Pulling your interval data from ComEd is not hard, but it is an afternoon, and there is a question worth answering first: what do you actually want to know, and does the bill in front of you already answer it?
Three of the four things people want interval data for are on the bill. One is not. It is worth knowing which is which before you spend the afternoon.
What your bill already settles
How big your peak was. The billed demand in kilowatts on your bill is your peak. It is the highest interval of the month, already measured, already on the page. You do not need the underlying record to know its size.
What that peak cost you. Multiply the billed kilowatts by the dollars-per-kilowatt rate on the same bill. That is the demand charge in dollars, and comparing it to the bill total tells you whether it is a large enough share of your cost to be worth any further work at all.
Whether the peak is a spike or your normal operation. This is the one people assume needs a file, and it does not. Take the kilowatt hours for the period, divide by the billed kilowatts, then divide again by the number of hours in the billing period. That ratio is your load factor: how close your average draw runs to your maximum draw.
A high ratio means a flat, heavily utilised building. Your peak is roughly what you do all day, there is no emptier interval to move load into, and peak shaving has very little to work with. A low ratio means your average sits well below your maximum, so the maximum is a spike, and a spike is a thing that can potentially be moved, staggered, or absorbed.
That single division reframes most equipment proposals, and it needs two numbers that are already printed on the bill.
What only the interval record can settle
When the peak happened, and whether it recurs.
A billing month contains roughly fourteen hundred readings. The bill preserves the largest one and discards the rest. So the bill can tell you that a spike exists and what it cost, and it can never tell you which Tuesday it was, what hour, what was running, or whether the same thing happened in eight of the last twelve months or exactly once.
That distinction is the whole decision. A peak that recurs on a predictable schedule is an operations problem with a cheap fix. A peak that happened once because two things coincided by accident may be worth nothing at all to chase. The bill cannot separate those two cases and the interval record can, and that is the only reason to go get it.
So the honest test
Work down in this order and stop as soon as you have an answer.
Size the demand charge as a share of the bill. If it is small, you are done, and no export is going to change that.
If it is large, run the load factor. If it comes back high, your peak is your operation rather than an event, and an interval file will show you a flat line confirming what the division already told you.
If the demand charge is large and the load factor is low, you have a spike that is costing real money and the bill has taken you as far as it can. That is the case, and in our experience it is the minority case, where the export is worth pulling.
A note on what this is not
If you arrived here looking for interval scale in statistics, or an interval error in a chat application, this is the electricity kind: the thirty-minute readings a utility meter records and a monthly bill is summarised from.
Do I need interval data to understand my demand charge?
Not to size it. Your bill already carries the billed demand in kilowatts and the rate applied to it, so the dollar cost is on the bill. Interval data is needed to find out when the peak occurred and whether it recurs, which the bill cannot say.
Can I calculate my load factor from my electricity bill?
Yes. Divide the kilowatt hours for the period by the billed kilowatts, then divide by the number of hours in the billing period. It tells you how close your average draw runs to your peak, which is the first indication of whether peak shaving has anything to work with.
What can interval data tell me that a bill cannot?
The timing and the recurrence of your peak. A bill preserves the single maximum reading and discards the roughly fourteen hundred readings behind it, so it cannot show which day and hour set the charge, what pattern it followed, or whether it repeats month to month.
Is it worth getting ComEd interval data for a small commercial account?
Only if the demand charge is a meaningful share of your bill and your load factor suggests the peak is a spike rather than your normal operating level. If either test fails, the export will confirm what the bill already told you.