The truth about cloud data egress fees: They’re not going away yet

Updated on 20/10/2026

It’s well understood that getting data into the cloud is often free. Depending on the route you take to engage with a cloud provider, there are numerous free credits, allowances and trial offers designed to make it easy to store your data with them.

Getting your data out of the cloud is an entirely different matter. Whether you’re migrating workloads to another provider, moving data between regions or transferring it to your own data centre, charges can apply. These are known as data egress fees, and the amount you pay depends on the provider, destination, volume and type of transfer.

Egress fees are often referred to as ‘hidden costs’ because they can be difficult to forecast and may only become apparent when the bill arrives. For businesses moving large volumes of data, these charges can quickly add up.

Egress fees have also attracted criticism because they can make switching cloud providers expensive, potentially contributing to vendor lock-in. Regulators have taken notice and hyperscalers have begun introducing changes to make cloud switching easier.

But while some providers have reduced or removed charges for specific use cases, egress fees have not disappeared. Understanding where they still apply is essential for businesses looking to control cloud costs.

Are hyperscalers making cloud switching easier by ditching egress fees?

Putting data into cloud storage or compute with a hyperscaler has been likened to buying a plane ticket on a budget airline. The list price is just the starting point and you can soon encounter additional charges for services such as baggage and seat allocation, or in this case for data storage, API access costs and data transfers.

Before you know it, the cost has increased, making other options look more attractive. But by this point, you may have invested heavily in your existing cloud environment and moving elsewhere can be expensive. This is one of the barriers to cloud switching that regulators have sought to address.

In the EU, the Data Act began applying on 12 September 2025. It introduces measures designed to make switching between cloud and other data processing services easier, including requirements to reduce contractual and technical barriers. However, the removal of switching charges, including data egress fees associated with switching, is subject to a transition period and is due to take full effect on 12 January 2027. Until then, providers may still charge for eligible costs incurred during switching.

Cloud providers have also introduced their own measures. In September 2025, Google Cloud launched Data Transfer Essentials at no cost for qualifying multi-cloud data transfers for customers in the UK and EU. The service is designed for eligible workloads running across multiple cloud providers, rather than every type of outbound transfer.

These developments are a step in the right direction for businesses that want greater freedom to choose where their workloads run. However, the details matter. Eligibility requirements, supported services and traffic types vary and charges can still apply outside the scope of these programmes.

Egress fees therefore remain an important consideration, particularly for organisations with large or recurring data transfers.

What cloud providers are changing about egress fees

The approach to data transfer charges differs between hyperscalers, so businesses should not assume that a policy introduced by one provider will apply to another.

Google Cloud’s Data Transfer Essentials programme allows qualifying traffic to be transferred between supported cloud environments without outbound data transfer charges. However, businesses must configure the service correctly, and other traffic continues to be billed at the applicable rates. The programme is not intended to cover every external data transfer or every mission-critical connectivity requirement.

Meanwhile, AWS provides a free data transfer allowance and allows eligible customers covered by its EU Data Act and UK customer switching provisions to request reduced data transfer rates for qualifying use cases. Standard data transfer charges can still apply, depending on the service and type of traffic.

For businesses using Microsoft Azure, the same principle applies: check the applicable pricing, contractual terms and eligibility requirements rather than assuming that switching-related provisions cover all data transfers.

The important distinction is between charges associated with switching providers and the ongoing costs of running a multi-cloud environment. A business may qualify for switching-related relief during a migration but still incur charges when regularly transferring data between cloud platforms, regions or on-premises infrastructure.

For now, businesses should review their cloud provider’s current policies and model their costs around their actual workloads, rather than relying on announcements about egress fee removal.

True costs of data egress from the cloud

Each hyperscaler has its own pricing framework for data transfers. Charges vary according to the volume of data being moved, where it originates, where it goes and which services are involved. Transfers between availability zones, within regions and across geographical regions can all have different pricing.

For this reason, there is no single per-gigabyte rate that accurately represents the cost of cloud egress across all providers and scenarios. Businesses should use current provider pricing calculators and rate cards to estimate the costs associated with their specific data flows.

For smaller organisations with limited data transfers, these fees may have little impact on the overall cloud bill. For larger enterprises moving terabytes of data regularly, however, they can become a significant operational expense.

The growing use of AI, analytics and distributed applications adds another consideration. Training data, model outputs, application data and replicated datasets may need to move between cloud environments, regions and data centres. Where these transfers happen frequently, even relatively small per-gigabyte charges can accumulate over time.

The answer is not necessarily to avoid the cloud or consolidate every workload with one provider. Instead, businesses need to understand how their data moves, identify where transfer costs are incurred and assess whether their connectivity model is still the most cost-effective option.

Private connectivity can form part of that assessment, particularly where organisations need to move large volumes of data between cloud platforms and their own infrastructure.

Using private connectivity to manage egress costs

For organisations moving substantial volumes of data, private connectivity can offer many benefits in addition to cost savings. By providing a more direct route between environments, it can help reduce exposure to public internet congestion and deliver more consistent network performance, including latency and jitter. It can also support security requirements by keeping traffic on private network paths, alongside appropriate encryption and access controls.

Software Defined Cloud Interconnect (SDCI) solutions such as Console Connect offer another advantage: flexibility. Rather than relying solely on a private connection tied to a specific cloud provider or data centre ecosystem, businesses can establish connectivity across different cloud platforms and locations.

Console Connect enables users to provision Layer 2 and Layer 3 connections to major cloud providers, including AWS, Google Cloud and Microsoft Azure, over its global network.

Its on-demand model allows businesses to activate dedicated connectivity when needed, making it suitable for use cases such as cloud-to-cloud migration, data replication and temporary high-volume transfers. Rather than maintaining a connection at a fixed capacity indefinitely, businesses can align connectivity with their requirements and pay for the services they use.

For example, a cloud-to-cloud migration case study describes how a customer used CloudRouter® to reduce egress costs by more than 50% compared with the public connectivity option assessed in the case study. The connection was used for three weeks, allowing the customer to complete the migration before disconnecting. The results demonstrate the potential savings in a specific scenario, rather than a guaranteed outcome for every business.

Ultimately, removing egress fees for certain transfers does not eliminate the need to manage cloud networking costs. Businesses still need visibility into where their data travels, what their providers charge and how their connectivity choices affect performance and expenditure.

Get in touch to discuss the best ways to transfer your data between environments and manage your cloud infrastructure costs.