ESGx Buildings is Measurabl’s scalable ESG intelligence solution that delivers asset-level environmental performance estimates across large portfolios—no direct data submission required. By leveraging Measurabl’s proprietary machine learning models and a robust database of 60,000+ buildings covering billions of square feet, ESGx Buildings provides instant visibility into estimated energy use, carbon emissions, and ESG characteristics for commercial real estate.
This product is designed for capital markets participants, lenders, and asset managers who need standardized, location-based ESG insights to support screening, benchmarking, and portfolio evaluation.
What data is included in ESGx Buildings?
Each delivery includes property-level data for each asset analyzed, including:
- Estimated energy consumption
- Estimated carbon emissions
- Presence of green building certifications
- Exposure to local ordinances
These insights enable deep analysis of ESG characteristics across large portfolios, without requiring building owners to share operational data.
📌 Note: ESGx Buildings provides results at the individual asset level. For company-level ESG metrics, see Measurabl’s separate product: ESGx Securities.
How are the estimates calculated?
The same validated machine learning models used in Measurabl’s Diligence product are applied to ESGx Buildings. These models require only four inputs:
- Property address
- Property type
- Gross floor area
- Year built
They are trained on more than 18 billion square feet of utility-billed building data and continuously retrained to reflect recent performance patterns.
How accurate are the estimates?
Model validation includes:
- Over 90% R-squared, indicating strong alignment between predicted and actual energy use
- Confidence scores based on average error margins by property type
Example: If actual energy use is 100 units, a high-confidence prediction would typically fall between 80–120 units.
- R-squared reflects how well the model tracks overall patterns.
- Error margin shows how precise each estimate is likely to be.