Commercial electric bills split into energy (kWh) and demand (kW)—the highest average power draw your facility imposed on the grid in a billing window, often fifteen minutes. You can use less total energy month over month and still pay more if everything starts at once. That is demand charge economics, and it punishes warehouses, shops, and food service harder than homeowners ever see.
How utilities measure peak kW
Interval meters record power every fifteen minutes (sometimes five). The highest interval average becomes your billing demand ratchet in many tariffs—meaning one bad lunch rush in July can haunt you eleven months. Ratchets are tariff-specific; read the rider.
Concurrent startup is the killer
HVAC blowers, compressors, conveyor motors, and DC fast chargers overlapping create a spike far above steady-state run amps. Soft starters, VFDs, and staged sequencing spread the peak without reducing production throughput if engineered correctly.
Smoothing the load profile
- Stagger charger schedules for fleet vehicles
- Pre-cool buildings before on-peak hours, then coast
- Shift batch processes to off-peak where quality allows
- Monitor power factor—inductive loads can incur penalties on some schedules
On-site storage and peak shaving
Batteries discharge during forecast peak intervals to cap grid draw. ROI depends on demand charge dollars per kW vs battery capex—often compelling above twenty dollars per kW-month in congested markets.
Operational visibility
Submetering by line or machine identifies which asset owns the spike. Without data, facilities blame "the rate" instead of the chiller sequence.
Contract and tariff hygiene
When renegotiating supply contracts or switching tariffs, model demand separately from energy. A slightly higher energy rate with lower demand charges can win for spiky loads.
Demand charges reward smooth operations, not just efficient ones. Map your fifteen-minute peaks, kill concurrent startups, and treat kW like dollars on the loading dock.