In short
- A kilowatt (kW) is a rate: how fast electricity is flowing right now. A kilowatt-hour (kWh) is a quantity: how much flowed over a period.
- Your bill charges for both, separately, on different lines and in different units.
- Energy charges are cents per kWh. Demand charges are dollars per kW per month, and the kW figure is the highest interval average in the period.
- Cutting kilowatt-hours does not necessarily cut kilowatts, and some efficiency measures cut one while raising the other.
- Load factor — the ratio between them — is the single most useful number you can derive from a bill.
The clearest analogy is water, and it is worth spelling out because everything else follows from it.
Kilowatts are the size of the pipe you need. They describe how fast electricity is moving at a given moment. If every motor, chiller and oven on site runs simultaneously, the flow rate is high, and the wires, transformer and substation feeding you must all be sized to carry it.
Kilowatt-hours are the volume in the tank at the end of the month. They describe how much flowed in total.
A utility incurs two quite different costs against those two quantities. Fuel, purchased power and losses scale with volume. The wires, transformers, substations and generation capacity that stand ready to deliver the peak flow scale with the rate — and they cost the same whether you draw that rate for eleven hours or for eleven minutes. That is why the bill separates them. Why utilities bill for demand at all goes into the cost causation.
The same energy, two very different bills
The point is easiest to see with two sites buying identical energy.
Both facilities consume broadly the same number of kilowatt-hours. The left-hand one runs steadily; the right-hand one has a sharp midday spike. Under a flat energy rate their energy charges are similar. Under any demand charge at all, the right-hand site pays considerably more, and it pays it every month, for a peak that lasted fifteen minutes.
Nothing about that outcome is unfair. The utility had to build enough capacity to serve the spike.
Load factor, and why it is the number to know
Load factor is energy divided by what would have been consumed had you run at your peak continuously for the whole period. It is the ratio of the two units, expressed as a percentage, and it tells you immediately whether your load shape is costing you money.
Load factor, worked
Both inputs are already printed on your bill.
- Energy in the billing period96,000 kWh
- Billed demand400 kW
- Days in the period31
- (Hours: 31 × 24)744 h
- (Energy at constant peak: 400 × 744)297,600 kWh
Load factor: 96,000 ÷ 297,60032.3%
A site at 32% is paying for capacity it uses under a third of the time. Sites with high load factors get little from peak management; sites like this one usually get a great deal. Figures illustrative.
As a rough orientation: a continuous process plant running three shifts tends to sit high, a single-shift manufacturer in the middle, and a facility with short intense production runs low. Where exactly your own site lands is a question your own two numbers answer in about ten seconds.
The number is diagnostic rather than prescriptive. A low load factor is not a fault — a business that only needs its equipment two days a week is running correctly by running it two days a week. What the figure tells you is that a tariff which prices capacity heavily is a poor match for how you operate, and that either the load shape or the tariff is worth changing. Which of the two is the cheaper thing to change is the question the rest of this site addresses.
It is also worth tracking month to month rather than calculating once. A load factor that collapses in one season points straight at cooling or heating; one that collapses in a single month with no seasonal explanation usually points at a one-off event that set a peak nobody noticed at the time.
Where the confusion causes real losses
Efficiency projects sold on the wrong unit. A lighting retrofit that cuts kilowatt-hours substantially may cut the demand charge by very little, if the lights were not on during the peak interval or if lighting is a small fraction of the load at that moment. The savings are real, but they are energy savings, and a business case that promises demand savings will miss. See six measures that do not cut your demand charge.
Shutdowns that create peaks. Turning everything off overnight or over a break saves energy. Turning everything back on simultaneously creates a fifteen-minute average that exceeds anything the site produces in normal operation, and that average sets the demand charge for the month. Staggered startup is the cheapest fix available anywhere on this site.
Blended rates used as marginal rates. Dividing the total bill by kilowatt-hours produces a blended cost per kWh. It is a fine budgeting summary and a terrible input to a savings calculation, because avoiding a kilowatt-hour does not avoid the demand charge that was blended into it.
Reading the units on your own statement
Anything priced in ¢/kWh or $/kWh is a volume charge and responds to how much you use. Anything priced in $/kW or $/kVA is a rate charge and responds to when you use it. Riders can be either, which is why riders and surcharges is worth reading before you model anything.
Once you can sort every line on the bill into those two buckets, you know which of your operational habits each one is charging you for — and the rest of this site is about changing the second kind.
