Using Interval Data to Manage Commercial Demand Charges

On a residential bill, the number that matters is energy: how many kilowatt hours you used. On most commercial and industrial bills there is a second number that often costs more and is far less understood. It is demand, measured in kilowatts, and it is billed on a single peak reading rather than a total. A facility can cut its energy use and still watch its bill barely move, because the demand charge did not change. Understanding where that peak comes from, and doing anything about it, requires interval data. This guide explains how the demand charge is calculated and how to use your interval feed to manage it.

Energy and demand are two different things

Energy (kilowatt hours) is how much electricity you consumed over the whole period. Demand (kilowatts) is how fast you were drawing it at your busiest moment. A shop that runs steadily and a shop that runs the same total load in short, intense bursts can use identical energy and be billed very different demand charges. The bill blends the two, which is why the total can move in ways that a look at consumption alone does not explain.

How the demand charge is calculated

The standard method is to take the highest short interval average of power during the billing period and multiply it by a dollar per kilowatt rate. Wisconsin utility We Energies describes billing demand as the highest 15 minute average kilowatt demand recorded during the month, which is the common approach across North American commercial tariffs. Some utilities use a 30 minute interval and a few use an instantaneous peak, but the 15 minute integrated reading is the usual basis.

The averaging window matters. Because the meter averages power over the interval, a brief spike, such as the inrush when a large motor starts, does not set your peak by itself. What sets the peak is sustained load across the full interval. That is good news, because it means the peak is usually driven by identifiable, controllable behaviour rather than by unavoidable momentary surges.

Why the numbers are large

Demand charges are not a rounding error. A 2017 National Renewable Energy Laboratory and Clean Energy Group survey of more than 10,000 utility tariffs across 48 states found that roughly 5 million of the 18 million commercial electricity customers in the United States could subscribe to a rate with demand charges above 15 dollars per kilowatt, which the researchers used as an industry benchmark for where managing peak demand becomes economically worthwhile. For a site that peaks at a few hundred kilowatts, a rate at or above that level turns the demand charge into one of the larger line items on the bill.

Many commercial tariffs add a ratchet, which makes the peak matter for longer than one month. Under a ratchet, your billed demand in a given month cannot fall below a set fraction of your highest peak over a trailing window of previous months. In practical terms, one bad afternoon can prop up your demand charge for the rest of the year even in months when your actual peak was much lower. Where a ratchet applies, avoiding an unnecessary peak in the first place is worth real money.

Why interval data is the only way to see the peak

Here is the problem the bill creates: it reports the peak as a single number, one kilowatt figure for the month, with no context. It does not tell you which day, which quarter hour, or which equipment produced it. Interval data does. With 15 minute reads you can find the exact interval that set the peak, look at what was running, and see whether it was a recurring pattern or a one time event. Without interval data you are managing a number you cannot see the cause of.

If your demand charges are a mystery on the bill, the fastest fix is to get the interval data flowing so you can actually see them. Book a walkthrough at ezgb.ca, tell us which accounts you want to watch, and we will set up a clean interval feed so the peaks stop being invisible.

Turning interval data into action

Once you can see the peaks, a repeatable process follows:

Find the peak intervals for each account and rank them, so you know which sites carry the largest demand charges.
Correlate the top intervals with operations: shift changes, HVAC startup, batch processes, simultaneous equipment starts.
Look for coincident loads, cases where several pieces of equipment happen to run in the same interval and stack into one high reading.
Test whether staggering or shifting some of that load flattens the peak without disrupting operations.
Track the peak month over month so you can confirm that a change held, and catch a new peak before a ratchet locks it in.

None of this promises a specific saving; the value depends entirely on your rate, your ratchet terms, and how much of your peak is genuinely movable. What interval data changes is that these decisions become evidence based. You are no longer guessing at what drives the bill; you are reading it directly from the meter.

How EZGB provides the interval feed

EZGB (Easy Green Button Connector) collects utility interval and billing data from authorized utility connections and from Green Button, whose Download My Data and Connect My Data methods are built on the NAESB ESPI standard. It standardizes the reads into consistent units and clean fields, keeps up to two years of history where the utility provides it, and delivers the result by REST API, CSV, or SFTP. That gives your analysts or your energy management software the interval detail they need to locate peaks, without anyone exporting files by hand each month.

To be clear about the boundary: EZGB supplies the interval and billing data. What you do with it, whether that is analysis, load shifting, or a storage business case, is your decision or your advisor's. The point of getting the data clean and current is that the demand charge stops being a number you accept and becomes one you can investigate.