One of the early changes I made after stopping working - and more specifically after ending my routine of spending most of my hours from 9-5 on a weekday in an office - was to change my electricity provider.
In some ways this may seem like an odd thing to do, but the goal here was to switch to an electricity tariff that charges based upon 'Time of Use'. And to explain why I did this, an explanation of how electricity pricing works is probably the best place to start... (and apologies, because there are a few technicalities to overcome on the way)
Electricity Pricing Explained
Traditionally you go to an electricity supplier and they offer you a fixed price for any electricity or gas that you might consume over a fixed period of time.
This was quite an easy problem for electricity suppliers and the power market to solve historically, because if the cost of gas could be fixed in advance, the most variable part of the power price would have also been fixed by that same action. Hurrah!
But - what makes things more complicated these days has been the rise of renewables, because now renewables - which have no running costs beyond basic maintenance - run at the expense of traditional power sources like gas power stations (which do have to pay for the cost of input gas etc).
Whether a power station will run or not, depends on whether there is a positive or negative net cash inflow from potential activity, considering all available income streams.
For a renewable project without subsidies, this means running for any positive power price - for any projects with subsidies, this means they will run at negative prices, so long as the cash flow is positive once subsidies are also factored in.
A renewable project still has a cost that is fairly high overall, but that is predominantly a capital cost that by the point of production has already been paid and is no longer consequential.
The result is that while gas power stations would historically run via a fairly stable pattern, based only upon the overall demand for electricity nationally - the activity of renewables (and thus the weather) now significantly alters their pattern of operation.
To show the impact of this, we should start by looking at the base case - i.e. a winter day, with high electricity demand and a low level of power generation from renewable sources :
 |
| Power generation by fuel in late January 2025 - Source Elexon |
For clarity, note that CCGT stands for Combined Cycle Gas Turbine and is a term used to describe a highly efficient gas power station (combined because as well as running a normal gas turbine, this is combined with an additional steam turbine which generates additional power from the exhaust gases).
On the 31st January 2025 (the middle day), there was a lot of power generated from gas power stations (pink area) - and this is important for this particular article, because it meant that the price of power was relatively high.
The reason for this is that gas power stations effectively sit in a queue based upon how cheaply they can produce power.
The power station at the front of this queue is the most efficient and likely newer power station - and if the market were to require three power stations to come online, he (or she) would be expected to come on, as he (or she) can produce power most cheaply.
This means that the most efficient plant is expected to run almost continuously through the year - but will need to price competitively, as otherwise a slightly less efficient power station might jump it in the queue by taking a smaller profit margin.
As an additional aside - some gas-fired power stations also provide waste steam to local industry and so may generate at a loss to ensure they deliver that steam. As a result these plants may run ahead of more efficient plants that don't provide steam to local industry. This technicality can just be ignored for narrative purposes, but it also highlights how these things get more complex as you look into them in more detail.
But, let us now imagine a very different day - like that one above on the 31st January - where the market not only has twenty power stations, but also needs them all to be operational on that day.
In this scenario, the power station at the end of the queue - the equivalent of that kid in PE class that no one wants to pick for their team - has it's day in the spotlight. This will be an older and fairly inefficient power station that must additionally recover all its maintenance costs from a precious few hours of operation each year.
When this equivalent of "that kid in PE class" comes online, the one thing that saves it is that it has no one behind it in the queue, and so it can charge a higher profit margin, making up for it's rare levels of activity.
It can be hard to imagine how this might work over a year, but fortunately this is quite straight-forward to do, for all we need to do is look at the level of generation by fuel type once again, but this time over a longer period of time (from September 2024 to September 2025) :
 |
| Repeating the same chart as above for a full year - Source Elexon |
What we can see is that in January there was a large spike in levels of gas generation (pink area) as overall generation levels were high, but wind generation levels (dark blue) were low. But later in the year, the pink area become much smaller compared to that peak level.
The most curious among you might also be wondering why there is no solar generation, despite the country having solar on roofs and solar farms in fields along many motorways and railway lines.
The reality is that it is there, but it is connected into the local distribution network - much like your home or place of work might be - and so from the perspective of the system operator it doesn't exist. Obviously it does exist, but it is reducing the total rather than showing up as a plot - we will see this clearer later on in the explanation.
Solar is not the only generation source where this applies as other renewable projects connect in a similar way, but these technicalities are not that important - what really matters is the changes in the level of activity at gas power stations that can still be seen clearly from the chart.
The result is that in a given year the power market has a small number of very expensive days, a reasonable number of generally more expensive days (high demand, low renewables), numerous days where power is not that expensive and an increasing number of days where power is very cheap (where renewable generation is very high against demand for power).
As a result, although you might always pay 24p/kWh for your power, your electricity supplier might occasionally be paying 99p/kWh (or more!), could have a reasonable number of 30-35p/kWh days, numerous days of 10-20p/kWh and an increasing number of days where power is close to zero or negative.
All this will vary throughout the day at a half-hourly level, and to give an idea of how it looks in practice, we can look at the following chart - which shows the price of power ahead of delivery in £/MWh for the 18th and then the 19th of September :
 |
| Intraday price of power - Source EPEX |
Here prices are negative overnight on both days as the levels of renewable generation push prices into negative territory, but then rise during the day as the need for higher levels of power generation from gas-fired power stations returns.
 |
| British electricity generation by fuel source - Source Elexon |
Coming back to that prior aside about solar - you can see on the second day that the total area of the chart dips in the middle of the day. Some of this is reduced demand for power in the middle of the day, but the majority of this dip is because power is being produced by solar on local networks - in effect reducing the need for additional generation from a national perspective. This is further squeezing levels of power generated at gas-fired power stations.
I'll admit that the explanation was longer than I'd hoped, but we are now making serious progress and can start looking at what this means in terms of our electricity bills.
Under the traditional approach used by electricity suppliers, all this activity above (at a half-hourly level) must be converted into some sort of estimate for how much we must be charged via any p/kWh fixed fee.
At the end of the day, if it costs a supplier more to supply our electricity than they receive from us, they probably won't be our supplier for too much longer - instead they will join the ranks of former but now bankrupted electricity suppliers.
So what they do is make an assumption about how we use electricity, and in their calculations they will assume that our use of electricity looks something like this :
 |
| An example profile for typical electricity consumption |
What they are doing here is normalising all the activity across the country and assuming that we are identical drones who all do the same thing.
Time of Use Tariffs
And so we finally get to the point of this article (finally, I hear you say).
A Time of Use tariff swaps out the above estimate for 'normal consumption' and swaps in your actual consumption through the day - with the cost of power varying for each half-hour in the day.
If your consumption is better than the normal profile then you should save money swapping to a Time of Use tariff, even without any changes in activity - whilst any further shifts in power consumption towards cheaper parts of the day will result in additional savings.
In terms of how this works, we can see the pricing profile I faced for a fairly typical day in September 2025 :
 |
| Octopus Agile Tariff for a typical September day (19th) - Source: Octopus App |
I picked this day to use as an example as it follows the normal pattern of a day (low price overnight, up in the morning, low in the middle of the day, and highest from 4-7pm, before dropping a bit in the evening. These prices only become known at 4pm the prior day, but since you know weather in advance of that, you should generally already have an idea of how a day should look well beforehand.
 |
| Electricity Consumption on 16th September 2025 |
 |
| Total cost of power (including standing charge) for the day |
I would estimate that I used about 2kWh for the first run in the oven and then about 1kWh for the second. That first cost me about 1p (0.5p/kWh), whilst the second cost me about 32p (32p/kWh).
Scores on the Doors - So What Are the Savings?
So has it been worth it?
I swapped to the Time of Use Tariffs in June 2024 and used to be paying 24p/kWh, which I have used as my benchmark for savings - and generally I have saved money during every month since except January 2025.
Focusing initially on that January month (where I paid 28p/kWh for my power), this was a month were it was not very windy and in addition where gas prices were higher than usual (in part because it hadn't been that windy all winter). This pushed up the price of power on those days and I ended up paying a premium as a result.
But otherwise I saved money every month - with savings dipping in winter and being particularly high recently :
The blue line shows how much power would cost based upon the 24p/kWh I would have paid extending my contract, whilst the pink line shows what I was paying via Time of Use. In green you an see the difference, and these three plots are all showing the total cost in pounds. For reference I have put the usage in kWh in grey on a second axis.
My electricity usage has risen over the period, and this is because I started running a dehumidifier to deal with condensation on windows that would lead to small patches of mould around the window frames. That is adding about 50kWh of additional electricity usage a month, but I feel that is worthwhile (I'll do a full article on the topic one day).
In terms of total savings, the overall figure is £104 for the full period, dropping to £75 for the last 12 months - but the savings have been higher more recently, and for the past four months my cost of electricity has averaged 16p/kWh.
The increased savings in recent months, are in part due to the time of year - with it being much easier to save money in the summer, when gas generation is typically low. But it is also because of specific weather conditions during each month, and also because my shifts in activity have now become habitual rather than intentional.
In very recent times I have been modifying my use of the dehumidifier to avoid the evening peaks, and have now - at the very end of the period - settled on a new pattern of running it for 7 hours in the morning, and then having no operation until the following day. For much of the period I have run it continuously, albeit with it setup to stop running once humidity is in an acceptable range.
Setting it off to run for 7 hours once a day in the morning now feels like the best balance between effort, impact and overall cost. This might bring the overall cost of power down a bit from the levels seen so far, although in some months it will run often if it is damp, whilst in drier months it will see less operation.
Long-term my belief is that I should save about £100 a year now that my habits have fully adapted - and in terms of whether it is worth it - I would say yes, but with a caveat.
A saving of £100 is not massive, but the impact upon me of these actions is very small, and so I'm happy to get a little financial boost from not doing very much at all. The return is fairly modest, but the level of effort to achieve it is lower still. Indeed I'm confident if I did nothing differently I'd still be making savings, just being home during the middle of the day.
If I had a heat pump or electric car, I think that is where the true savings would lie, as my gas bill amounts to over two-thirds of my energy bill - with electricity being the smaller share of that energy bill.
But my conclusion to the whole experiment is that if you are at home through the day during the week, you might as well use a Time of Use tariff. At the same time you don't need to do very much, just work out the big consumers of power, and if there is no real consequence to doing so, use them at slightly different times.
With an electric car or heat pump, I suspect you could make real savings on your total energy bills - but for everyone else, while it won't completely change your financial situation, there's a bit of free money to be had from it all the same.
Additionally you can also take comfort in the fact that shifting to a lower priced period is going to help make the electricity system very slightly greener, which is also a nice little win.
Solar Panels
Ironically though, what has come out of it as a big takeaway is that I now definitely don't see the point in buying solar panels (unless I end up in a house that is more off grid than usual).
You see, when the sun shines I already get discounted electricity thanks to the solar panels that already exist nationally.
If I did have solar panels I would have to work even harder to use the power in the middle of the day to justify their installation, or I would need to further sink in money by adding a battery - which in turn might go and burn my house down.
If I were off grid a combination of solar, a battery and a heat pump would make a lot more sense than relying on oil or gas deliveries, but living where I do this is not a consideration.
Instead I realistically can't see much additional saving from having those solar panels - and in many ways that is a good conclusion, because I don't want the faff of the installation and maintenance anyway.
Comments
Post a Comment