# Outputs and Interpretation | Embodied Emissions methodology
Source: https://greenpixie.com/docs/methodology/cloud-embodied-emissions/outputs-and-interpretation

Primary outputs, intended use, and interpretation guidance.

## 11.1 Primary Outputs [#111-primary-outputs]

The methodology produces the following metric for each unit of cloud usage:

* **Embodied emissions**, expressed in tonnes CO₂e per usage unit of the SKU. The estimated manufacturing, transport, and end-of-life emissions of the hardware attributable to the unit of cloud usage, together with an allowance for the data centre shell.

## 11.2 Intended Use [#112-intended-use]

These outputs are intended to support organisational sustainability reporting under frameworks such as the Corporate Sustainability Reporting Directive, Streamlined Energy and Carbon Reporting, and SEC climate disclosure; Scope 3 emissions measurement; instance family and processor selection; rightsizing and workload placement decisions; and comparative analysis of the embodied emissions of cloud hardware across instance families, hardware generations, and cloud providers.

## 11.3 Interpretation Guidance [#113-interpretation-guidance]

Outputs are estimates calculated from published hardware specifications, published life cycle assessments, and billing data, and reflect the best available information at the time of calculation. Actual emissions from the manufacture of production cloud hardware may vary.

An embodied emissions figure is a share of the embodied emissions of hardware the customer does not own, apportioned to the capacity a unit of usage occupies and amortised across the operating life the hardware is assumed to have, together with an allowance for the data centre shell. The figure covers the period of use.

Three points bear on how an individual figure is read, being the modelled host configuration, the derivation used where no die area is published, and the inferred or averaged component generation. The host configuration a figure is apportioned against is modelled, as described in Section 6.3. The model covers both the memory population and the drives of the host. Where no die area is published for a processor, the die area is estimated from the process node and the thread count, as described in Section 7.2. The figure for that instance carries a wider uncertainty than one built on a published die area. Where the memory or NAND flash generation of a part is inferred, the figure is calculated at the die density of the inferred generation. For a part whose memory generation is not known, the figure is calculated from an average taken across memory generations, as described in Section 6.4.

Embodied emissions are one component of the environmental impact of a cloud workload. The operational electricity of the same workload is calculated by the [*Greenpixie Methodology for Cloud Compute Energy Measurement*](/methodology/cloud-compute). Its facility overhead, grid carbon intensity, and water consumption are calculated by the [*Greenpixie Methodology for Cloud Emission Measurement*](/methodology/cloud).

## 11.4 Non-Intended Use [#114-non-intended-use]

The outputs should not be treated as certified measurements, contractual guarantees, or regulatory compliance instruments unless explicitly agreed in writing. They should not be read as the embodied emissions of hardware owned by the customer, or as a total for a cloud workload. They should not be used as the sole basis for carbon offsetting calculations without independent verification.