Request Access to Climatiqs data tool kit.
All posts
June 27, 2022

Climatiq’s emissions: Q1 2022 Report & Re-forecast

Climatiq’s emissions: Q1 2022 Report & Re-forecast

Earlier this year, we outlined our approach to measuring and forecasting our emissions. We decided against taking a typical annual reporting approach, with a commitment to both improving how we measured and providing full transparency to our own process of emission measurement and reduction.

After updating our carbon forecasting and reporting approach with much more granular and geographically applicable emission factors from EXIOBASE (rather than the older, global numbers from the GHG Protocol Scope 3 Evaluator), we have downgraded our forecast emissions to 17.88 tCO2e for the period September 2021-2022 (down from 24.8t).

Our actual emissions after 6 months were 9.18t against a straight-line 6-month projection of 8.94t, meaning we are 2.6% over budget at the halfway mark of the year, with the 6 coldest (and therefore most carbon intensive months) behind us. This leaves us in good stead to hit both our overall and per-employee targets.

”Hotspots” and areas for improvement are:

  • Homeworker (gas) heating emissions are 1.29t of a total 1.14t for the whole year. This comes from reporting the coldest 6 months first, but also an underestimation of how often and how long team members run their heating - a learning to take into next year’s forecast.
  • 11% over budget on flights (3.46t of the 6-month 3.1t target)
  • 120% over budget on edge-caching - which distributes our API globally to reduce latency and emissions (0.267t of 0.12t target)

Areas where we are doing well:

  • 26% under budget on train travel (0.47t of 0.64t)
  • 68% under budget on hotel stays (0.47t of 1.48t)
  • 69% under budget on database usage (0.11t of 0.34t target)

Next Steps

As we scale up our infrastructure, it will become more important to understand in detail the footprint that our emission intelligence services have - both for our own accounting and for clients who are using them. We intend to use our own services, such as the cloud computing endpoint, to understand this better, and will be working closely with partners in our tech stack such as Fastly and Fauna to provide more detailed estimates on a per-API-call emission basis.