The evolution of frameworks and potential conduct obligations could reshape vendor behaviour well beyond fees, forcing clearer standards for formats, APIs, and contractual transparency.

---

Final assessment and practical takeaways​

Google’s decision to waive data transfer fees for multicloud workloads in the EU and UK is a meaningful, customer-facing concession that reduces one of the most visible costs of switching or operating in multicloud. It also raises the bar for competitors and helps translate regulatory pressure into operational benefit for customers. However, the headline should not obscure persistent complexities: licensing, cross-cloud engineering, contractual detail, and practical network constraints still require attention.
For WindowsForum readers and IT decision-makers, the immediate action items are clear:

• Treat this as an opportunity to pilot multicloud patterns with a measured, technical approach.
• Use formal vendor processes and document exchanges to secure written commitments on costs and migration support.
• Focus on licence portability and interoperability as the next strategic priority — price relief is necessary, but not sufficient, to break vendor lock-in.

The EU Data Act has changed the economics of cloud exit in Europe and the UK; Google’s Data Transfer Essentials amplifies that change with a no-cost offer for parallel workloads. That combination makes this a pivotal moment for organisations to reassess portability strategies — but success will come from careful planning, contractual clarity, and technical validation, not from headline pricing alone. (reuters.com)

---

Source: Republic World Google Scraps Some Cloud Data Transfer Fees in EU, UK

 

[ChatGPT]

Google Cloud’s decision to waive multicloud data-transfer charges in the EU and UK — through a new “Data Transfer Essentials” option — has shifted a regulatory tug‑of‑war into a full commercial play, and it matters for every IT team wrestling with vendor lock‑in, migration economics, and multicloud resilience.

Background​

The European Union’s Data Act — due to take effect on September 12, 2025 — forces cloud providers to price data‑egress (transfer) fees on a cost‑reflective basis for certain cross‑provider use cases. The aim is to reduce switching costs and promote competition among hyperscalers long criticised for using egress pricing as a deterrent to migration. Several major cloud vendors updated guidance and product terms in the weeks leading up to the Act’s entry into force.
Google Cloud’s response was to go further than the law’s baseline in headline terms: introducing Data Transfer Essentials and billing qualifying multicloud “parallel” traffic at zero cost for eligible traffic in the EU and the UK. The company framed the move as encouraging multicloud strategies, improving operational resilience, and lowering migration friction.

What Google announced (and how it differs from “at cost”)​

Data Transfer Essentials in plain terms​

• What it is: A configurable option in Google Cloud Platform (GCP) that meters multicloud traffic separately and, for qualifying use cases, bills that traffic at zero cost.
• Eligibility framing: Google’s public messaging targets workloads running in parallel across two or more clouds — a common pattern for resilience, burst scaling, or best‑of‑breed pipelines. The exact operational definition of “in parallel” and the list of eligible services are described in high‑level product statements and are expected to be clarified in detailed product documentation or service terms.

How this stacks up against the EU Data Act and rivals​

• The Data Act’s baseline: The regulation requires cloud providers to avoid charging supra‑competitive egress fees — effectively compelling “at cost” pricing for transfer fees in certain interoperability or portability scenarios. Google’s waiver goes beyond the legal minimum by eliminating charges for qualified transfers rather than passing on marginal cost.
• Competitor moves: Microsoft rolled out an “at cost” mechanism for EU customers that requires a support request and administrative steps to invoke, while Amazon Web Services signalled reduced‑rate options for eligible cases rather than a blanket waiver. Those responses are more conservative than Google’s zero‑cost positioning.

Why this matters: strategic and technical implications​

Lowering the most visible switching cost​

For many organisations, egress fees are the single most visible economic barrier to migration. Removing or sharply reducing that fee for parallel multicloud workloads immediately changes the financial calculus for pilot migrations, hybrid deployments, and burst‑offload patterns. The net effect is to make testing, prototyping, and staged migration far less risky from a billing standpoint.

Competitive signalling and regulatory optics​

Google’s public move is both a customer‑facing product change and a regulatory statement. By volunteering a more permissive commercial posture than the law strictly requires, Google positions itself as proactively aligned with regulators’ competition goals and attempts to reshape customer perceptions about which provider is easiest to exit. That signalling may influence procurement conversations and enterprise vendor short‑lists in the short term.

The battleground shifts to licensing, features, and operational lock‑in​

Even if egress prices fall or disappear in selected scenarios, other lock‑in levers remain:

• Software licensing complexity: Vendor or third‑party licences that price differently on alternative clouds (notably some Microsoft licences) can still make migration uneconomic. Licences, not data transfer costs, can be the dominant migration blocker.
• Proprietary services and APIs: Workloads built around a provider’s unique managed services, proprietary APIs, or data formats still require engineering effort or refactoring to move. Egress deals address cost, not portability of architecture.
• Operational and performance realities: Large bulk transfers remain engineering challenges — bandwidth, staging, consistency, and migration windows are technical constraints that money alone does not solve.

Technical caveats and “fine print” to watch​

Defining “parallel” workloads​

Google’s eligibility language centres on the idea of processing in parallel across multiple clouds. In practice, that term can be interpreted narrowly in product terms to require:

• Simultaneous, coordinated processing pipelines that meet certain API/metadata requirements, or
• Specific networking configurations or service pairings that allow metering as “multicloud parallel” traffic.

Organisations should expect service documentation and contract terms to define this precisely; pilots are essential to confirm real eligibility before relying on assumed cost savings.

Metering, observability, and audit trails​

When an egress waiver applies only to metered “qualifying” traffic, auditability becomes crucial. IT teams must be able to:

• Identify and tag flows that qualify,
• Produce logs and reconciliations to support billing disputes, and
• Validate that multi‑provider test runs are being classified as expected by vendor billing systems.

Potential commercial rebalance​

Providers can — and historically have — adjust pricing and packaging across other line items. Removing a visible egress charge does not guarantee lower total cost of ownership if:

• Compute or storage pricing is restructured,
• Support tiers, enterprise services, or managed offerings are repriced, or
• Licensing surcharges or interconnect services acquire new premiums.

Practical checklist for Windows/Enterprise IT teams​

• Audit current egress volumes and identify candidate workloads that could exploit waived transfers. Focus on high‑egress pipelines, backups, and migration artifacts.
• Inventory license dependencies (OS, database, middleware, third‑party libraries) and quantify the cost difference of running those services on alternative clouds. Treat licensing as a first‑order migration risk.
• Run small, controlled pilots that implement true parallel processing between clouds, and verify that vendor billing classifies traffic as eligible for waiver. Collect screenshots, logs, and support case IDs.
• Open formal vendor support requests and secure written confirmation of eligibility and duration for waived transfers; negotiate contractual clauses where migrations are material to business continuity.
• Reassess disaster‑recovery and high‑availability designs to take advantage of lower egress friction, while preserving attention to latency, consistency, and failover automation.

Commercial and regulatory analysis​

Short‑term winners and marketing effects​

Google’s zero‑cost message is a potent marketing differentiator and could accelerate short‑term experimentation on GCP. By making the economic case for multicloud testing easier, Google may gain share among risk‑averse procurement teams and cloud architects looking to de‑risk projects.

Medium‑term market dynamics​

Regulatory pressure — notably from EU bodies and the UK’s Competition and Markets Authority (CMA) — is already shifting incentives. Even where providers implement “at cost” frameworks, the market will increasingly compete on:

• Ease of data portability (tools, guides, migrations services),
• Licensing fairness and clarity, and
• Native cross‑cloud interoperability features.

Those battlegrounds are likely where durable competitive advantage will be fought, not in headline egress pricing alone.

Risk of regulatory fragmentation​

The Data Act is an EU‑level instrument; the UK has its own regulatory tools and the CMA has been an active investigator of cloud market dynamics. Enterprises operating across jurisdictions should be prepared for variations in implementation and interpretation between the EU, UK, and other regions — meaning a single global playbook won’t fit every market.

---

Strengths of Google’s move​

• Simplicity for procurement: A zero‑cost headline cuts through complex bill estimates and creates a straightforward negotiation point with finance teams.
• Regulatory alignment: By exceeding the minimal legal requirement, Google reduces the friction of regulatory approval processes and gains goodwill with regulators looking to boost competition.
• Customer experimentation: Lowered transfer costs make it easier to trial multicloud patterns (analytics on one cloud, storage on another, failover between providers), which can accelerate adoption of best‑of‑breed architectures.

Risks and limits​

• Eligibility opacity: The value depends on precise product definitions of “parallel” — firms should be cautious until contractual language and operational metering are explicit.
• Commercial rebound risk: Providers could shift margin to adjacent products, meaning the elimination of transfer fees is not a guaranteed reduction of total spend.
• Technical effort remains: Migration engineering, data consistency, and performance tuning still require substantial project resources and timeline commitments.

How this affects Windows administrators and enterprise architects​

Windows shops and enterprises that rely on Microsoft technologies must treat this moment as a policy and procurement opening, not a turnkey migration signal. Tactical recommendations:

• Use the Data Act and vendor concessions as negotiation leverage in contract renewals, explicitly requesting written terms on egress, migration support, and licensing portability.
• Pilot cross‑cloud directory and identity integration scenarios (Azure AD + GCP IAM) to identify licensing or SSO wrinkles prior to large migrations.
• Revalidate disaster‑recovery playbooks in a multicloud context: automation, DNS failover, and cross‑cloud backup restores need verification even if transfer costs fall.

Verification and known unknowns​

Several claims circulating in press summaries require careful verification before treating them as settled facts:

• Reports quoting specific contract backlogs (for example, a cited figure of ~$106 billion in cloud contracts and a conversion expectation attributed to Google Cloud leadership) should be treated as unverified unless confirmed by official financial filings or direct company statements. Those numbers are time‑sensitive and dependent on disclosure conventions; they are not validated within the vendor documentation and regulatory summaries consolidated here. Exercise caution when using them in procurement or investor analysis.
• Assertions about customer lists (for example, naming particular AI labs or customers relying on GCP for specialized GPU clusters) may be accurate in public reporting but should be verified against vendor case studies or contractual confirmations before relying on them for strategic decisions. The regulatory and product documentation examined here focuses on pricing and policy, not exhaustive customer rosters.

When facts are time‑sensitive (stock prices, contract values, or customer disclosures), the only reliable route is direct verification from official company filings, vendor product pages, or primary vendor support confirmations.

Tactical playbook for the next 90 days​

• Establish a cross‑functional migration task force: procurement, legal, SRE, network, and application owners. Draft a prioritized list of pilot candidates.
• Execute a controlled pilot to prove the multicloud pipeline, capture vendor metering evidence, and confirm zero‑cost treatment in practice. Escalate any discrepancies through formal support channels and retain documentation.
• Review and rewrite renewal clauses: insist on explicit egress treatment, migration support credits, and license portability assurances where migration is business‑critical.
• Maintain a vendor neutrality checklist: ensure critical services have contingency plans that do not depend solely on egress economics — e.g., database replication strategies, container registry mirrors, and IaC portability.

Final assessment​

The EU Data Act has altered the contractual baseline for cloud exits, and Google’s Data Transfer Essentials amplifies that change by offering a customer‑friendly option that eliminates one visible barrier to multicloud operations. This is a meaningful win for customers in the short term: it lowers experimentation costs and forces competitors to respond. However, price concessions do not equal automatic portability. Durable multicloud freedom will be decided by licensing fairness, interoperability standards, and honest, enforceable contractual terms — not egress headlines alone. Organizations should treat the new environment as an operational opportunity, not a migration panacea: pilot deliberately, secure written guarantees, and keep a close eye on how providers respond commercially across compute, storage, and licensing lines.
The market has turned a regulatory nudge into an active commercial battle. For IT teams and Windows administrators, the next months are an ideal window to pilot multicloud patterns, tighten contractual protections, and rework DR and portability plans — while remaining skeptical of any single headline that promises instant, cost‑free migration without engineering work or licensing consequences.

---

Source: CryptoRank Google Cloud changes data transfer fee model with new EU laws on the horizon | Tech Google | CryptoRank.io

 

[ChatGPT]

Google has announced a dramatic change to how it charges for cloud data movement in the European Union and United Kingdom, waiving fees for certain multicloud transfers through a new “Data Transfer Essentials” option that it says is available at no cost — a move timed to land just ahead of the EU Data Act taking effect on 12 September 2025. (reuters.com)

Background​

The EU Data Act is a sweeping regulatory effort to reduce friction in the European data economy by making it easier for firms and public bodies to access, share and port data across platforms and providers. Among other measures, the Act requires cloud providers to make data-porting options available at cost — effectively limiting the practice of charging exorbitant transfer or “egress” fees that can lock customers into a single provider. The regulation becomes applicable on 12 September 2025. (digital-strategy.ec.europa.eu)
Cloud incumbents — principally Amazon Web Services (AWS), Microsoft Azure and Google Cloud — together control the lion’s share of global cloud infrastructure. Recent market measurements place the three at roughly two-thirds of the market, with per-quarter variance: Synergy Research Group and Canalys show AWS leading, Microsoft second and Google Cloud a more distant third. These market-share figures frame the competitive and regulatory attention now focused on cloud switching costs and interoperability. (srgresearch.com)

---

What Google actually changed​

The headline: zero fees for specified multicloud transfers​

Google’s public messaging says it will make data transfers “in-parallel” between Google Cloud and other cloud providers in the EU and UK free for customers who opt into its new Data Transfer Essentials offering. In practical terms, that means organizations running the same, in-parallel workloads across two or more cloud platforms — for resilience, latency distribution, or cost optimisation — will not see intra-workload transfer charges from Google under this program. Reuters’ coverage of the announcement quotes Google Global Risk & Compliance Senior Director Jeanette Manfra and frames the move as going beyond the minimum legal requirement of charging at cost. (reuters.com)

How customers will access the benefit​

Google says the program is opt-in: customers must register for Data Transfer Essentials to receive the zero-fee treatment. The company positions the change as a direct support for multicloud architectures and for reducing vendor lock-in that has frustrated customers and regulators alike. The opt-in requirement suggests Google intends the offer to be trackable and administrable within billing and support systems rather than an open-ended waiver.

Limits and scope — what isn’t covered​

Google’s public statements emphasise in-parallel workloads and same-organization transfers. That implies typical CDN or internet egress scenarios, or transfers between unrelated customer accounts, are not the intended use-case. The announced policy appears targeted at operational patterns where an organization intentionally runs identical or cooperative workloads across providers for resilience or regulatory diversity — the classic multicloud design pattern that European regulators want to encourage.

---

Why this matters: the EU Data Act in context​

From “at cost” to zero: legislative minimum vs. commercial practice​

The EU Data Act sets a baseline that cloud providers may offer data transfers at cost — i.e., recouping only the incremental expense required to move bytes. Legislators saw “at cost” as a pragmatic compromise: providers should not pocket unreasonably large exit fees, but they can still recover real network or operational costs. Google’s move to waive fees entirely for specific multicloud transfers goes beyond that statutory baseline and converts a legal minimum into a commercial concession for a subset of customers. (digital-strategy.ec.europa.eu)

Market power and vendor lock-in​

The practical effect of data egress fees has long been to increase the friction of moving: even when data is portable in technical terms, moving petabytes is costly in time and money. European competition authorities have repeatedly flagged the cloud market’s concentration — a regulator’s attention Google itself cites — making any credible reduction in switching friction noteworthy for enterprises and public sector buyers. Google’s announcement is thus more than a pricing tweak; it’s a stake in the debate about whether hyperscalers are facilitating or restricting portability. (srgresearch.com)

---

How competitors have responded (and already changed pricing)​

Microsoft’s “at cost” approach (August 2025)​

Microsoft updated its guidance in late August 2025 to provide at-cost refunds or credits for internet egress between Azure and other providers in the EU — a direct implementation of the Data Act’s allowance. Microsoft’s support documentation details the steps customers should take to request at-cost credits, what qualifies, and the qualification criteria including billing address and routing constraints. That guidance was published on 26 August 2025, indicating Microsoft moved early to align commercial policy with the impending law. (learn.microsoft.com)

AWS: reduced-rate requests for eligible EU customers​

AWS’s public documentation and its network FAQ make clear that EU customers can request reduced data transfer rates for eligible scenarios under the Data Act framework. AWS also has for some time offered specific free allowances (e.g., CloudFront free tiers) and researcher discounts, and it announced earlier global waivers for customers migrating off AWS, though AWS’s model relies more on case-by-case processing than an across-the-board zero-fee program. AWS’s FAQ points customers to support channels to seek rate reductions or clarifications. (aws.amazon.com)

How Google’s approach differs​

• Google’s move is explicit, public, and framed as a no-cost offer for a defined multicloud use-case, rather than a credit- or request-based program.
• Microsoft’s approach provides an at-cost mechanism that still requires customer action and validation.
• AWS’s line is to offer reduced rates via requests and maintain some pre-existing waivers for specific migrations.

Taken together, the three major providers are now aligned in intent — reducing switching friction — but they differ in mechanics and commercial aggressiveness.

---

Technical and commercial caveats enterprises must evaluate​

1) Opt-in vs. automatic application​

Google requires opt-in for Data Transfer Essentials. Enterprises should assess the administrative overhead: registration, configuration, and whether the opt-in must be renewed or is bound to specific subscriptions or projects. This is relevant for procurement teams negotiating contracts and for billing reconciliation. Without careful onboarding, teams risk assuming waivers apply universally when they do not.

2) The in-parallel constraint​

Google’s public language focuses on in-parallel workloads — not bulk archival offloads, public internet delivery, or third-party account transfers. Organizations will need to document architectures that qualify, and design telemetry that demonstrates the workload topology if disputes arise. Technical architecture teams should map which traffic types are eligible and which will still produce egress charges under standard pricing. Reuters’ reporting highlights this constraint as central to Google’s offer. (reuters.com)

3) Network route and ASN requirements (Azure precedent)​

Microsoft’s at-cost policy makes clear that routing and ASN details matter; Azure requires information like the autonomous system numbers of external endpoints when customers request credits. That precedent suggests providers will demand technical proof of eligible traffic, which could increase procurement friction if customers lack the required network engineering details. Enterprises should prepare ASN and routing documentation ahead of any large-scale migration or multicloud implementation to avoid delays. (learn.microsoft.com)

4) Non-technical lock-ins persist​

Eliminating or reducing egress fees removes one major financial friction, but it does not automatically remove other lock-in factors: proprietary managed services, data format incompatibilities, deeply embedded contracts, and differentiated AI/ML toolchains can still make switching expensive. Google itself has insisted that removing egress fees is meaningful but not a cure-all for restrictive licensing practices and product differentiation. Enterprises should therefore view fee changes as necessary but not sufficient for true portability. (cloud.google.com)

---

Strategic implications for buyers and IT leaders​

Short-term actions​

• Review contracts and billing lines for egress and cross-cloud transfer fees across all providers.
• Identify multicloud workloads and confirm whether they meet in-parallel criteria for Google’s program.
• Prepare ASN and routing details if you plan to rely on Microsoft’s at-cost credits or AWS reduced-rate requests. (learn.microsoft.com)

Medium-term programmatic changes​

• Reassess vendor negotiation strategies. The existence of a zero-fee program gives buyers leverage to negotiate favorable terms beyond the data-transfer line item, including discounts for managed services or guarantees around data portability timelines.
• Standardise data formats and replication strategies to reduce the actual technical cost of switching. Porting-ready data and infrastructure-as-code can materially reduce migration timelines and cost.
• Build clear observability around cross-cloud traffic so any dispute over qualifying transfers can be resolved with accurate telemetry.

Risk management and compliance​

Public sector and regulated entities should map how these commercial changes intersect with data residency, sovereignty, and audit requirements. Reduced egress costs do not alter where data resides or which local laws apply; those remain contract and policy matters separate from transfer pricing.

---

Competition and regulatory optics​

Google’s timing — publishing the change days before the Data Act becomes applicable — is strategic. It positions Google as a champion of openness and multicloud interoperability while placing competitive pressure on AWS and Microsoft to match or beat its commercial generosity in ways that may not be cost-neutral.
Regulators in Europe and the UK have been explicit about their desire to curb anti-competitive tactics such as onerous switching fees and restrictive licensing. The three hyperscalers’ responses — AWS’s case-based reductions, Microsoft’s at-cost refund mechanism, and Google’s zero-fee opt-in — create a landscape where market behavior is visibly realigning with regulatory intent. The commercial competition over transfer fees may now shift to other battlegrounds: managed AI services, proprietary accelerators, and data-processing feature sets.

---

Financial and engineering realities: who really pays?​

Zero-fee for data movement is not free in economic terms — network capacity, interconnect peering, and cross-cloud engineering carry real costs. Providers that internalise those costs are effectively subsidising certain customer behaviours, perhaps to win long-term platform lock-in in other areas such as database, AI, or developer tooling revenue.
From a network economics perspective:

• Peering and direct interconnects are the lowest-cost paths for large-volume transfers; companies that maintain dense peering ecosystems can bear transfer traffic more cheaply.
• Indirect transfers over transit ISPs cost more; providers may carve out allowances only where the cheapest routing applies or where interconnects are available.
• Some providers will likely trade off transfer revenue for upsell opportunities into differentiated services or long-term contracts.

Enterprises should therefore understand the profile of their traffic: short bursts, constant cross-cloud replication, or periodic bulk moves all have different cost and operational implications.

---

What this means for multicloud strategies​

The practical outcome is that multicloud just became materially less expensive in a specific set of use-cases — those that match Google’s Data Transfer Essentials design. But successful multicloud is not only about costs; it is a disciplined engineering and governance approach.

• Resilience-first teams can more confidently place redundant workloads across clouds without fearing crippling egress bills.
• Data gravity remains relevant: the largest and most active datasets may still be most cost-efficient to process where they already reside.
• AI and platform lock-in: as providers differentiate on model availability, performance and integrated tooling, the economics of data movement will be only one factor in workload placement decisions.

---

How to operationalise the change: a checklist​

• Inventory your cloud workloads and classify them by portability and cross-cloud traffic patterns.
• For multicloud candidates, document ASN and routing information and register for provider-specific programs (e.g., Google’s Data Transfer Essentials opt-in; Microsoft support request process). (reuters.com)
• Establish observability (flow logs, VPC flow, interconnect telemetry) to prove eligibility for at-cost credits or waivers.
• Negotiate contract language that captures the new commercial reality: include explicit terms for transfer rates, opt-in program duration, and audit mechanisms.
• Reassess your vendor lock-in risk map with the new pricing variables in place.

---

Strengths and risks of Google’s announcement​

Strengths​

• Immediate commercial relief for selected multicloud scenarios, lowering an important cost barrier.
• Positive regulatory signalling: proactive alignment with the EU’s policy objective can reassure public customers and procurement bodies.
• Tactical differentiation: in a market where Google seeks to grow share, this offers a marketing and procurement advantage that can accelerate wins in Europe and the UK. (reuters.com)

Risks and unanswered questions​

• Opt-in mechanics and eligibility verification could add administrative friction and slow adoption if not straightforward.
• Scope creep risk: customers may assume the waiver applies broadly and face unexpected bills if traffic is later reclassified.
• Sustainability of the offer: commercial pressures or network economics may prompt future recalibration; Google has precedent for earlier concessions (e.g., eliminating migration egress fees globally in 2024), but long-term permanence is untested. (cloud.google.com)
• Regulatory capture risk: if providers comply only superficially, regulators may need to monitor implementations to ensure the spirit of the Data Act is fulfilled.

Where public statements are paraphrased from company blogs or news wire reporting, some operational details remain to be seen in the providers’ formal policy pages and contractual amendments. That means procurement teams must validate vendor promises against contract language and billing behavior.

---

The big picture: cloud portability as a competitive lever​

This episode is a case study in how regulation, competition and engineering economics interact to change market outcomes. The EU Data Act set a legal floor; providers responded with a spectrum of commercial measures — from Microsoft’s compliance-focused at-cost refunds to AWS’s request-based reductions to Google’s more aggressive zero-fee opt-in for multicloud transfers.
For enterprise buyers, the cumulative effect is positive: switching-related friction is being reduced. But meaningful portability will require sustained changes beyond pricing: standardised data formats, open APIs, and interoperable management tooling remain essential.

---

Practical next steps for WindowsForum.com readers​

• Audit current egress costs across clouds and project savings under Google’s Data Transfer Essentials where applicable.
• If you run multicloud in the EU or UK, validate opt-in and eligibility now rather than later to capture the new commercial terms.
• Talk to your network team about ASN and route documentation; vendors like Microsoft already require that level of detail for at-cost credits. (learn.microsoft.com)
• Treat data-transfer pricing changes as part of a broader vendor strategy — negotiate firm commitments on portability, documentation, and dispute resolution.

---

Conclusion​

Google’s decision to waive specified EU and UK multicloud data transfer fees via Data Transfer Essentials is a meaningful, tactical advance for customers seeking true multicloud flexibility. It transforms a regulatory minimum — the Data Act’s “at cost” rule — into a commercial concession for a defined use-case, and escalates competitive pressure on AWS and Microsoft to make their own offerings both compliant and customer-friendly. At the same time, the change is not a panacea: enterprises must still navigate technical eligibility, contract language, and other forms of lock-in that money alone cannot eliminate. The EU Data Act has changed the bargaining landscape; the real winners will be organisations who use this window to standardise data formats, document network topologies, and negotiate enforceable portability guarantees into contracts so that movement becomes as frictionless as policy makers intended. (digital-strategy.ec.europa.eu)

---

Source: TechRadar Google slashes UK and EU cloud data transfer fees ahead of EU Data Act

 

[ChatGPT]

Google has moved the cloud market’s chess pieces with a tactical — and loudly publicised — concession: for customers in the European Union and the United Kingdom, Google Cloud will no longer charge data‑egress fees for qualifying “in‑parallel” multicloud transfers through a new offering called Data Transfer Essentials, a commercial move timed to coincide with the EU Data Act coming into force on 12 September 2025. This is not merely compliance with the regulation’s minimum “at‑cost” requirement — it is a deliberate, customer‑facing policy that is likely to reshape procurement conversations, migration economics, and how IT teams plan multicloud architectures across Europe and the UK.

Background and overview​

The EU Data Act establishes a legal baseline intended to reduce switching costs in the European cloud market by requiring certain cross‑provider data transfers to be priced on a cost‑reflective basis. The regulation became applicable on 12 September 2025 and was designed to make it easier for organisations — private and public — to move data and processing between providers without being blocked by excessive egress fees or contractual barriers.
In the run up to the Act taking effect, the three major hyperscalers responded in different ways:

• Google Cloud announced Data Transfer Essentials, stating the service will be available at no cost for qualifying multicloud, in‑parallel workloads in the EU and UK.
• Microsoft Azure updated its guidance and launched a process for EU customers to request at‑cost transfers, requiring an administrative request to trigger the accommodation.
• Amazon Web Services (AWS) signalled it will offer reduced rates or case‑by‑case accommodations for eligible migrations and interoperability scenarios.

Taken together, these moves mark a material shift in market behaviour: providers are aligning commercial practice with regulatory intent — and, in Google’s case, going one step beyond the statutory minimum in headline terms.

---

Why this matters: the economics of cloud exit​

For many organisations, data egress fees were the clearest and most visible barrier to migration or multicloud strategies. When a cloud provider charges for every byte that leaves its environment, the cost of staged migrations, active‑active architectures, and cross‑cloud replication can quickly balloon.
Google’s decision to eliminate those charges for specified, in‑parallel use cases changes the financial calculus for several common scenarios:

• Pilot migrations and proof‑of‑concepts become far cheaper to execute because organisations no longer face uncertain egress bills for repeated testing.
• Active‑active resilience patterns — where identical workloads run simultaneously on two or more clouds — become more economically viable.
• Cross‑cloud analytics pipelines (for example, ingest in one provider, heavy analytic processing in another) can be balanced on technical rather than purely cost grounds.

However, price is only one axis of vendor lock‑in. The dominant remaining frictions are software licensing, proprietary managed services/APIs, data formats, and contractual terms that govern service portability.

---

What exactly did Google change — and what “in‑parallel” likely means​

Google’s public messaging brands the change as Data Transfer Essentials: an opt‑in framework that meters multicloud traffic and applies a zero‑cost treatment to qualifying flows between Google Cloud and other providers when workloads are processed in parallel.
Key characteristics of the announced change:

• Scope: EU and UK customers; intended for traffic that is part of parallel processing workflows across multiple cloud providers.
• Opt‑in: Customers must register or opt into the Data Transfer Essentials offering to receive the no‑cost treatment.
• Metered and auditable: The approach implies metering of qualifying traffic to ensure only eligible flows receive the waiver.
• Commercial posture: Google frames this as going beyond the Data Act’s minimum “at cost” requirement by electing to waive charges in qualifying scenarios.

The operational definition of “in‑parallel” is deliberately high level in public statements. In practice, providers almost always define eligibility precisely in product documentation or service terms; that definition can include requirements such as simultaneous workload execution, specific API/metadata flags, or particular networking configurations. Organizations should expect vendor documentation to contain the fine print and to clarify which Google services and interconnect patterns are included.

---

How Microsoft and AWS responded — a comparison​

The hyperscalers’ responses differ in style and administrative burden:

• Microsoft Azure implemented an at‑cost process for EU customers. Practically, that requires creating a formal support request and providing metadata such as subscription identifiers and autonomous system numbers (ASNs) for the non‑Azure endpoint. The approach is consistent with the Data Act’s legal baseline but introduces an administrative step.
• AWS has signalled reduced rates for eligible migrations and interoperability scenarios. AWS historically handles special cases via support and partner programs; its response is more case‑by‑case than Google’s blanket opt‑in waiver.
• Google Cloud made the most visible commercial concession by packaging an opt‑in, zero‑price alternative targeted specifically at multicloud in‑parallel use cases.

These differences matter in procurement negotiations: Microsoft’s route provides a formal mechanism but adds process friction, AWS’s case‑based model requires engagement and justification, while Google’s public, no‑cost positioning aims for simplicity and a stronger marketing message.

---

Strengths of Google’s move​

• Immediate reduction of a visible switching cost: By removing the egress bill for qualifying flows, Google makes it materially cheaper to pilot multicloud and staged migrations.
• Regulatory signalling: Going beyond “at cost” positions Google as proactively aligned with the Data Act’s goals, which may appeal to public sector buyers and procurement teams focused on compliance and openness.
• Competitive differentiation: An easy, no‑cost opt‑in product is a compelling procurement talking point and can influence vendor short‑lists.
• Operational flexibility for resilience: Teams can more easily evaluate active‑active patterns, cross‑cloud DR, and specialised processing without immediate cost barriers.

---

Important caveats and risks​

• Eligibility and fine print: The devil will be in the definitions. In‑parallel may be interpreted narrowly in product terms and exclude common egress patterns (for example, transfers involving separate customer accounts, CDN egress, or third‑party recipients).
• Metering and observability: When a waiver applies only to metered “qualifying” traffic, enterprises must be able to prove which flows qualify. That requires robust telemetry (flow logs, interconnect metrics, tags) and reconciliation processes to support billing disputes.
• Cost rebalancing: Providers can and historically have reallocated economics across product lines. Removing egress fees does not rule out increases or bundling elsewhere — in compute, storage, support tiers, or managed AI services.
• Licensing remains the dominant lock‑in lever: Software licences that are priced differently or restricted on other clouds — notably certain enterprise database or middleware licences — can continue to make migration uneconomic regardless of transfer charges.
• Operational complexity: Moving petabytes across providers is an engineering challenge. Bandwidth limits, migration windows, consistency, and application reconfiguration still require planning and investment.
• Policy permanence: Commercial policies can evolve. Organisations should avoid assuming a zero‑fee position is permanent without securing contractual guarantees or documented commitments.

---

Technical and contractual details to verify before relying on zero‑cost claims​

Organisations should confirm the following before planning significant migrations or operational changes:

• The exact technical definition of “in‑parallel” and a list of included services and networking topologies.
• Whether inter‑account transfers, cross‑organization projects, or cloud‑to‑cloud transfers across different tenancy models are included or excluded.
• The metering mechanism and what logs or proofs Google requires to validate qualifying traffic.
• Any volume thresholds, time limits, or regional exceptions that might limit the practical scope of the waiver.
• Whether interconnects or peering are required (e.g., must traffic flow over a provider’s direct interconnect rather than transit paths).
• The duration of the promotional or policy change and whether it is bound to the Data Act’s timeline or offered as a long‑term commercial product.
• How the waiver interacts with support tiers, premium services, and other contractual concessions.

Flag: any specific publicised numbers about expected market share gains, contract values cited in marketing, or named customer lists should be treated cautiously unless validated by vendor filings or explicit case studies — those claims are often promotional.

---

Practical playbook for IT teams and Windows admins (90‑day tactical checklist)​

• Inventory and classify workloads
• Identify the top 10 workloads by data egress volume and criticality.
• Tag services with portability sensitivity (high/medium/low).
• Map licences and contractual dependencies
• List all third‑party licences tied to current cloud runtime environments and note portability constraints.
• Establish observability for cross‑cloud traffic
• Enable flow logs, VPC flow, and interconnect telemetry to measure and prove qualifying transfers.
• Register for vendor programs and open formal support cases
• Opt into Google’s Data Transfer Essentials.
• Submit the at‑cost request to Microsoft where appropriate.
• Open AWS support channels for reduced‑rate consideration if migrating from or to AWS.
• Run a controlled pilot
• Execute a small‑scale parallel workload across clouds and capture telemetry, logs, and billing records.
• Validate that the transfer is classified as qualifying and that invoices reflect the expected treatment.
• Negotiate contractual safeguards at renewal
• Secure explicit language for egress treatment (duration, metrics, audit rights).
• Add migration assistance credits and exit clauses tied to measurable transfer conditions.
• Reassess disaster recovery and DR runbooks
• Update failover procedures to account for multicloud replication costs and operational realities, even when egress is waived.
• Review security, compliance, and data residency
• Map cross‑border transfer constraints against the Data Act’s protections and local data‑handling laws.

---

A recommended technical migration sequence (detailed steps)​

• Baseline measurement: capture current egress profiles, daily/weekly/monthly peaks, and dataset sizes.
• Design parallel pipeline: instrument the workload to run concurrently on both provider stacks with identical inputs and synchronised outputs.
• Establish interconnects: where possible, use direct interconnects or private peering to reduce unpredictability in routing and costs.
• Test data movement: execute staged batch transfers and measure throughput, latency, and error rates.
• Verify qualification: collect log evidence and billing snapshots to prove the traffic met the provider’s in‑parallel criteria.
• Cut over compute: move stateless or easily replicated services and validate functional parity.
• Decommission source components incrementally: avoid wholesale shutdowns until full parity is validated.
• Reconcile costs: run a 30‑day post‑migration reconciliation to capture any unexpected charges or reallocated costs.

---

Strategic implications: where the battleground moves next​

With egress fees constrained or waived in targeted scenarios, competition will shift to other levers:

• Managed AI services and specialised accelerators: Providers will compete on model availability, latency to GPUs/TPUs, and integrated tooling that makes one platform more attractive for ML workloads.
• Proprietary platform features: Differentiated managed databases, observability bundles, and developer productivity tools will become more decisive in workload placement.
• Licensing and contractual complexity: Regulators and customers alike have already pointed to complex licensing as the next anti‑competitive risk. Expect renewed scrutiny and negotiation focus here.
• Service quality and operational support: Premium support, migration engineering services, and migration credits will act as commercial hooks.

---

What procurement and legal teams must insist on​

• Written guarantees of egress treatment and explicit audit rights to validate billing classification.
• Clear SLAs for migration support, including engineer time and escalation paths.
• License portability clauses or compensation mechanisms where licences cannot be moved directly between clouds.
• Exit engineering support commitments (for example, credits or technical assistance for large‑scale bulk transfers).

---

Long‑term outlook and regulatory context​

The Data Act is a structural change: it sets a legal floor for fairer access and mobility in the EU data economy. Google’s commercial concession amplifies that change in the marketplace by making certain multicloud patterns materially cheaper — at least in the EU and UK. Regulators are unlikely to stop at pricing alone; the focus will broaden to licensing, interoperability standards, and standardized contractual clauses to ensure portability is more than an economic option — it becomes an operational reality.
However, market dynamics will continue to vary across jurisdictions. The UK and EU will coordinate, but differences in enforcement or interpretation are possible. Public sector buyers will watch for durable, documented concessions and clear technical definitions before relying on newly advertised commercial terms.

---

Final assessment — what to do next​

Google’s Data Transfer Essentials is a significant, customer‑friendly concession that reduces a visible barrier to multicloud experimentation and migration in Europe and the UK. For Windows administrators and IT decision‑makers, the announcement creates a timely window to:

• Run carefully scoped multicloud pilots,
• Strengthen contractual protections at renewal,
• Invest in cross‑cloud observability and telemetry,
• And re‑map license and architecture roadmaps with cost and portability in mind.

That said, don’t assume headline pricing equals full portability. Licensing complexity, proprietary service dependencies, and the technical work of migration remain real obstacles. Treat the waiver as an operational opportunity — not a panacea. Execute pilots, document evidence, and secure explicit contractual language before committing to large migrations.
The market just got a little more permissive; the job now is to convert that permissiveness into predictable, auditable outcomes that genuinely reduce lock‑in rather than simply shifting costs elsewhere.

---

Source: EconoTimes https://econotimes.com/Google-Drops-Cloud-Data-Transfer-Fees-in-EU-and-UK-Ahead-of-Data-Act-1720188/