Updated on 09/12/25 by Ben_OVO
The total electricity consumption I am billed for is correct but the allocation of cost to peak versus off peak use is wrong and seems to vary month to month.
It isn’t surprising that the half-hourly usage figures don’t tally with the reported peak and offpeak meter register advances, although the difference won’t normally be significant.
A SMETS2 meter configured for multi-rate operation with different tariffs at different times will normally see to two actions at the same time; at the start of an offpeak period, the meter will both start recording usage on the offpeak register and also switch on power to heating equipment like storage heaters. The reverse happens at the end of an offpeak period.
The most common arrangement of this sort is the venerable Economy 7 (E7) plan, where electricity is charged at a cheaper rate between midnight and 07:00. Smart meters are precision instruments, so the switch would take place at the same millisecond in every E7 system. Suddenly increasing the demand by 10-15kW for many households in a neighbourhood could cause problems for the network joining them all to a substation. For this reason, every SMETS2 meter has a built-in randomized offset that delays the switching times (both for the register and for the heating load) by a few minutes for each offpeak period. The offset is unique and immutable for every meter, and it will usually be less than 10 minutes.
Well and good. However, records of half-hourly consumption aren’t subject to the same offset. They will be recorded with millisecond precision for 48 periods each day. This means that - taking the aforementioned E7 plan as an example - the first bucket of each day contains ~10 minutes’-worth of usage at peak rate and ~20 at offpeak. The same applies at the end of the period, but it’s often less significant, because it’s likely that the heavy load will have reduced before the end of the 7-hour slot.
You are billed for what the meter registers record, not for how much the half-hourly data buckets contain, so there is no question of overcharge in a situation like this. It can happen, though, if the switching times of heavy loads like storage- and immersion heaters aren’t controlled by the meter but by their own timers. It’s important to ascertain the precise switching times of your system to avoid this sort of extra expense.
That is the usual explanation for the difference between meter advance and usage data. There is also the question of the change to and from British Summertime. Meters won’t normally alter their behaviour at the change, but some less modern time switches, like many of those controlling overnight immersion heaters, will need adjusting by an hour twice a year. Alternatively, they can be set to switch on at, say, 01:30 instead of midnight, and off again at 07:00 - it’s unusual for an immersion heater to take more than 3 or 4 hours to heat a tankful up from cold.
If you’d like to share some of the figures you’ve been examining, we could take a look to see if there’s something else afoot.
