This has been a subject that I’ve wanted to investigate for some time because I want to know whether getting solar panels or a home battery was worthwhile in terms of a good investment. However, the subject is very difficult to analyse and it was going to take some effort to untangle this particular Gordian Knot and I wasn’t sure if the effort was going to be worth it. I figured that solar panels would most likely be a benefit. I saw a home battery as being ‘parasitic’ on the solar panel scheme in that it doesn’t generate any power in itself and probably wouldn’t return the value invested in it. It’s also difficult to analyse but I’ve finally convinced myself that I want to know the answer and that is why we are here.
Solar Panels – These provide power during the year. They have an average installation cost and a yearly output related to the installed peak power. Anything produced over and above what is required in the house at the time can be exported as a Solar Export Guarantee (SEG) payment at the SEG rate or stored in a Home Battery.
Home Battery – Allows you to store excess solar panel power generation for later use. If the battery is full the excess can still be exported at the SEG rate. The analysis in this blog only covers this use. It’s also possible to store solar power for export at a later time (Solar + time shifting) or just have a battery (no solar) for time shifting. Both of those are not covered here and may make a difference to the economics of having one. It’s also possible to run the house independent of the grid during a power cut but there will be additional costs to allow that to happen.
RESTRICTIONS on the application of this blog. The results and conclusions in this blog apply to my usage data ONLY but I believe that the conclusions are still broadly valid for other data usage. As a summary my data usage looks like this:
Yearly domestic power: 3992 kWh (includes an A2AHP and an Immersion Heater)
Distribution of use:
I also have an EV as well but I’ve very carefully separated out the 3088kWh that it uses as it’s not part of this analysis. This blog also doesn’t cover charging an EV from solar panels or a home battery. That may also change the economics of a home battery.
Solar Panels
The yearly output of the system varies according to location but for Southern England each kWp of installed panels produces on average 850 kWh per year. The day to day output of the panels varies according to the season, the time of day and the weather. For each 1 kWp of panels you can expect to get the following daily energy outputs.
Sizing the solar array for a house
This variation is part of the issue which makes it difficult to size a system according to your energy use. During the winter, when your electricity usage is generally higher, the panels produce less. However, you are likely to be able to use all the power that they produce. In the summer the output is much higher but your demand less and there is going to be an excess of energy produced which can be exported for a SEG payment. There is a hint on how to size an solar array in the section called The GOOD.
This chart shows the importance of having SEG payments if you haven’t got a battery. The percentage used can be very low in the summer. The green ellipse is where I simulated being away on holiday for a week and reduced the power usage to a background level. As you can see the export has a corresponding increase during that time.
The Solar Export Guarantee (SEG)
In order to encourage the uptake of domestic solar panels the government introduced a requirement through Ofgem that larger energy suppliers have to offer a Solar Export Guarantee (SEG). The only requirement is that the SEG has to be greater than zero so rates from 1p to as much as 40p per kWh are offered from suppliers. These rates can change at any time. Pay Back Time for Array Size, Tariff and SEG. Without the SEG it would take a lot longer (2025 installation rates) to recover the cost of the installation. The graph shows how I would be able to recover cost with an average 3.5kWp solar system with no battery and a Tariff of 25p (£0.25). As you can see the Pay Back Time (PBT) is strongly dependent on the SEG rate that you can get. The PBT is also very dependent on when you use your energy which means that the graph is very dependent on my data. If I were to use more energy during the hours of sunshine then the graph would change.
STOP PRESS: I was looking for typical electricity usage patterns on the internet and came across this graph (from ResearchGate.net) which gave me data for the winter and summer and then averaged that for spring and autumn. I then ran all my scenarios again and basically found there was no fundamental change in the results. The last variation I’m going to test is to cause a daily randomisation of the data both for the solar insolation and the usage data and then see if there is any significant average departure from the basic findings. I’ll update that in a following blog.
The Siren Call of the Battery
When you have excess solar power you are able to export the power to the grid through a Grid Tie Inverter which means you can get a payment for what you can’t use. If you are paying 25p a unit for supply but only 10p for export it (SEG) would seem beneficial to have a way of storing your excess solar power to use it later. After all, why sell the energy at 10p a unit when you could be offsetting your need later on that would cost you 25p a unit. It seems like a no brainer, but it’s not that simple and it’s an easy trap to fall into.
To give you a clue, one scenario where you definitely wouldn’t want to use a battery is where the SEG is maybe 40p a unit because why would you not export the solar power immediately for the SEG? Why would you store the power to use it later at 25p? You wouldn’t unless you had another reason. From the graph above the solar panel system alone will pay back in about 8 years. So it would seem that somewhere around the point where the SEG approaches the Tariff a battery becomes a liability.
Pay Back Time
This is the time it takes to generate enough income from your system to pay it back. In very simple terms we can divide the cost of the system by the revenue it generates each year. There are two basic criticisms of this approach.
Firstly you may be doing the whole solar / battery scheme to go green. If that’s the case then this whole blog is just academic for you.
Secondly, maybe you should be taking into account future energy costs which could well increase but to offset this there are two factors. You could invest the money and get interest on it or pay off part of a mortgage/loan and not pay interest on that portion. Later battery systems are likely to be lower cost. It’s possible that solar have reached a minimum but batteries are still expensive for what they are and there could be a reduction in the cost of those as battery technology and volume manufacturing matures.
I’m just going to take the simple case that the PBT = Cost / Yearly Income as a relative indicator. To make reasonably accurate payback time calculations I’ve used the average cost of Home Batteries solar in Ref [1] and that for Solar Panel Systems in Ref [2] at the end.
The GOOD
The table shows what I mentioned in earlier that if you have a SEG of zero (or no SEG arrangement) then you
really do want to capture as much power in a battery as possible to use later on. For a typical 3.5kWp system,
without any battery your system PBT is 22 years whereas with a 4kWh battery it becomes 19 years. The chart also gives a good clue how to size the system (for my data) where you can see a green diagonal area extending from a 3kWp / 3kWh system down to a 6kWp / 9kWh system. Obviously the size of your roof will determine how big a
solar system you can have and mine would be a maximum of 3.5kWp. Along the 3.5kWp row there are ‘green’ values of mainly 18 years for a Tariff rate of 25p. As the payback time for all the 18s is the same you could choose any of the battery sizes 5-8kWh without any difference in payback time apart from the initial cost.
The corresponding Yearly Income table is shown here. A bigger battery would usefully capture more power but would be more expensive but the PBT is still very much the same so if you were looking at maximising your income, which would be real after the PBT then a bigger battery would do that.
So far, so good.
The term ‘Yearly Income’ also covers the amount that you don’t have to pay because you used solar power, either
directly or stored in a battery. The other part of ‘Yearly Income’ is a real amount that you would see in the export part of your electricity bill. Another factor to consider is I’ve not considered the losses in storing and retrieving power from a battery. Losses are probably in the region of 10-15% for the round trip.
The BAD
However if you have a SEG of 40p then the situation is very different. Here are the tables for that.
If you have a 3.5kWp / 4kWh system then your payback time is 13 years (intersection of 3.5kWp and 4kWh).
When the SEG was zero (previous table) the payback time was 18 years and it has now reduced to 13 years which looks like a good thing. However, when you look at the table for Battery = 0 (above, left) then the payback time is only 8 years. This is because you stored energy that you could have got 40p a unit for and then you used it later to mitigate a Tariff rate of 25p. You lost 15p for every unit you stored in the battery.
So what is being said here is that there is a benefit to getting a battery but certainly not if the SEG rate is more than the Tariff and perhaps a whole lot less (spoiler). As a diagram this is what we think we have in terms of a having a battery with a SEG. Any SEG less than the Tariff should be ok. Only it isn’t.
Although the problem is related to the Tariff rate, it’s not the only factor that that sets the dividing line between GOOD and BAD areas. It’s the Tariff rate minus the effective cost of the battery system. In fact the PBT becomes worse when the SEG is only 7p. You’ve paid out £3,000 for a battery and a SEG of only 7p makes the PBT worse. The battery does however increase the Yearly Income from £410 to £580.
The corresponding GOOD / BAD diagram looks like this. You have a much reduced good area left of the 7p line. The effective cost of the battery system in Tariff/SEG rate terms is 18p. (The difference between 25p and 7p)
On the PBT graph for a 3.5kWp solar system and my data you can see how the lines sort of pivot around the 5p point for different sizes of battery. The black line is for no battery. Any battery line (2,4,6, 9 kWh) above the black line means the payback time is longer although the Yearly Income will be higher.
Is the cost model right?
You always need to question the accuracy of any cost modelling you do as it can be expensive if you get it wrong. You have to get everything right in order to be able to make predictions and to explain what is going on. The Appendix explains how I put together the Cost Model and hopefully it is correct. One of the things I do to check the CM is to put in cases where you would expect a particular type of result.
Cost Model Sense Check
Case 1 – Set the Battery Cost to Zero with a high SEG rate.
Expected result: There should be no change in the PBT with
increasing battery size.
Actual Result: Not quite right. There’s a smaller but significant increase in PBT with battery size even with a free battery. It’s basically correct but something is missing. Going back to what I had said previously, with a SEG higher than the Tariff then a battery is preventing the export to the SEG at 40p and saving it for a time when you could have got the supply for 25p. For each kWh stored you are losing 15p a unit.
Case 2 - Zero The Battery Cost with a high SEG rate but change the Tariff so that it is the same as the SEG.
Actual Result: The PBT stays constant. This means the 18p (which is subtracted from the 25p to give the 7p pivot point) is made up of the battery cost and the difference between the SEG and the Tariff rate. One day I might try and find out how to calculate the 18p figure but I don’t think it’s going to be easy and I’ve no reason to do that at the moment.
Case 3 – A relationship between the Solar System size and Battery System size.
The Cost Model shows a minimum PBT relationship that is not unexpected. The best battery size scales with the solar system size. It shows a mismatch between a 1kWp solar and a 9kWp battery. You have the high cost of the battery but you are not using it because there is not enough solar power being generated to fill it.
The GOOD, the BAD and the UGLY
Do you think having a battery could get any worse? Unfortunately it’s going to get UGLY as well. You might be thinking, I really don’t care about the payback time because if I have a 3.5kWp / 9kWh system, eventually the Yearly Income will be pure profit. But after 18 years the battery is probably not going to have a lot of capacity left if it’s working at all. Add a number of failed inverters during that time and the payback time is going to be even longer.
Adaptive behaviour
In the same way as when I started using the Agile Octopus tariff I started concentrating my use towards times when it was less expensive, so users of a solar and battery system may be able to reduce the PBT by modifying their usage behaviour to better advantage but without knowing the savings made it would be difficult to know how to change your pattern of usage.
Conclusion
I think this is a classic case of simpler is better.
First Place: Solar System + Good SEG rate will beat most systems with a battery and will be cheaper. With solar panels lasting 25+ years you will get a system payback and then profits. The only downside I can see is that if SEG rates drop (quite likely) then the payback time will increase. Ref [3]
Second place: If you can’t get a SEG then a battery will improve the percentage of the produced solar power that you can use but look to match the size of the battery with the size of the solar. But you are not going to compete with a Solar System + Good SEG and the battery/inverter is likely to die before the payback time. The advantage here is that you are independent of variable and dropping SEG rates. Batteries are useful if you get a lot of power cuts and you need to run the house independent of the grid. But that isn’t a business case, it’s an insurance policy against disruption during a power cut.
And Finally: The whole Solar/Battery scene is changing all the time but the fundamentals will remain the same. If the Battery cost per kWh were to halve (to £0.27k per kWh) then you could expect the following for a battery system with no SEG. For a 3.5kWp / 4kWh system it would take two years off the payback time. The Yearly Income doesn’t change. That is probably as low as a battery system cost will go. The installation cost will probably stay the same. It’s interesting to note that EV Batteries are a lot cheaper than Home Batteries and even the much older ones are less than £0.15k per kWh (£6,000 for a 41kWh battery) when new. Second hand batteries are a lot less but an ultimate payback time of less than 15 years is not likely.
And if you don’t agree with what I’ve put in this blog …
Let me know where you think I have gone wrong. It’s certainly true that if the SEG = Tariff there is no point in storing the energy if the Tariff rate is constant (different if it’s variable) and that is the basis that I saw that a battery may be more of a liability than a help. The surprise was that the cost of the battery (very expensive for what it does) shifted the break-even point to such a low value of SEG as to make it more or less pointless unless you have another use for it like covering power cuts. If you do have an alternative calculation then I would like to know.
Appendix
Calculation of the Pay Back Time (PBT) for each configuration
1) You will need the Half Hour Panel Power Output (HHPPO) for its location and orientation with respect to due north for each half hour of the day and for each day of the year. You get this by multiplying the daily insolation pattern during that day by the insolation pattern for each day of the year. The output is the average energy delivered by the panel in that half hour. This avoids the need to know the precise weather.
2) Find the minimum between the HHPO and the HHPU to give the Immediately Usable Power (IUP) for each half hour.
The Immediately Usable Power (IUP) (IUP) = MIN(HHPO,HHPU)
The Immediately Usable Power replaces electricity at the Tariff rate.
3) Find the total solar excess power (SEP) output per day: SEP = SUM(HHPO) – SUM(IUP) and this is the power that is either exported or sent to a battery (BATT = energy stored in the battery) or exported.
If there is no battery or the battery is full the power is exported at the SEG rate = SEP - BATT
If the power is stored in a battery then it replaces electricity at the Tariff rate = BATT
In combination: For a given value of SEG and Tariff = 25p
For each size of solar system = 1, 2, 3,3.5, 4, 5, 6kWp
For each size of battery = 0, 1, 2, 3, 4, 5, 6, 7, 8, 9kWh
For each day calculate Year Income(YI) = SUM (IUP * Tariff + BATT * Tariff + (SEP-BATT) * SEG)
The Pay Back Time (PBT) = Cost ( Solar (size) + Battery(size) ) / Year Income (YI)
Limitations of this study
Without collecting data at the time it wouldn’t be possible to know what the panel output would be for each half hour for every day of the year. We do know the average power output of a 1kWp system for a year given the location details and we also know what the insolation would be for each half hour, for each day on average. Larger systems are scaled accordingly. The assumption is made that, on average, the weather will be random and the pattern of domestic usage is what it is for me and that the totals at the end of the year should be accurate enough to model the differences in solar system and battery system sizes if not the absolute values.
The assumption is made that the SEG rates will be constant for the payback time. This may not be the case and rates are evolving into Time of Use like Import Tariffs. Ref [3].
System Costs (Default system is Solar = 3.5 kWp, Battery = 4kWh)
The average Solar Cost can be described by the equation: £1.48k + £1.47k x kWp Ref [2]
The average Home Battery Cost can be described by the equation: £0.97k + £0.52k x kWh Ref [1]
This gives us the table below which shows the cost of each combination of Solar Panels and Battery sizes.
As an example a 3.5kWp array plus a 4kWh battery will cost: £6.65k + £3.05 = £9.70k (rounded to 10k in the table)
The Collection of Pay Back Time and Yearly Income tables for a Tariff of 25p for different battery sizes and for a range of SEG rates
References:
Quite a lot of information was obtained by asking ChatGPT which has provided quite comprehensive answers so I’ve just taken extracts and put them in here to prevent the blog getting too big
Ref [1] – Average cost of a Home Battery installation in the UK - 2025
Q: In the range of 1 to 10kWh in a step of 1 can you give me the average installation cost of a home battery
system in the UK?
Q: Can you give me the slope and intercept of the average values from that table?
The linear regression on the average UK battery installation costs gives:
Slope: £536.97 per kWh (the incremental cost per additional kWh of capacity)
Intercept: £966.67 (the fixed cost associated with the installation)
So, the approximate cost model is: Cost ≈ £0.537k × (kWh) + £0.967k
Ref [2] – Average cost of a Solar Power installation in the UK - 2025
Q: Can you give me the average cost of a solar installation in the UK for a range of kWp from 1kWp to 6kWhp
in steps of 1 as a table?
Q: Can you give me the slope and intercept of the average values from that table?
The linear regression on the average UK solar installation costs gives:
Slope: £1,471.43 per kWp (the incremental cost per additional kW of capacity)
Intercept: £1,483.33 (the fixed cost associated with the installation)
So, the approximate cost model is: Cost ≈ £1.471k × (kWp) + £1.483k
Ref [3] – The possible evolution of SEG rates
Q: What is likely to happen to the SEG price over the next 10 years if solar power production continues to
increase at the forecast rate If solar power production in the UK continues to increase as forecast—tripling to around 47 GW by 2030—the Smart Export Guarantee (SEG) price is likely to be affected by several market dynamics. Here's what could happen over the next 10 years:
1. Downward Pressure on SEG Prices (Likely). As more households and businesses export solar electricity back to the grid, the supply of distributed generation will increase. This increased supply could push down the SEG prices, especially during peak solar hours when the grid may be saturated with solar energy.
2. Time-of-Use SEG Tariffs- To counter oversupply at peak times, energy companies may increasingly offer time-of-use SEG tariffs, where export prices vary by time of day. SEG rates might drop during the day when solar is plentiful, and increase in the evening when demand is high and solar output drops. (Ed - Octopus do this)
3. More Dynamic Pricing Linked to Wholesale Markets- The SEG scheme may evolve to include real-time pricing linked to wholesale electricity markets, meaning export
prices will fluctuate more frequently. This could reward battery owners who can store excess electricity and export it
when prices are higher.
4. Government Intervention (Uncertain)- If falling SEG prices undermine public interest in installing solar panels, the government could intervene with
minimum SEG rates, similar to early Feed-in Tariff (FiT) schemes. However, given the UK's market-driven approach post- FiT, this is less likely unless adoption rates drop significantly.
5. Grid Constraints May Affect Local SEG Rates
In areas with grid congestion, network operators may curtail exports or offer lower prices, which could further
fragment SEG value across regions.
Summary:
Short-term (1–3 years): SEG prices may stay relatively stable with gradual shifts.
Mid to long-term (4–10 years): Likely trend toward lower average SEG rates, more variable/time-based pricing, and
greater value for those with storage or smart export tech.
Here’s a graph of how SEG prices could evolve under different scenarios from 2025 to 2035
Note: The only upward trend is Dynamic Pricing where higher SEG rates apply outside of the solar generation period and you will need a battery to time-shift
Further Meanderings
Just how cheap does a battery system have to be in order to make it worthwhile having the SEG as well?
I just had to investigate how the value of 18p comes about with the difference between 25p and 7p (the value of the
SEG where having a battery makes no sense). We know that there are two components to this. The difference between the SEG and the Tariff and the cost of the battery. I didn’t find out how to calculate it but I did see how the two parts of the battery cost changed the breakeven point
Installation = £0, Per kWh of battery = £0 This shows that for a zero battery cost the advantage or disadvantage of having a battery pivots around the Tariff rate in the way that the GOOD / BAD diagram illustrated.
Installation = £0.97k, Per kWh of battery = £0 The breakeven point (where a battery line crosses the zero battery
line) increases with increasing battery size.
Installation = £0, Per kWh of battery = £0.54k The reverse effect to the scenario above. The breakeven point is
better for smaller batteries.
Installation = £0.97k, Per kWh of battery = £0.15k In this last one I’ve kept the installation cost the same (unlikely to
decrease from what it is now) and put in the lowest possible battery cost and this is the best it is ever going to get.
With a SEG rate of 10p then you have a breakeven condition. The advantage is that if the SEG price were to drop then the payback time gets better but you would be better not having a SEG rate in the first place.
The Battery / SEG conclusion
To be honest, if you can export your surplus power then a battery doesn’t provide any advantage as its cost just
increases the payback time. In other words, if you can get a decent SEG, a battery is just a parasitic addition to the
system unless it performs another function like a backup in a power cut or where you have a ToU SEG where you are shifting power from the solar generation period into the demand peak (power arbitrage).
