Carbon & Cost Savings Methodology

Last updated: July 8, 2026

GreenOps helps you identify cloud infrastructure waste and estimate the associated cost and carbon impact. The figures shown in the platform are estimates intended for prioritization and planning, not audited financial or carbon accounting statements.

1. Cost savings estimates

What we measure

Resource usage metrics from AWS APIs (CPU, memory, disk, network, request counts).
AWS pricing data for the resource type and region.
Resource configuration (instance type, storage class, provisioned capacity, etc.).
Usage patterns over time (e.g., idle periods, low utilization).

How we estimate savings

For each finding, we compare the current resource state with a recommended state. For example:

Right-sizing an over-provisioned EC2 instance to a smaller instance type.
Moving infrequently accessed S3 data to a lower-cost storage class.
Deleting unattached EBS volumes or idle load balancers.
Scheduling non-production resources to stop outside business hours.

The estimated monthly savings is the difference in cost between the current state and the recommended state, using on-demand pricing where applicable. Reserved Instance, Savings Plan, or spot pricing discounts are not applied unless explicitly configured.

Limitations

Savings are potential; actual savings depend on whether the recommendation is implemented and whether usage patterns remain stable.
We do not account for every AWS pricing dimension (e.g., data transfer, IOPS, request charges may be simplified).
Pricing data is refreshed periodically and may not reflect real-time AWS rates.
Recommendations do not consider application-level constraints or business requirements.

2. Carbon emissions estimates

What we measure

Estimated electricity consumption of cloud resources based on usage metrics and cloud provider energy models.
Regional grid carbon intensity factors, typically from public sources such as the International Energy Agency (IEA) or equivalent third-party datasets.
Cloud provider-provided average PUE (Power Usage Effectiveness) data for data centers.

How we estimate carbon impact

We convert resource usage into estimated kilowatt-hours (kWh) using simplified cloud energy models, then multiply by the regional grid carbon intensity to estimate kilograms of CO₂ equivalent (CO₂e).

For example, an idle EC2 instance running in a region with high carbon intensity will show a higher estimated CO₂e impact than the same instance in a low-carbon region.

Limitations

Carbon figures are modeled estimates, not measured emissions.
We use average emission factors, which may not reflect the exact mix of energy sources feeding a specific data center at a specific time.
We do not perform life-cycle assessments (LCA) of hardware manufacturing or disposal.
Carbon estimates are not a substitute for a formal carbon audit or GHG Protocol inventory.
Cloud provider renewable energy purchases or carbon offsets are not reflected in these estimates unless the underlying emission factor dataset accounts for them.

3. How the CLI scanner works

By default, the CLI scanner (npx greenops-scan) runs locally on your machine. It reads metadata from your configured AWS profile using read-only APIs and produces a report on your laptop. No cloud data is sent to the GreenOps platform unless you explicitly connect the account.

When you connect an account to the GreenOps platform, the same read-only metadata is stored so that findings can be tracked, resolved, and reported over time.

4. Recommended use

GreenOps findings are intended to:

Prioritize optimization opportunities.
Estimate the scale of possible cost and carbon reductions.
Support conversations with engineering, finance, and sustainability teams.
Provide inputs for broader CSRD, GHG Protocol, or sustainability reporting processes.

They are not intended to:

Guarantee specific savings or emission reductions.
Replace formal cloud cost audits or carbon inventories.
Serve as the sole basis for financial or compliance decisions without independent verification.

5. Updates and changes

This methodology will evolve as we improve our models, add more cloud providers, and incorporate better data sources. Significant changes will be noted in this document and, where appropriate, communicated to users.

6. Questions

If you have questions about how a specific estimate is calculated, contact us at privacy@greenops.cloud.

Carbon & Cost Savings Methodology

Last updated: July 8, 2026

1. Cost savings estimates

What we measure

Resource usage metrics from AWS APIs (CPU, memory, disk, network, request counts).
AWS pricing data for the resource type and region.
Resource configuration (instance type, storage class, provisioned capacity, etc.).
Usage patterns over time (e.g., idle periods, low utilization).

How we estimate savings

For each finding, we compare the current resource state with a recommended state. For example:

Right-sizing an over-provisioned EC2 instance to a smaller instance type.
Moving infrequently accessed S3 data to a lower-cost storage class.
Deleting unattached EBS volumes or idle load balancers.
Scheduling non-production resources to stop outside business hours.

Limitations

Savings are potential; actual savings depend on whether the recommendation is implemented and whether usage patterns remain stable.
We do not account for every AWS pricing dimension (e.g., data transfer, IOPS, request charges may be simplified).
Pricing data is refreshed periodically and may not reflect real-time AWS rates.
Recommendations do not consider application-level constraints or business requirements.

2. Carbon emissions estimates

What we measure

Estimated electricity consumption of cloud resources based on usage metrics and cloud provider energy models.
Regional grid carbon intensity factors, typically from public sources such as the International Energy Agency (IEA) or equivalent third-party datasets.
Cloud provider-provided average PUE (Power Usage Effectiveness) data for data centers.

How we estimate carbon impact

For example, an idle EC2 instance running in a region with high carbon intensity will show a higher estimated CO₂e impact than the same instance in a low-carbon region.

Limitations

Carbon figures are modeled estimates, not measured emissions.
We use average emission factors, which may not reflect the exact mix of energy sources feeding a specific data center at a specific time.
We do not perform life-cycle assessments (LCA) of hardware manufacturing or disposal.
Carbon estimates are not a substitute for a formal carbon audit or GHG Protocol inventory.
Cloud provider renewable energy purchases or carbon offsets are not reflected in these estimates unless the underlying emission factor dataset accounts for them.

3. How the CLI scanner works

When you connect an account to the GreenOps platform, the same read-only metadata is stored so that findings can be tracked, resolved, and reported over time.

4. Recommended use

GreenOps findings are intended to:

Prioritize optimization opportunities.
Estimate the scale of possible cost and carbon reductions.
Support conversations with engineering, finance, and sustainability teams.
Provide inputs for broader CSRD, GHG Protocol, or sustainability reporting processes.

They are not intended to:

Guarantee specific savings or emission reductions.
Replace formal cloud cost audits or carbon inventories.
Serve as the sole basis for financial or compliance decisions without independent verification.

5. Updates and changes

6. Questions

If you have questions about how a specific estimate is calculated, contact us at privacy@greenops.cloud.