A school roof generates most of its electricity in the weeks the buildings are empty. We say that first because nobody else does, then we measure the gap, and then we arrange the survey, the design, the grid application and the installation through our MCS-certified partner.
UK wide / RAAC and asbestos settled first / term and holiday load measured
Book a school roof survey
School rooftop solar is a photovoltaic array mounted on an education building that converts daylight into electricity the school consumes before it draws anything from the grid. The solar panels sit on the existing roof, the inverter turns direct current into the alternating current the distribution board already uses, and any electricity the site does not consume is exported under a Smart Export Guarantee tariff, the government backed scheme Ofgem has administered since it launched on 1 January 2020, which obliges larger licensed suppliers to offer a price for units sent back to the grid. Every unit the buildings take from the array is a unit nobody has to buy.
A solar system is not a generator and it does not replace the energy contract. Solar panels cut the volume of electricity a school buys during daylight hours, so the energy bill falls in proportion to what the site consumes on the day rather than to what the roof produces across the year. That distinction does more work on a school than it does anywhere else, and it is the reason the next section exists.
Three things decide whether solar is worth doing on schools and colleges: the roof, the timetable and the grid connection. Most enquiries reach us because energy costs have risen faster than the budget, which is a fair reason to look, and because solar power is now the cheapest generation a site can put on a building it already owns. What we find is that the building and the calendar decide the outcome far more than the panels do, and we say so plainly when the answer is that a site should wait.
A school's electricity demand collapses in precisely the weeks its roof generates the most, which makes self-consumption on an education building structurally worse than on a warehouse or a factory. This is the single most useful thing we can tell a school about solar, and almost nobody selling it says so.
Roughly half the generation arrives when nobody is in the building. Set the median monthly profile for the 397 towns on this site against a 190 day school year and 49 percent of the output lands while the school is open, 51 percent while it is shut (EU PVGIS v5.2, SARAH3, modelled as described under fig. 2). A business that opens every weekday catches 71 percent of the same roof's output on its working days. The school gives up 22 percentage points purely to school holidays, and that gap is the whole reason this site exists.
It is worth separating the two halves of that loss, because only one of them is a school problem. Weekends take 29 percent of the year's output, and every organisation in the country carries that. The 22 point gap is the school holidays on their own, and it is concentrated rather than spread: July and August between them carry 26 percent of annual output and the buildings are largely empty across both, while December, January and February carry 9 percent and those are the months a school is full, lit and heated. The daily shape compounds it, because the teaching day ends in the middle of the afternoon while the array is still producing strongly.
Four things improve that picture, and a school with two or three of them has a genuinely good case rather than a marginal one. A battery moves generation from the middle of the day into the morning warm up and the after school hours. Holiday letting, a community sports facility or a summer scheme keeps a load under the roof in August. A swimming pool, a commercial kitchen, a data room or mechanical ventilation gives the site a base load that runs whether or not the children are in. And accepting a lower export rate for the holiday surplus, rather than pretending it will be consumed, at least produces a model that survives contact with the meter.
The Department for Education concedes the point in its own words, and it is worth reading closely. Its Education Estates Strategy of 16 February 2026 estimates that on average a typical school could save up to £25,000 a year if it had solar panels with complementary technologies installed such as batteries. The conditional clause at the end of that sentence is the whole argument on this page. The headline figure depends on storage, because without something to shift the generation the school cannot use what it makes.
We measure the split rather than assume it. The survey reads half hourly consumption for the supply across a full twelve months, sets it against modelled generation for the site, and reports what proportion the school would actually use. A scheme sized to the roof will look better on paper and worse on the bill than a scheme sized to the timetable, and the difference is not small. Savings quoted from anything other than that measured share are a modelling assumption wearing the clothes of a forecast.
| Month | Session days | Output, kWh per kWp | School open | School closed |
|---|---|---|---|---|
| Jan | 20 | 24 | 15 | 9 |
| Feb | 15 | 38 | 20 | 18 |
| Mar | 21 | 73 | 49 | 24 |
| Apr | 12 | 104 | 42 | 62 |
| May | 17 | 120 | 66 | 54 |
| Jun | 21 | 123 | 86 | 37 |
| Jul | 11 | 124 | 44 | 80 |
| Aug | 0 | 105 | 0 | 105 |
| Sep | 20 | 82 | 55 | 27 |
| Oct | 17 | 51 | 28 | 23 |
| Nov | 21 | 30 | 21 | 9 |
| Dec | 15 | 21 | 10 | 11 |
| Year | 190 | 895 | 436 | 459 |
A school building suits solar when it carries a large uninterrupted roof, a load beneath it worth serving, a structure with capacity to spare, and a covering with years of service life left. Most sites hold several buildings put up in different decades, and on a typical estate only one or two are worth an array, which is why the survey looks at the whole site rather than the block somebody happened to ask about.
Area is not the scarce thing here. The Department for Education puts the school estate at around 55 million square metres of roof, with the potential to support an additional estimated 0.8 to 1.9 GW of solar power in its current condition (Education Estates Strategy, 16 February 2026). The constraint has never been how much roof the country's schools own. It is which of those roofs will carry an array, and which of them have a load underneath worth serving.
Age is the variable that separates a school from almost every other commercial roof we look at. A site can carry a Victorian pitched slate block, a 1960s system built teaching wing with a flat roof, a 1990s sports hall and a new build primary in the same grounds, and the four take four different answers. The two findings that matter most both belong to the older buildings. Reinforced autoclaved aerated concrete may be present in the roofs of anything constructed or modified between the 1950s and 1990 (DfE identification guidance, published 14 December 2022 and last updated 20 July 2026), and asbestos containing materials were used extensively in school construction from the 1950s until the material was banned in Great Britain in 1999 (HSE, Asbestos in schools).
Neither finding rules a school or a college out. Both change the order of the work, and both are the reason a school quotation and an industrial quotation for the same kWp look nothing alike. What we will not do is tell a school that a particular roof is safe or unsafe. That is a judgement for a competent structural engineer working to the Institution of Structural Engineers' guidance on RAAC panel investigation and assessment, and our part is to establish which buildings need that assessment before anything is designed on top of them.
Usually the largest uninterrupted roof on the site and often the best candidate. Evening lettings, weekend club hire and a summer holiday programme give it a load in the hours and weeks the teaching blocks have none, which is worth more here than roof area alone.
On a site of a certain age this is a flat or shallow roof over a system built frame, and it is the building most likely to fall in the RAAC and asbestos window. Large area, light electrical load, and the one that has to go to a structural engineer before anything else happens.
The strongest match on any school estate. Circulation pumps, dehumidification and water heating run through daylight hours all year, and where the pool stays open to the community through August the load holds up in exactly the weeks the rest of the site empties.
Refrigeration runs every day of the year including the holidays, and the servery, extract and dishwash draw hard across the middle of the day in term time. A modest roof with a genuinely useful load under it, which is the opposite of most of the estate.
Structurally the most straightforward roof we survey: known construction, known capacity and a covering with its full service life ahead of it. The load beneath it is small and it closes for the whole of August, so the case rests on export value or on a battery rather than on consumption.
A longer day, more IT, more mechanical ventilation and a term calendar that differs from the school one. Colleges also let space more readily outside term, so the self-consumption share on a college block is commonly better than on a secondary school of the same size.
Illustrative layout for an estate of this size. The array is drawn at about 345 kWp, roughly 53 percent of the 650 kWp these three roofs would hold, because a school sizes to the load it can use in term time rather than to the roof it has: generation it cannot use is exported at a much lower rate than it pays to import. The hatched block carries no array because its structure has not been assessed, which on a school of that era is the usual starting point rather than an exception. Your own figures, and which of your blocks can take an array, come from the roof survey.
A school solar survey is a structural, electrical and commercial assessment that establishes what the buildings can physically carry and what the resulting system would be worth against the school's own consumption. It carries no fee, it carries no obligation, and the reports are the school's to keep whoever ends up doing the work.
The age of the estate sets the order. Where a block falls in the RAAC or asbestos window, the structural and asbestos position is settled before an array layout is drawn rather than after, because a design resting on an assumption about a 1970s roof is not a design. Where a covering is near the end of its life, re-roofing first is normally cheaper than lifting an array off in five years to do it then.
We also ask what the estate plan says. A block scheduled for replacement, a new teaching wing, a kitchen refit, a move to a heat pump or a planned change in how the site is let through the holidays all shift the load an array is being sized against. A system meant to last two decades should be sized against the electricity the school will use in ten years rather than the electricity it used last year, and on a school the holiday letting question changes the answer more than anything else on that list.
Annual generation is the product of the system size in kWp and the yield for the location, expressed in kWh of electricity for every kWp installed. Across the towns on this site, modelled yield runs from 704 to 1,028 kWh per kWp a year (EU PVGIS v5.2, SARAH3), and every location page carries the figure for that area alongside its source, so a roof in Cornwall and one in Aberdeenshire are not quietly treated as the same number.
Value follows consumption rather than generation. Electricity used on site displaces the day rate the school would otherwise pay, which is the highest value outcome available. Electricity exported earns the Smart Export Guarantee rate from the chosen supplier (Ofgem), which is materially lower. On most commercial buildings that split is a detail. On an education building it is the whole question, because the calendar pushes so much of the generation into weeks when nobody is there to use it.
The monthly shape is what a school should look at, not the annual total. The brightest month produces roughly six times what the dullest one does across our towns, and the annual figure hides exactly the thing that decides the return. We model twelve months of generation against twelve months of half hourly consumption for the same site, and the resulting self-consumption share is the number worth arguing about.
One dot per town, positioned at its centroid and shaded by modelled annual yield in kWh per kWp. Source: EU PVGIS v5.2.
The spread between the best and worst town on this map is 324 kWh per kWp, which is about 46 percent. On a 500 kWp array that is roughly 162,000 kWh a year of difference for the same money spent.
Free solar panels for schools normally means one of three arrangements, and only one of them is genuinely free at the point of use. The phrase does a lot of work in this market and it is worth taking apart before a governing body signs anything.
The first is a grant, where public money pays for the equipment and the school keeps the electricity. The Great British Energy Solar Partnership is the live example. As at 16 July 2026, 245 schools and colleges already had government funded solar panels, a further 100 were announced as joining, backed by up to £40 million, and 150 schools and colleges across Yorkshire and the Humber, the East Midlands and the South East were to pilot a different model, with the private sector installing and maintaining the panels at no upfront cost (gov.uk). The Partnership is an investment of up to £255 million in total across schools, colleges, NHS sites and military facilities. Where a school is in it, the panels genuinely are free to that school.
Now the part that almost nothing else written on this subject will tell you, and for most readers it is the most useful sentence on the page. There is no application. The published government material describes schools being selected and targeted, primarily clustered in areas of deprivation in the North East, the West Midlands and the North West, with at least ten schools in each region of England. No government page publishes an application form, an application portal or a closing date, because the allocation is made centrally and the school is contacted. So if a supplier hands a business manager an application guide for the current round, the question to ask is which gov.uk page it came from. We have looked and we cannot find one. A school cannot apply its way onto the Great British Energy programme, and a proposal whose funding case rests on doing so is not a funding case.
The 150 school pilot is worth watching for a different reason. A model where the private sector installs and maintains the panels at no upfront cost is, in substance, the second arrangement below wearing a government badge, and it should be read with the same care as any other contract of that shape.
The second is a power purchase agreement, which is the offer most schools mean when they say free solar. A third party funds and owns the array, installs it at no capital cost, and sells the school the electricity it generates under a contract that commonly runs fifteen to twenty five years. Nothing is paid up front, which is the attraction. What is given up is the cheapest electricity on the site for the length of the term, plus whatever the contract says about roof access, about who repairs the covering underneath, and about what happens on a lease assignment or an academy conversion. A PPA can be a sound answer for a school with no capital and a strong load. It is not free, and it should be read by whoever reads the school's other long contracts.
The third is a community energy scheme, where a local society raises the capital, owns the array and shares the benefit with the site. Community energy has put solar on a good number of British schools and colleges, and the terms are usually more generous than a commercial PPA, but they depend on a local organisation existing and on a share offer succeeding, so they are an opportunity rather than a plan. Our funding page sets the three routes out against each other, with the published position on each.
The funding route a school can use is set by what kind of school it is, because an academy trust, a maintained school and a further education college each sit under a different capital regime. Getting that wrong wastes a term, and it is the most common reason a promising scheme never reaches a quotation. Great British Energy installations aside, which are allocated rather than applied for, almost everything below is capital the school or its authority already controls.
An academy trust holds its own capital and its own balance sheet. Devolved formula capital accumulates across the trust, the Condition Improvement Fund runs as a competitive bidding round for smaller trusts, standalone academies and sixth form colleges, and larger trusts receive a school condition allocation to spend against their own estate priorities. A trust that can pool several years of devolved formula capital across a number of schools is in a materially stronger position than a single primary school trying to fund an array from one year of it, and pooling is the single change that most often moves a trust scheme from marginal to worth doing.
A maintained school sits inside its local authority's capital programme, and the authority is the body that holds the funding and, in most cases, the freehold. That makes the local authority a necessary party rather than an interested one. Some authorities have run their own solar programmes across the schools they maintain, which is usually a better route than a single school going alone, and it is worth asking the question before anything else.
Across both, the public sector decarbonisation route has funded heat and energy efficiency work on the school estate in phases rather than continuously, and government has run its clean energy support for schools and colleges in announced programmes rather than as an open door. We treat none of those as an available route until we can point at a current, open, published one, and at the time of writing we are not pointing at any. We do not state grant amounts we cannot source, we do not present a closed or centrally allocated programme as something a school can bid into, and we will tell a school plainly when the honest answer is that no scheme is open to it this year. That answer is more common than the market admits.
One structural point is worth making to any trust with a net zero commitment in its annual report. A capital route the trust controls, such as pooled devolved formula capital, leaves the trust owning the array and counting the generation against its own emissions. A route where a third party owns the equipment does not, or not straightforwardly, and a trust that has told its members it is working towards net zero should settle that question before it signs rather than after.
A battery is a storage system that holds generation the site cannot use at the moment it is produced and releases it later, and on a school it does more work than on almost any other building type. Battery storage converts exported electricity into on-site electricity, which raises the value of every kWh the solar panels make.
The school day is the reason. Heating and ventilation come on before the array does, the teaching day ends in the middle of the afternoon while generation continues, and after school clubs, evening lettings and a caretaker's site run all draw power in hours the roof does not cover. A battery shifts the middle of the day into both ends of it.
It is not a fix for the summer holidays, and anyone presenting it as one is overselling. A battery stores a day, not a season, so six weeks of empty buildings in the highest yield months stay six weeks of export whatever is installed. The honest case for storage on a school is the daily shape and a restrictive export limit, not August. We size any battery against twelve months of half hourly data rather than a rule of thumb.
The cost of a school solar system is set by the array size, the roof it lands on and the grid work the connection needs, and the payback is the electricity the school stops buying, valued at the rate it would otherwise have paid, set against that cost. On an education building the second half of that sentence is where the care goes, because the self-consumption share is lower than a generic model assumes.
Cost for every kWp falls as the array grows, because access, design, the grid application and commissioning are largely fixed and spread across more panels. Roof condition is the variable that moves a school quotation most: a sound 1990s sports hall roof and a 1970s teaching block needing structural assessment before anything is fixed to it are two different projects at the same kWp. We ask our partner to break a quotation into those lines so a business manager can see which is which. We do not publish a price per kWp, because a figure given without seeing the buildings is a guess, and a governing body should not be asked to approve a guess.
The tax position is where school advice most often goes wrong, because it gets copied across from commercial property. Capital allowances, the special rate pool and full expensing are reliefs against corporation tax on trading profits, and most schools are not paying corporation tax on trading profits at all. An academy trust is an exempt charity and a maintained school is part of its local authority, so the allowances a trading company would claim on an array are simply not in play. Do not let a proposal show them in the model.
VAT is the relief that does apply, and it works differently by school type. An academy trust recovers VAT on its non-business activity through the refund scheme in section 33B of the Value Added Tax Act 1994, and a maintained school recovers through its local authority under section 33. Where a site is let commercially, part of the recovery can be restricted, which is worth checking with the trust's own advisers before a holiday letting argument is used to justify the scheme. Our cost page sets out what each of these does to a real quotation.
A school spends public money under procurement rules and reports the decision to a governing body or a trust board, which makes the evidence behind a solar scheme as important as the scheme itself. This is the part of the public sector process that has no equivalent in private property, and it is the part suppliers most often ignore.
Three practical consequences follow. Quotations have to be comparable, which means they have to be priced against the same scope, which in turn means somebody has to have surveyed the roof before any of them were written. A long contract such as a power purchase agreement is a commitment the school's next business manager inherits, so it belongs in front of whoever reviews the school's other long term liabilities rather than being signed off as an energy decision. And the case put to governors needs to show the self-consumption assumption in the open, because that is the assumption the whole saving rests on and the one a generic model gets wrong on a school.
Our survey pack exists for that reason as much as for the engineering. It gives a business manager one set of measured facts about the estate that every bidder prices against, and it is the school's property whoever wins the work. Where a trust runs its own framework or the local authority has a route already in place, that route usually beats going alone, and we will say so.
One last point on how a case is written up. The strongest paper we see for solar for schools puts the self-consumption share, its source and its weakest assumption on the first page, states the savings as a range rather than a single number, and says what would have to be true for the low end to happen. Governors approve that paper quickly because it does not ask them to take anything on trust. A paper that leads with a headline saving and buries the modelling improves nobody's position, least of all the business manager who has to defend it in three years.
A rooftop array gives a school a working piece of energy infrastructure that pupils can see, measure and argue about, which is a genuine benefit and also the one most often inflated in a sales document. It is worth having and it is not worth paying a premium for.
The practical version is modest and real. Most inverters publish generation data, a display in a public part of the building makes it visible, and the underlying figures support work in science, geography, mathematics and design across primary and secondary. A primary school can use the daily curve directly; a sixth form or college can work with the raw half hourly export. Either way, ask for the monitoring and the data feed to be specified at design stage, because retrofitting a display and an export of the data afterwards costs more than including it.
The wider case is about the estate rather than the curriculum. A school that generates part of its own power reduces the emissions attributed to its electricity, which matters to trusts reporting against a net zero commitment and to local authorities reporting net zero progress across their own estate. Clean energy generated on the roof is the part of a school's net zero position that is genuinely within its control, unlike the grid it buys the rest from. That is a real outcome, it is measurable, and it sits alongside the savings rather than instead of them.
Installation is a fixed sequence that runs from survey through to commissioning, and on a school it is governed by safeguarding, by access and by the calendar rather than by how long the panels take to fix down. Rooftop solar on a school building in England is usually permitted development, so planning is rarely the item that sets the date.
The connection is the long pole, which is why we submit it early rather than after a contract is signed. Anything above 3.68 kW per phase runs under G99, and the network operator sets the export limit.
Work starts with the survey and the design, then the grid application, because anything above the smaller threshold connects under G99 rather than by G98 notification and the network operator sets the export limit. Then access, mounting, the modules, the DC and AC electrical work and the tie in to the board. Then commissioning, testing, the MCS certificate and handover of the operation and maintenance file. Our partner installs to that sequence on every site, and where a block has to be surveyed structurally first, the design waits for the engineer rather than the other way round.
The summer holidays are the window, and that is the one scheduling fact every school already knows. Six weeks with the buildings empty is enough to install most rooftop arrays, it removes the safeguarding complication of contractors moving through an occupied site, and it means scaffold and a mobile platform can stand where they need to. The consequence is that the grid application has to go in early in the spring, because the connection is the long pole and a school that starts the conversation in June has usually lost that year's window. School and college installations therefore cluster in August, which is also when installers are busiest, so an early decision buys a better programme as well as a better price.
Maintenance afterwards is light: an annual inspection, a check on the inverters and clearing anything that has grown up to shade the array. An annual energy review is worth running alongside it, because a change in how the site is used through the holidays changes the self-consumption share the whole case was built on. Inverters carry the shortest working life in the system, so plan on replacing them once inside the array's life and treat it as a maintenance cost rather than a surprise.
Send the postcode, roughly when each block was built and what the site is used for outside term. We come back with what the roofs can carry, what an array would generate against your own consumption term by term, and what the funding routes look like on those numbers.
No survey fee, no obligation. Lenzie Consulting Ltd arranges the survey and passes your details to our MCS-certified installation partner so they can quote.