There's Crazy - then there's Octopus

BTW the bill is headed:

Your energy account

That's reminiscent of the BT bills that I used to get before I switched to Zen. BT's were of the format: Think of a number, divide it into several parts, charge back for some of them, and finally forget any result and think of a new number to put on the bill. Zen's bills stayed the same year after year....

What granularity does your smart gas meter send of its readings? 0.1 m3 is quite a lot of gas, so presumably it must report in a smaller granularity to the mothership?

It sounds like you've used so little gas that rounding is an issue.

Theo

We're with Octopus and a smartphone app that talks to the In Home Device (the little console) reports daily usage in kWhr to one decimal place. Octopus's phone app (24 hours later) reports it to 2 dp, and their web site reports it to 3 dp.

I've always wondered how much accuracy is justified and how much is multiplying a fairly inexact volume in m^3 by a m^3-to-kWhr conversion factor, and reporting the result to spurious accuracy.

I've noticed a funny with Octopus and gas. When their app and web site eventually deign to give a reading for "yesterday", often as much as 15 hours after midnight when a new day begins, the figure is slightly different to the figure for that same day, when reported a few days later.

For example, our website result for Mon 15 June was reported yesterday as 40.2 kWhr (*) but today it's reported as 41.1 kWhr. That later figure remains the definitive reading if I check again in a few days/weeks.

It's as if they give a preliminary reading when they first quote a reading, and then they slightly adjust it, sometimes upwards and sometimes downwards, the following day to a final reading.

(*) That's also the figure given shortly after midnight on the phone app which talks to the IHD.

Who knows? 0.1m3 would power my boiler for two minutes.

You've edited out the answer, i.e

There is no way that 1005.0 minus 1005.0 can produce 0.4 - let alone

-0.4 ! The same applies to any X minus X, whatever the granularity.

See above

I've noticed that change of reading, too - and have reported it to Octopus. No interest shown.

At the moment, I cans see 48 16th June figures and 3 17th June figures.

Try downloading a month or two's data. Unbelievably slow.

There is no way that 1005.0 minus 1005.0 can produce 0.4 - let alone

-0.4 ! The same applies to any X minus X, whatever the decimal granularity.

And I didn't pump just short of 1,000 kWh of gas back into Octopus over the first few days of June 2025 !

AIUI there's about 11 kWh per m3 (seasonally adjusted).

So 0.4 kWh is 0.4/11 = 0.036 m3.

The front of your meter only displays in units of 0.1 m3, but the smart meter backend reports in units of 0.001 m3 (if NY is correct).

So your customer reading on 5 June was sent as 1005.000.

But maybe the smart meter reading on 1 June was 1005.036. That got rounded down to 1005.0 for displaying on the front of the meter (as is traditional for meter reading).

However the backend knows the actual consumption to the nearest 0.001 and if you'd not sent in a manual reading it would have still recorded 1005.036 if you had used no gas between those times, and would have issued a bill with exactly zero consumption. However your manual reading took precedence, and hence there was a small negative adjustment.

Theo

I've just checked the gas meter and, except for the bracketed words above, you are absolutely right! As of now - 1005.035m3 on the gas meter display, 1005.0 on the remote display and the bill. Many Thanks.

At the end of the conversation about electricity the agent asked about the gas figures and I replied that the figures should be the same, no gas having been used since mid-April.

There's many a slip twixt cup and lip!

Now, how about the other phantom "you used -999.90 kWh in the same period the previous year."? Is that -999.90 a truncated 999.9 recurring?

Where are the sanity checks - particularly regarding negative figures?

That reminds me of a conversation I had with a workmate when I started work 60 years ago. He had just moved into a new address and registered himself as the new user of electricity. By return of post he got a bill for £0 0s and 0d (it was before decimalisation).

As it was a bill for nothing, he ignored it. He subsequently got a red bill (ie overdue) with a warning that if he didn't pay the £0 0s 0d within 7 days the case would be passed to a debt collection agency. So he put the red bill in the supplied return envelope with a cheque to pay "Zero pounds, zero shillings and zero pence" for his outstanding bill. That satisfied the supplier's computer, but the supplier put that cheque with several others in a pack for their bank - and it crashed the bank computer because that hadn't been programmed to process a zero value.

The bank having a brief computer problem got a 1 inch mention in the local paper but it didn't say what had gone wrong.

I have noticed from my bills that the conversion from M3 to KWh isn't always exactly the same. It varies slightly depending on the time of year, presumably based on ambient temperature because the calorific value for a given volume will depend on the temperature expansion affecting the volume at the time the gas was used. That variability in the calculation could explain your 0.4 if the two readings were far enough apart to have different expansion values.

Also the exact composition of the gas, which will depend on its source.