Chicago, Illinois, USA
#00160
Case study of
#00167 Attribute cloud emissions from provider billing and usage data across multiple providers
Implementer
Thoughtworks (open-source project and sponsor)
Timeline
Since Mar 1, 2021
Location
Description
Cloud Carbon Footprint (CCF) is an open-source, multi-cloud tool launched by Thoughtworks in March 2021. It reads itemised billing and usage data from AWS, Google Cloud, Azure and Alibaba, converts it to energy using heuristics such as Etsy's Cloud Jewels, then applies region-specific carbon intensity, covering scope 2 and scope 3 (embodied) emissions. The tool is modular: React dashboard, Node/Express backend, CLI and API, plus an on-prem package. Its methodology is published and reportedly reviewed by PwC. In Thoughtworks' own internal deployment, the tool surfaced an emissions spike traced to Google Cloud Composer, with three GCP projects contributing roughly 90% of total emissions; the responsible teams then tuned those workloads and the spike narrowed substantially.
Funding
Lessons learned
- A single consistent methodology across clouds is worth more for decision-making than each provider's more accurate but non-comparable native figure — use CCF for cross-cloud baselining and trend-spotting rather than exact accounting.
- Tagging and labelling cloud resources before deploying CCF is what turns an aggregate emissions number into an actionable, team-level hotspot; without it the output is not actionable.
- Billing-based estimation trades precision for coverage and immediacy; the appropriate use case is identifying relative hotspots, not regulatory-grade exact accounting.
Links
1Documented Jul 13, 2026