Substantiation
How HarmonyGrid Calculates Your Savings
Every savings number on this site links back to this page. We publish the method before we publish the number, because property-management procurement asks — and because the FTC requires it.
Baseline Method
We establish your pre-optimization energy baseline using 12 months of historical utility data (via Green Button API or bill upload). For new enrollments without history, we use a weather-normalized regression model using NOAA HDD/CDD data and EIA regional consumption benchmarks for your property type.
Counterfactual
We compute what you would have paid without HarmonyGrid by projecting your baseline consumption forward using your actual utility rate schedule, adjusted for weather. This is the "do nothing" scenario.
Savings Attribution by Lever
We attribute savings to four non-overlapping mechanisms:
- Consumption reduction — HVAC setpoint optimization, pre-cooling, scheduling (measured in kWh avoided × rate).
- Time-of-use price optimization — load shifted from peak to off-peak windows on applicable TOU rates (kWh shifted × rate differential, counted only if TOU rate applies).
- Solar self-consumption — solar kWh consumed on-site rather than exported at wholesale rates (avoided export × retail-wholesale spread).
- VPP/DR revenue — grid dispatch payments received from ConnectedSolutions, Renew Home, or other programs (actual payments received, not estimated).
These four categories do not overlap. TOU optimization is only counted where a TOU rate is active and load-shifting is verified. Energy cost reduction counts consumption reduction only. VPP revenue is actual cash received.
Sensitivity
Savings estimates carry a ±20% range reflecting variation in weather, occupancy, and equipment performance. We report actuals vs. estimates monthly and update projections quarterly.
CalTRACK / OpenEEmeter Alignment
Methodology note: This approach follows CalTRACK methodology (OpenEEmeter, Apache 2.0 license) for savings verification, consistent with DOE IRA Home Energy Rebate standards.
Architecture Note
This methodology page describes the SavingsAttributionService and BaselineEngine components identified in our implementation roadmap. These components are in active development. Estimates today use regional benchmarks; we switch to customer-specific attribution in Sprint 2–3. Until then, every savings number on this site is an estimate based on regional utility benchmarks, not a customer-verified result.
Sources we cite
- EIA Electric Power Monthly, Table 5.6.A — average retail price of electricity by state and sector.
- Eversource ConnectedSolutions — $325/kW/yr residential battery program (2026 rates).
- ACEEE — 2022 rental energy consumption benchmarks.
- CalTRACK / OpenEEmeter — savings measurement methodology (Apache 2.0).
- CPUC SGIP — Self-Generation Incentive Program rates and waitlist status.