The 15 Minutes That Set Half Your Power Bill
A single 15-minute window can decide more of your electric bill than every kilowatt-hour you used the rest of the month.
That window is your demand charge. Across 2026 commercial-rate analyses, demand charges run 30 to 50 percent of a multi-site electric bill, and they are set by the single highest 15-minute interval in the billing cycle. Spike your draw for fifteen minutes one afternoon in July, and you pay for that peak across the whole month, every month, until you beat it.
It is easy to read the bill as one number and assume the only way down is to use less. That is half the picture. The other half is when you use it.
What a demand charge actually measures
Your utility bills you two ways. Consumption is the total energy you pulled, measured in kilowatt-hours. Demand is the fastest you pulled it, measured in kilowatts, sampled in short intervals.
The demand piece exists because the grid has to be sized for your worst moment, not your average one. If your stores all crank rooftop units at the same time on the first hot afternoon of the season, the utility has to stand ready to deliver that peak. They charge you for the privilege, and the charge is keyed to the highest interval they recorded.
For a fleet, the math gets unforgiving. A few hundred locations all ramping cooling in the same hour stacks into a coincident peak that no single store would ever produce on its own.
Why summer is the expensive season
Cooling load is the driver. When outside air climbs, compressors and rooftop units pull harder, and they pull hardest in the same mid-afternoon window across every store in a region. That synchronized ramp is exactly what sets a demand peak.
Rates are not helping. U.S. commercial electricity averaged 14.12 cents per kilowatt-hour in early 2026, up 10.7 percent year over year. Many budgets assumed 2 to 3 percent and are absorbing double digits instead, with delivery and demand line items doing much of the work.
So the bill goes up two ways at once in summer: you consume more cooling, and you set higher peaks doing it.
The lever most fleets leave on the table
Here is the part that surprises people. You can cut the demand charge without cutting comfort, because the charge is about timing, not total.
Staggering matters. If every location recovers its setpoint at 2 p.m., they peak together. Spread that recovery across a window, and the coincident peak flattens. One 40-location QSR chain cut peak demand 18 percent through HVAC scheduling discipline alone, with zero capital outlay on equipment. No new hardware. Just smarter timing.
Setpoint discipline matters. Small, pre-planned setpoint shifts during the hours your utility is most likely to set the peak shave the top off the curve. A two to four degree adjustment during a peak window is usually invisible to customers and worth real money on the bill.
Visibility matters most. You cannot manage a peak you only see thirty days later on a statement. By the time the bill arrives, the expensive 15 minutes happened weeks ago and you have already locked in the charge. Interval-level data tells you which stores set the peak and when, while you can still do something about it.
Where Smart Setpoints come in
This is what GlacierGrid's Smart Setpoints and Adaptive Recovery is built for. Instead of every unit slamming back to setpoint at the same moment, recovery is sequenced and paced so the fleet does not build one coincident peak. The system reads the conditions that drive a peak and adjusts before the spike forms, not after, which is the difference between managing a demand charge and paying it.
What to do before the next bill
You do not need a capital project to start. Three moves this week:
- Pull an interval-data report from your utility for one high-cost store and find the 15-minute peak. Look at what was running.
- Check whether your locations recover setpoint on the same clock. If they do, stagger them.
- Identify your utility's peak-pricing window and pre-plan a modest setpoint shift inside it.
The stores that win on demand charges are not the ones using the least power. They are the ones that never let the whole fleet spike at once.
If you want to see where your peaks are set across your locations, a 90-day GlacierGrid pilot is free and shows you the 15 minutes that are costing you.
