Load factor electricity: how to calculate and improve it

Load factor is the ratio of the energy a facility actually used in a period to the energy it would have used had it run at its peak demand for every hour of that period. The concept, and why it drives the effective price of electricity, is covered in load factor: how the shape of your demand drives cost. This article is the working guide: the formula, worked examples straight off a bill, monthly versus annual figures, what the published ranges mean, and the levers that move the number.

The US Energy Information Administration defines load factor as the ratio of the average load to peak load during a specified time interval. Everything below is that one ratio, computed from data every commercial account already receives on its monthly bill.

The load factor formula

Load factor equals kWh used in the period divided by (peak kW in the period times hours in the period). Hours are the days in the billing period times 24, so a 30 day bill has 720 hours and a 31 day bill has 744. Peak kW is the metered maximum demand, which for most utilities is the single highest 15-minute average power draw during the billing period. Use the metered peak, not the billed demand, if a ratchet has lifted the billed figure above what was actually drawn; the ratio is meant to describe how the site ran, not how it was priced.

Average demand is the same calculation from another angle: kWh divided by hours gives average kW, and average kW divided by peak kW is the load factor. Both routes give the same answer, so use whichever your data makes easier.

A worked example from a monthly bill

Take a bill that shows 180,000 kWh of consumption, a metered peak of 600 kW and a 30 day period. Hours are 30 times 24, or 720. Peak kW times hours is 600 times 720, or 432,000 kWh: the energy the site would have used had it held its peak for the whole month. Load factor is 180,000 divided by 432,000, which is 0.417, or 41.7 percent. Average demand is 180,000 divided by 720, or 250 kW, and 250 divided by 600 gives the same 41.7 percent.

The third row is the trap in month-over-month comparisons: nothing about the site changed, but a longer period lowered the ratio. Always compute with the actual days on the bill, never a nominal 30. The fourth row is the other trap: the ratio improved, but the bill went up, because more energy was bought. A ratio has two halves, and only one of them saves money on its own.

Annual load factor vs monthly load factor

Annual load factor uses the year's total kWh, the single highest peak of the year and 8,760 hours (8,784 in a leap year). If the site above used 2,000,000 kWh across the year and its highest monthly peak was 650 kW, annual load factor is 2,000,000 divided by (650 times 8,760), or 35.1 percent. That is lower than most of its monthly figures, and it should be: the annual number is held down by the one worst peak, usually a summer cooling afternoon, while every monthly number resets to its own peak.

Use both. Monthly load factor shows operational drift and is the one to trend on a dashboard. Annual load factor is the one that matters for tariff selection and for demand ratchet clauses, because both are governed by that single worst peak.

What counts as a good load factor

Reference bands differ by source, so treat them as orientation rather than targets. One commercial rate comparison site publishes typical values by building type: data centers around 88 percent, manufacturing 68, hospitals 65, offices 42, retail 38 and schools 30. An energy management software vendor places most offices, restaurants and schools with proper nighttime and weekend setbacks in the 40 to 60 percent range, with 80 percent and above indicating steady round-the-clock demand, and values under 20 percent pointing to spiky, intermittent loads such as field lighting or irrigation.

Two readings follow. A 30 percent load factor at a school is normal, and a 30 percent load factor at a cold storage plant is a fault or a metering error. And a high load factor in an office is not automatically good news: it can mean equipment that never shuts off. Load factor is a diagnostic, and the diagnosis depends on what the building is supposed to do.

Why a low load factor raises the cost per kWh

Commercial tariffs bill energy per kWh and demand per kW of monthly peak. A national laboratory survey of more than 10,000 US utility tariffs found that demand charges typically range from 30 to 70 percent of a commercial customer's electric bill, and that nearly 5 million commercial customers can subscribe to tariffs with demand charges above 15 dollars per kW. The demand charge is fixed by the peak; the number of kWh it is spread across is set by the load factor.

Put numbers on the example above. Assume, for illustration, an energy rate of 10 cents per kWh, and take a demand rate of 12 dollars per kW. As billed, energy costs 18,000 dollars and demand costs 600 times 12, or 7,200 dollars: 25,200 dollars for 180,000 kWh, or 14.0 cents per kWh all-in. Trim the peak to 450 kW and demand falls to 5,400 dollars, the total to 23,400 dollars and the blended rate to 13.0 cents. Same energy, same tariff, about 7 percent less cost, and the only thing that changed was the ratio.

The same source that publishes the building-type table summarizes the effect bluntly: a business at 35 percent load factor pays roughly twice the demand cost per kWh as one at 70 percent. Ratchets make it worse. The most common ratchet locks minimum billing demand at 80 percent of the highest recorded peak for up to 12 months, so one bad month lowers the effective load factor on eleven more.

The practical levers

• Schedule around the peak. Move flexible loads (batch processes, EV fleet charging, water heating, ice making, test runs) out of the hours when the building's own peak forms. This lowers the denominator without touching consumption.
• Stagger start-ups. Chillers, air handlers, ovens and compressors that all energize at 6 a.m. create a peak that lasts one interval and costs a month. Sequenced starts, soft starters and optimum-start logic in the controls spread the same work over more intervals.
• Set a demand limit in the control system. A demand-limiting routine that sheds non-critical load when the running 15-minute average approaches a setpoint caps the peak directly.
• Add storage where the peak is short. Batteries or thermal storage discharge during the peak interval and recharge off-peak; the tariff survey cited above was written to size exactly that market.
• Respect the ratchet calendar. If the tariff carries a ratchet, the month that sets it matters more than any other. Know which month that is and manage it hardest.
• Fill the valleys only with useful work. Running a process overnight raises kWh and load factor, and can lower the blended rate under time-of-use pricing, but it raises the bill if the added energy was not needed.

Computing load factor for every account, automatically

Four fields per bill are enough: kWh, metered peak kW, period start and period end. The difficulty is getting them out of hundreds of bills a month in dozens of layouts, where the peak may be labelled billed demand, maximum demand or metered kW, and the period may run anywhere from 27 to 34 days. Once those fields are extracted and validated, monthly and annual load factor is a formula that runs on every account, and the accounts can be ranked.

MartinAI reads bills of any layout across electricity, gas, water and steam, extracts consumption, metered and billed demand and period dates as structured fields, validates them against the tariff and prior periods, and computes load factor for every account and every period inside its energy management workflows. Where interval data is available through Green Button or a meter feed, the same record shows when the peak formed, which turns a low ratio into a specific hour and a specific piece of equipment. The load profile primer covers how to read that curve, and reducing peak demand charges covers the playbook once the target is known.