How Cloud Egress Fees Work
Cloud providers generate revenue not just from compute, storage, and managed services — they also charge for the movement of data. Egress fees apply when data leaves the cloud provider's network boundary. The primary egress scenarios that affect enterprise buyers are:
- Internet egress: Data transferred from cloud storage or compute instances to the public internet — typically to end users, downstream systems, or partner organisations. This is the highest-volume egress category for most enterprises.
- Cross-region egress: Data transferred between regions within the same cloud provider — e.g., from AWS us-east-1 to eu-west-1. These fees are lower than internet egress but significant for distributed architectures with data replication across regions.
- Cross-provider egress: Data transferred from one cloud provider to another — the most expensive scenario per GB and the direct cost of switching primary cloud provider.
- On-premises egress: Data transferred from cloud to an organisation's on-premises or co-located infrastructure. Relevant for hybrid architectures where data moves bidirectionally between cloud and data centre.
Published egress rates at the major hyperscalers are broadly similar: AWS charges up to $0.09/GB for internet egress (with volume discounts above 10TB/month), Azure up to $0.087/GB, and GCP up to $0.08/GB. These rates are applied to gross data transfer volume — there is no standard allowance for replicated, cached, or redundant data.
Why Egress Fees Are Designed This Way
Cloud providers invest hundreds of billions in global network infrastructure — fibre, undersea cables, Points of Presence, Content Delivery Networks. Egress fees partially reflect the genuine cost of this infrastructure. But the pricing is not cost-based — it is strategic. Egress fees are calibrated to create switching costs that increase proportionally with the amount of data you store in the cloud. The more you use the cloud, the more expensive it becomes to leave it.
This dynamic is well understood by cloud providers' commercial teams. They are not going to volunteer to reduce your switching costs — you have to negotiate it. See our complete cloud contract guide and the Cloud Contract Framework white paper for the full commercial framework.
Calculating Your True Egress Exposure
Before you can negotiate egress terms, you need to understand your actual and projected egress exposure. Most enterprise cloud teams have incomplete visibility into their egress costs because cloud billing dashboards aggregate egress into a single line item rather than breaking it down by source, destination, and scenario.
A true egress exposure analysis requires:
- Current monthly egress volume by category. Pull your cloud billing data and segment egress by type: internet, cross-region, cross-provider, and on-premises. Most enterprises are surprised to find that cross-region egress — driven by data replication and distributed architectures — is a material cost centre.
- Egress growth trajectory. Model your egress cost trajectory over the contract term. Organisations moving workloads to the cloud during an EDP term often see egress costs double or triple as more data is accumulated in cloud storage.
- Switching cost calculation. Quantify the one-time egress cost of moving all cloud data to an alternative provider or on-premises environment. Multiply your total stored data volume (in TB) by the applicable egress rate. This is the financial ceiling on your ability to switch cloud providers — the number you need to negotiate down before signing any long-term commitment.
- Multi-cloud egress flows. If you operate workloads across multiple cloud providers, map the data flows between providers. Inter-cloud egress is the highest-cost category and is almost never optimised.
"A global media company we advised had accumulated 12 petabytes across AWS S3 without ever calculating the switching cost. The theoretical exit bill was $1.1 million. That number — which we surfaced before any EDP renewal — was the starting point for a negotiation that delivered egress waiver provisions and a 28% core discount."
EU Data Act: The New Regulatory Baseline
The EU Data Act, which came into full force in September 2025, fundamentally changed the regulatory landscape for cloud egress fees in Europe. Article 25 of the Data Act requires cloud providers to ensure that customers can switch to another provider without facing switching charges — including egress fees — that prevent or unreasonably impede switching.
In practice, this means that AWS, Azure, and GCP operating in the EU must offer EU-based customers the ability to transfer all their data to another provider or to on-premises without incurring egress charges. The providers have implemented this through updated EU contract terms and a waiver application process for customers invoking their Data Act portability rights.
Leveraging the EU Data Act in Non-EU Negotiations
For organisations operating outside the EU, the Data Act does not apply directly. However, it creates a powerful commercial precedent. If cloud providers have accepted — under regulatory pressure — that charging egress fees to customers who want to switch is unreasonable in the EU, then the same argument can be made commercially in non-EU negotiations.
We have successfully used the EU Data Act as a commercial reference in North American and APAC enterprise negotiations: "You have committed to providing egress-free data portability for your EU customers under the Data Act. We are requesting equivalent provisions in this agreement. If the position is commercially reasonable for the EU, it is commercially reasonable here." This framing moves the conversation from 'this has never been done' to 'this is your standard EU practice' — and is significantly more effective.
Negotiating Egress Waivers and Caps
Egress negotiation requires a different approach from core discount negotiation. Cloud providers' commercial teams have limited authority to waive egress fees unilaterally — it typically requires network or infrastructure team approval alongside commercial approval. This means the negotiation timeline is longer and the framing needs to be more precise.
Egress Waiver for Switching Scenarios
The most valuable egress provision is a contractual waiver of egress fees if you choose to switch providers at the end of the commitment term. Frame this as a data portability right: "At the end of this agreement, we require the right to transfer all data from [provider's] infrastructure to our designated destination without incurring egress fees." This is increasingly achievable, particularly for EU-operating customers or for those referencing the EU Data Act baseline.
Monthly Egress Caps
An alternative to full waiver is a monthly egress cap — a maximum monthly egress fee beyond which no additional egress charges are applied. This provides cost predictability for architectures with variable or high-volume internet egress. Monthly egress caps of 50–70% of projected peak egress costs are achievable in EDP negotiations for commitments above $10M annually.
Reduced Egress Rates as Part of EDP
The most common egress concession in enterprise cloud negotiations is a negotiated reduction in the egress rate — moving from the published $0.09/GB to $0.05–$0.06/GB as part of the EDP structure. This does not eliminate egress as a switching cost but materially reduces the ongoing cost burden and reduces the switching cost calculation. For high-volume internet egress scenarios (streaming, large-scale data distribution), a 40% reduction in the egress rate can save $1M+ annually at scale.
Egress Included in Commitment
Negotiate for egress charges to count toward your EDP or MACC commitment minimum. By default, AWS and Azure may exclude egress from the commitment calculation — meaning high-volume egress reduces your progress toward the minimum spend threshold. Ensuring egress counts toward the commitment makes meeting the minimum easier and reduces shortfall risk.
Data Portability Rights in Cloud Contracts
Egress fee negotiation is one component of a broader set of data portability provisions that should appear in every enterprise cloud agreement. Complete data portability rights include:
- Format portability: The right to export all data in open, non-proprietary formats — not just in the provider's native format. Particularly important for managed database services where data may be stored in provider-specific schemas.
- API portability: Documentation of all APIs used to access your data, with sufficient detail to replicate access through a different provider's interface.
- Metadata portability: The right to export metadata, configuration, and infrastructure definitions (IAM policies, network configurations, access controls) in addition to raw data.
- Assisted migration: A contractual right to request technical assistance from the cloud provider for a data migration — particularly valuable for large or complex data estates where DIY migration is technically challenging.
- Transition period access: The right to continue accessing your data for a defined period (typically 90–180 days) after contract termination, without additional charges, to facilitate an orderly migration.
Architecture Choices That Reduce Egress Risk
Contract negotiation can reduce egress costs, but architectural choices determine egress exposure in the first place. For organisations designing new cloud architectures or migrating from on-premises, building egress efficiency into the architecture from the outset is more effective than trying to negotiate away high egress costs after the fact.
Key egress-reducing architectural principles include keeping data processing close to data storage (avoiding moving data across regions for processing), using Content Delivery Networks to cache and serve frequently accessed data at the edge (dramatically reducing origin egress), designing APIs to return only the data fields required rather than full records (reducing application-layer egress), and standardising on open data formats from the outset to reduce the cost of future migration.
Our cloud contract negotiation service includes a commercial architecture review that identifies egress reduction opportunities alongside contract term improvements — because the most effective cloud cost strategy combines good architecture with well-negotiated commercial terms.