Egress Fees

What are Egress Fees?

Egress fees are charges cloud providers apply when data is transferred out of their network, such as to the internet, another cloud or an on-premises data center, and sometimes between regions or zones.

Updated November 2026 · 5 min read

Egress fees are what cloud providers charge for data leaving their network. Moving data into the cloud, known as ingress, is usually free, but moving it out to users on the internet, to another cloud provider or to an organization's own data center is charged per gigabyte. Providers also commonly charge for transfers between their own regions and, in some cases, between availability zones within a region. For applications that serve large amounts of content or move data frequently between locations, these charges can become a significant part of the cloud bill.

Egress fees are often overlooked when workloads are designed or migrated, because they depend on how data flows rather than on the resources being used. They also make it more expensive to move data away from a provider, which contributes to concerns about vendor lock-in. For IT financial management, egress fees are a cost that needs specific visibility and design attention, and an important factor in multi-cloud and hybrid strategies.

Definition of Egress Fees

Egress fees are usage-based charges for outbound data transfer from a cloud provider's network or between its locations. They are typically priced per gigabyte, often with tiered rates that decrease at higher volumes and a small free allowance each month.

Data transfer charges are part of cloud billing and are usually reported under networking or data transfer categories, separately from the compute and storage resources that generate the traffic.

Types of Data Transfer Charges

Internet egress covers data sent from the cloud to users and systems on the public internet, such as web pages, video streams, file downloads and API responses.

Inter-region transfer covers data moved between a provider's regions, for example for replication or disaster recovery. Inter-zone transfer covers data moved between availability zones within a region, which some providers charge for, affecting highly available architectures.

Transfers to other clouds or to on-premises data centers are generally charged as internet egress, unless they use dedicated connections that have their own pricing.

Recent Changes

In response to regulatory pressure, including the European Union's Data Act, and to customer demand, major providers have introduced options to waive egress fees for customers moving their data off the platform entirely. These waivers usually have conditions and apply to switching providers rather than to ongoing data transfer.

Content delivery networks, private connections and certain partner arrangements can also reduce egress costs for regular traffic.

Because pricing and waiver terms change, organizations should check current provider terms when planning architectures or migrations.

Importance of Egress Fees in IT Financial Management

Egress fees can grow quietly as applications attract more users or as data pipelines move more data between locations. Because they are not tied to a specific resource, they are often missed in cost reviews.

They also influence strategic decisions. Decisions about where to place workloads, which providers to use and how to design data platforms all depend partly on what it costs to move data.

The cost of moving data out of a provider affects the economics of multi-cloud architectures, hybrid deployments and future migrations.

Hidden Cost Growth

For content-heavy applications, such as video platforms, software downloads or data services, egress can be one of the largest cost categories. Growth in users or data volumes increases egress costs directly.

Architectural choices, such as replicating data across regions or splitting services across zones, can generate steady transfer charges that were not considered when the design was chosen.

Allocating egress costs to the applications that generate them, as part of cloud cost allocation, helps teams understand and manage them.

Lock-In and Strategy

High egress fees make it costly to move large datasets from one provider to another or back on premises. This can discourage organizations from changing providers or using several, contributing to vendor lock-in.

In multi-cloud cost management, architectures that move data frequently between providers can incur substantial ongoing transfer costs.

Business cases for migrations and for cloud repatriation should include the one-time cost of moving data and any ongoing transfer costs.

How to Manage Egress Fees

Managing egress fees involves making data transfer costs visible, designing architectures that minimize unnecessary transfer and using lower-cost delivery options.

Engineering teams play the central role, since most egress is determined by architecture and application design. The FinOps team's role is to make the costs visible and help teams compare options.

Visibility

Break down data transfer costs by type, such as internet, inter-region and inter-zone, and by application or team. Detailed billing data shows which resources and services generate transfer charges.

Use network flow logs and monitoring tools to understand which applications and data flows drive the largest transfers. Start with the largest flows, which usually account for most of the cost.

Include expected data transfer costs in architecture reviews and cost estimates for new workloads. Data transfer is easier to design for at the start than to reduce after a system is running.

Reducing Transfer Costs

Use content delivery networks to cache and serve frequently accessed content closer to users, which usually reduces both egress costs and latency. Content delivery network pricing should still be compared with direct egress for each workload.

Keep data and the services that process it in the same region and, where practical, the same zone. Compress data and avoid repeatedly transferring the same data between locations.

For large, steady transfers between cloud and on-premises environments, consider dedicated connections, which may offer lower per-gigabyte rates than internet egress.

Examples of Egress Fees in IT Financial Management

The examples below show egress fees being identified and reduced.

Both use illustrative figures. The first concerns internet egress, and the second concerns transfer between regions.

In each, understanding data flows was the key to reducing costs. The changes were technical, but the trigger was cost visibility.

Serving Downloads

A software company serves product downloads directly from cloud storage, generating 500 terabytes of internet egress a month at an average of $0.06 per gigabyte, about $30,000 a month.

It places a content delivery network in front of the storage, which caches popular files and charges a lower effective rate. Monthly delivery costs fall to about $18,000, and download speeds improve for users. The change takes a few days to implement and pays for itself in the first month.

Cross-Region Replication

Another company replicates a large analytics dataset between two regions every night for reporting, transferring about 200 terabytes a month at $0.02 per gigabyte, roughly $4,000 a month, plus storage in both regions.

Analysis shows the second region's reports could run in the primary region. Users of the reports see no difference in speed. The company stops replicating, saving the transfer and duplicate storage costs, and keeps a smaller backup copy for disaster recovery.

Conclusion

Egress fees are charges for data transferred out of a cloud provider's network or between its regions and zones. They are often overlooked but can become significant for content-heavy or data-intensive applications, and they affect the cost of using multiple providers or leaving one.

In IT financial management, managing egress fees requires visibility into data transfer costs and attention to architecture. Allocating transfer costs to applications, reviewing data flows, using content delivery networks and keeping data close to where it is processed help organizations control these costs.

Cloud Billing

Cloud billing is the process by which cloud providers measure customers' usage of services, apply prices and discounts, and issue invoices and detailed billing data, usually on a monthly cycle.

Cloud Cost Allocation

Cloud cost allocation is the process of assigning cloud costs to the teams, products, environments and business units that use them, using account structures, tags and rules for dividing shared costs.

Vendor Lock-In

Vendor lock-in is a situation in which an organization becomes so dependent on a provider's products, services or data formats that switching to an alternative would be costly, slow or risky.

Multi-Cloud Cost Management

Multi-cloud cost management is the practice of tracking, allocating and optimizing costs across two or more public cloud providers, using consistent data, policies and reporting to give a single view of spending.