Just got my first bill with complete coverage by Sense during that time and I was surprised to see there is a large difference reported in kWh usage.
Sense shows 573 kWh while my bill shows 761 kWh over the same period.
I’ll be reviewing my install and taking some meter readings myself to see what’s up, but just wondering if others have seen such differences and if so were you able to correct them?
From my experience… Obtaining an “accurate” comparative analysis of the Sense kWh vs Energy Bill has proven that rarely are they comparable. Historically, Sense is notably higher than the actual Energy Bill. The Sense kWh serves only as a baseline value, mitigating “surprises” in my upcoming Energy Bill.
I no longer perform this ineffective comparative method. I now simply tracks Sense’s average monthly kWh totals, effectively identifying my Sense-baseline usage, while doing the same for the Energy Bill.
Really? Very surprised to hear that actually. Hmm. Thanks very much for the data points.
I think @invoice is not getting a proper comparison of his Sense usage to his utility. Your usage is off by 25%, something is not right with that. You should be closer to +/- 2% You are either not measuring all your load or you are not getting the proper comparison of statements. You have to know exactly when the utility reads the meter and when Sense reads the data for the month in question.
I tell Sense that my billing is the 1st of the month, so they use that to figure my monthly usage. My utility gives the next meter reading date on the statement, sometimes that is on the 2nd or maybe the 30th, depends on holidays etc. I take the utility usage , divide that by the number of billing days for an average daily usage. If the reading is on the 2nd, then subtract 2 times that number from the total utility usage and you will be pretty close to the Sense number. Hope this helps.
Sense monthly energy consumption statistics should be within 1% - 2% of your monthly energy bill. The greater discrepancy you’re seeing could be from the installation process or the dates of measurement of your utility bill not lining up with Sense (Data Export is a great tool to do this comparison). If either of these are not the root cause here, please reach out to support@sense.com so we can take a deeper look at this for you.
I’d concur with @JustinAtSense and @jkish here. My billing dates vary, as does the length of the billing period…it’s frustrating. I export my Sense data at the daily level every month and make sure to line up the dates exactly with my exported utility data. I’ve always been within less than 1%. With a difference as large as yours, it sounds like some other external factor is at play here.
I have been comparing against my utility bill for the past couple years at an hourly level. Even after having Sense measuring both usage and solar and doing the net calculation, I am seldom off by more than a % or two. When I am off, it is because of dropped data.
I would suggest that you try analyzing hourly or daily if your utility offers you the data. You might find that:
Thanks, 1-2% lines up more with my expectations.
My bill is also a variable number of days but it was possible to get it to line up in the app usage. Before reading all these responses, I opened my panel and the CT clamps did look pretty canted, like almost 45°. I readjusted them and tried to make sure not to jostle them when putting the cover back on. It’s pretty cramped in there.
I tried turning on/off some previously detected devices and nada. So I reset my data.
I actually forgot all about the data export. Probably should’ve checked that first 
I also had not considered a sub-panel. Good suggestions. Doh!
Anyways, I’ll take a meter reading at midnight and at midnight again tomorrow. Then I can compare against the usage in the app and the data export. To be continued…
I too have variable number of days for my bill, not had a full month yet but it will not matter what it is as my house panel is a sub panel. the main panel feeds my well, a shed and my house. It will also feed my shop in about 6 months i hope. I am working on getting the wifi to cover the area where the main panel is so I can move sense out there to see the “whole” picture.
Wonder if that will cause the time so far to be wasted when i move the sense unit to the main panel.
Okay, so over 24 hours there is a difference of 2%. My sump pump, furnace, and a laser printer have already been detected again. I think what actually happened though is that there can be a delay between what is recorded and what is shown in the app and that might have caused my original issue. Just a guess based on my observations this morning.
I checked the usage for the day a few minutes after midnight after reading the meter and it said my monitor was offline. I knew this wasn’t the case as I had another app instance open on a different device at the time and it only showed this message when checking the usage for that day. I checked it again later and it showed maybe half the day. I checked again this morning and it is now showing the total kWh for that day. So, maybe if I had just waited a day, I would have seen the missing data??? I don’t know, 200ish kWh is a lot of unaccounted for energy though. Could have a been multiple issues going on.
@brian4 Heh, don’t do what I did and just starting moving and deleting stuff. I probably should have contacted support first and got their advice. If nothing else, the time will not be wasted as you will have some info on how many of your devices can be detected and how accurately in 6 months time.
@ramon,
For me it, the Sense summary data in the Trends or Devices sections (of both phone/tablet app and web app), or via export, aggregates and posts every hour, so you see the data up until the last complete hour. So here, it is about 7:30AM and I can see data up until the last complete hour - 6AM. 7AM data will be posted a few minutes after 7AM.
One more note - just did my end of year, hourly comparison of my Sense Net (Total Usage - Solar Production) vs. my utility’s billing net meter billing data. As expected, except for maybe 60 or so hours out of the 6220 hours this Sense has been in operation, I have seen hourly results that coincide within +/1% of Sense. The colors show the biggest 25 errors (red), the next 25 biggest errors (yellow), and the next set of 25 (errors):
Removing them, I can generate a linear model with coefficients below, that means SenseNet = 1.000970 x PG&E data - 2.216Wh
Coefficients:
(Intercept) Sense3$PGENet[keepers3]
-0.002216 1.000970
If I look at the distribution of the remaining errors they are very balanced around this model, most within 100Wh
And if I look for where the 75 biggest errors occurred in time, that I dropped from my final analysis, most took place during periods where I was having known issues:
None have happened since early October.
Update: I just reviewed my latest bill since my reinstall with complete coverage and the difference reported by Sense and my bill was about 3kWh. That’s a difference of less than 1% and less than $1. 
I have a “smart” energy meter from the electric company that updates usage every 15 minutes and over the last two weeks since installing sense the two have matched daily usage with a bias of well under 1%. Could it be your utility bill wasn’t being generated from an actual meter reading? Utility companies don’t always read meters every month… they just estimate then make up for it with an actual reading every 2nd or 3rd month to save employee hours.
Good point.
For future reference (since @ramon’s issue was a different one):
Most Utility bills will call out whether a reading was “read”, be it from an online smart meter or by a human with a wireless reader or a human’s eyeballs, versus “estimated”. If you see “estimated” it could be way off if your use has changed significantly from the last year for the same period (usually how estimates are made). “Estimated” is usually an indication of a human not being able to get eyeballs on a meter to do a reading.
I wouldn’t put it past certain Utilities to exploit that anachronistic necessity to impose an anachronistic reading to something supposedly Smart, i.e. if their “smart” systems glitch they could also give you an “estimate”.
If you get that type of reading it’s going to come down to exporting Sense data and doing a matchup if you want to verify the bill in the short term. In the long term, things will get recalibrated after an actual reading takes place.
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.
