From the widgets on the website:
- October = 153.24 kWh
- November = 303.1 kWh
- October = 140.3 kWh
- November = 303.7 kWh
I have seen both behaviors on multiple reports and widgets including electric and water meters from two Omnimeter V4s.
Report vs. Widget Discrepancies
Report vs. Widget Discrepancies
I am seeing discrepancies between widget and report values. The calendar widget, which aggregates monthly usage, shows a monthly total that does not match the monthly total the associated report. Here are some examples:
From the widgets on the website:
I have seen both behaviors on multiple reports and widgets including electric and water meters from two Omnimeter V4s.
From the widgets on the website:
I have seen both behaviors on multiple reports and widgets including electric and water meters from two Omnimeter V4s.
- Attachments
-
- meters.png (45.27 KiB) Viewed 40117 times
Re: Report vs. Widget Discrepancies
Yes I can see what you are saying about November, it looks like the data in encompass is correct. The data in your report is off by 0.6 kWh for some reason (I would guess it has something to do with the timezone shift, meaning a month in your timezone will be just slightly different than a month in the wrong timezone). We should be able to get to the bottom of this and sort it out pretty quickly.
As for October, it is off by quite a bit more. Encompass is also correct in this case. The report is off because like you say, the report is dropping the last day of the month. This started from a desire to compare one month to the previous month. But if one month has 30 days and the next month has 31 days, you could either compare 30 days to 31 days or 30 days to 30 days. If you compare 30 to 31, then the months with more days will always look like higher consumers of electricity. If you compare 30 to 30 then you have the issue of dropping a day. There is no perfect solution to this. We are leaning toward comparing 30 days to 31 days in this case. Maybe adding something like a kWh per day value, which could be a good idea. Let us know which you think would be a better comparison to use.
If you ever really want to look at the data behind the scenes that we are using with your meter, it would be best to go to the source and look at the Summary API data for your meters. Here is a link to the Summary API data section of out Developer Portal: https://documents.ekmmetering.com/api-d ... ummary-api
And please by all means always point out issues you are seeing and improvements you want, we are making these tools for you
As for October, it is off by quite a bit more. Encompass is also correct in this case. The report is off because like you say, the report is dropping the last day of the month. This started from a desire to compare one month to the previous month. But if one month has 30 days and the next month has 31 days, you could either compare 30 days to 31 days or 30 days to 30 days. If you compare 30 to 31, then the months with more days will always look like higher consumers of electricity. If you compare 30 to 30 then you have the issue of dropping a day. There is no perfect solution to this. We are leaning toward comparing 30 days to 31 days in this case. Maybe adding something like a kWh per day value, which could be a good idea. Let us know which you think would be a better comparison to use.
If you ever really want to look at the data behind the scenes that we are using with your meter, it would be best to go to the source and look at the Summary API data for your meters. Here is a link to the Summary API data section of out Developer Portal: https://documents.ekmmetering.com/api-d ... ummary-api
And please by all means always point out issues you are seeing and improvements you want, we are making these tools for you
Re: Report vs. Widget Discrepancies
I see what you mean regarding the comparison periods. Interesting.
I think a kWh per day metric would be the most straightforward solution since that's arguably what you are really comparing when it comes to long periods of time. The total is relevant in other contexts, so having both would be helpful. With that said, leaving days off of a period is not an intuitive behavior and I didn't notice that October ended on the 30th until I had been confused for a few days. I would argue that the day boundaries should be handled in the most intuitive way even if it means sacrificing the ability to make exact comparisons in other places. I agree there is no perfect solution.
I have some broader feedback regarding reports, timezones, and other Encompass features that I will post in a different thread.
I think a kWh per day metric would be the most straightforward solution since that's arguably what you are really comparing when it comes to long periods of time. The total is relevant in other contexts, so having both would be helpful. With that said, leaving days off of a period is not an intuitive behavior and I didn't notice that October ended on the 30th until I had been confused for a few days. I would argue that the day boundaries should be handled in the most intuitive way even if it means sacrificing the ability to make exact comparisons in other places. I agree there is no perfect solution.
I have some broader feedback regarding reports, timezones, and other Encompass features that I will post in a different thread.
Re: Report vs. Widget Discrepancies
I'm seeing a different issue between the Comparison Chart widget and the report. On the widget, the graphs appear to match the data, but the report shows a very different set of charts. See below.
This chart is from a report that covers 12/1/19-12/31/19 vs 1/1/20-1/31/20.
Here is the widget that covers the same time period for the same meter.
It looks kinda similar for January, but December isn't even close. Maybe the first day of December is different because of the timezone, but I think there's a lot more going on, which should be obvious based on the other differences. The report colors also don't match the legend, but the widget appears to be correct.
My experience has been that the widgets are typically more correct than the reports. The data I get from reports is untrustworthy and I am starting to not use that feature as a result. Not being able to pick a timezone has caused some issues as well with my tenants since they have access to the physical meters and they've accused me of making rounding errors in my favor. Once I went through the hassle downloading the data directly and showing it to them, they calmed down, but it's still a problem that could have been avoided.
The reports also take a long time to send. One of my tenants needs the report on the first day of the month, but the automatically generated report doesn't show up until the second day of the month, so I have to manually generate it. There's no setting for the day the report is generated.
This chart is from a report that covers 12/1/19-12/31/19 vs 1/1/20-1/31/20.
- report.png (16.7 KiB) Viewed 38612 times
Here is the widget that covers the same time period for the same meter.
- widget.png (17.8 KiB) Viewed 38612 times
It looks kinda similar for January, but December isn't even close. Maybe the first day of December is different because of the timezone, but I think there's a lot more going on, which should be obvious based on the other differences. The report colors also don't match the legend, but the widget appears to be correct.
My experience has been that the widgets are typically more correct than the reports. The data I get from reports is untrustworthy and I am starting to not use that feature as a result. Not being able to pick a timezone has caused some issues as well with my tenants since they have access to the physical meters and they've accused me of making rounding errors in my favor. Once I went through the hassle downloading the data directly and showing it to them, they calmed down, but it's still a problem that could have been avoided.
The reports also take a long time to send. One of my tenants needs the report on the first day of the month, but the automatically generated report doesn't show up until the second day of the month, so I have to manually generate it. There's no setting for the day the report is generated.
Re: Report vs. Widget Discrepancies
Hello bandtank,
Thank you very much for bringing this to our attention. We would like to see the full pdf report that you are comparing the dash widget to. It does look like something strange is going on. Would you be willing to email us the full pdf report so that we can check it out? Please email it to support@ekmmetering.com
Sorry that you are seeing this issue, hopefully we can get it sorted in encompass.io very soon.
Yeah it can take quite a while for the encompass back end to generate all of the necessary pdf reports and email them out to all customers. You are right that this is not ideal, but there are just so many (particularly at the first of the month) that it creates quite a backlog. We are always working to improve encompass, reports and billing have been our main focuses of late. Please keep checking in to see if the improvements are working for you.
Thank you very much for bringing this to our attention. We would like to see the full pdf report that you are comparing the dash widget to. It does look like something strange is going on. Would you be willing to email us the full pdf report so that we can check it out? Please email it to support@ekmmetering.com
Sorry that you are seeing this issue, hopefully we can get it sorted in encompass.io very soon.
Yeah it can take quite a while for the encompass back end to generate all of the necessary pdf reports and email them out to all customers. You are right that this is not ideal, but there are just so many (particularly at the first of the month) that it creates quite a backlog. We are always working to improve encompass, reports and billing have been our main focuses of late. Please keep checking in to see if the improvements are working for you.
-
INTELLECTRIC
- Posts: 10
- Joined: Wed Nov 25, 2020 6:40 am
- Contact:
Re: Report vs. Widget Discrepancies
I am seeing something very strange with the widget numbers as well. I have six water meters on a system, the data downloaded for the month of April and displayed on the info widget showed 639.89m³, but the April Calendar widget showed 853.865m³. Looking back, I noticed that February (the installation month) and March were also inflated by ~25%. To make this more confusing, the monthly report showed the 639.89m³ reading, but an upstream water meter that feeds all six, only showed 601m³ for April. What could the calendar widget be looking at to come up with such an inflated number?
- April Widget.png (18.91 KiB) Viewed 11929 times
- April Data.JPG (102.33 KiB) Viewed 11929 times
Re: Report vs. Widget Discrepancies
Thank you, at first blush I think this may have something to do with daylight savings time, but Im not sure. What is the EKM-Omnimeter number and Pulse Input number you are looking at?
-
INTELLECTRIC
- Posts: 10
- Joined: Wed Nov 25, 2020 6:40 am
- Contact:
Re: Report vs. Widget Discrepancies
Sorry for the delay getting back to you on this. The meters are 000300013982, 000300013987, and 000300014018. I don't have the Omnimeter number just at the moment.
Re: Report vs. Widget Discrepancies
Hi, thank you very much for letting us know. We actually saw this same issue last week, so have been working on a fix. Will you take a look at your calendar widget again? We think we have it fixed.