Capex (buy outright) for commercial solar.

You pay for the system up front and the asset goes on your balance sheet. A well-designed commercial array is a long-life asset, typically rated for around 25 years, so you carry it and you keep the saving it produces for that whole period. Because you own it from day one, the generation it offsets against your import bill is yours in full, and the surplus you export is yours to monetise through an export route. The value sits in self-consumption: an exported unit pays far less than a unit you avoid buying, so we design for using the power on site first.

Capex also gives you the tax relief in the year you buy. Solar PV is a special-rate asset, and the Annual Investment Allowance gives a 100% first-year deduction within the £1m annual AIA limit, which brings the relief forward rather than writing the cost down over many years. Full expensing does not apply to solar, and the former super-deduction did not either, so the Annual Investment Allowance is the route to the first-year deduction. That is the honest hook, not tax advice. The exact relief depends on your profits, your other capital spend against the £1m limit and your accounting period, so confirm the position with a qualified accountant or tax adviser before you rely on it.

What you are buying is generating plant with a long design life, not a subscription. Commercial modules are typically warranted on a performance curve running to around 25 years, and a well-built array carries on producing beyond it at a reduced output. That matters to the capex case in a way it does not to any other route, because once the system has covered its own cost every further year of generation is margin. A funder's agreement, by contrast, tends to run for a fixed term and take the good years with it.

Owning the asset also means owning its upkeep, and we would rather put that on the table than leave it as a surprise. Inverters are the shortest-lived part of a commercial system and should be planned for as a mid-life replacement rather than treated as a fault when it happens. Monitoring, periodic inspection, cleaning where the site needs it and the electrical retest interval all sit with you. None of that is large against what the array generates, but it belongs in the model, so we set the operations and maintenance position out alongside the generation figures rather than after the order.

The other half of ownership is control. You decide when to add battery storage, when to extend the array onto a second roof, who maintains it and what happens at end of life. Under a lease the asset is yours but the finance agreement is not, and under a power purchase agreement neither is. If the building or the business changes, the route that leaves you holding the asset is the one that leaves you with options.

A capital request for generating plant is judged like any other capital request, which means the paper has to stand on its own. We build it from four things. Your half-hourly consumption data for a full year, because the value of the system is set by how much of its output you use on site rather than by how much it generates. An on-site drone survey flown by our own qualified, insured pilot, which gives real roof geometry, real shading and real obstructions instead of a satellite trace. A PV*SOL model built on that survey, which is the bankable-grade software lenders model against. And the connection position with your distribution network operator, because a G99 application is what usually paces a commercial programme.

From those we produce the split between self-consumed and exported generation, the resulting import saving, the year-one capital-allowances position on the spend, and the maintenance cost over the life of the asset. We model the conservative end rather than the best case, because a board paper that overstates the return is the one that gets found out in year two. Where a figure is sensitive to something outside our control, such as your import tariff at renewal, we show it as a range with the assumption named rather than as a single number.

We do not publish a per-kWp figure or a headline payback on this page, and we will not put one in an email before the survey. One costed figure for your building, produced after the roof has been looked at, is worth more to a finance director than a rate that has to be revised. The first feasibility read costs you nothing.

The relief that makes capex strongest only has value against tax you actually pay. If the business has little or no taxable profit in the year of the spend, or has already used its Annual Investment Allowance headroom on other capital, the first-year deduction does less work and the case rests on the energy saving alone. It may still be a good case. It is simply a different one, and your accountant should size it before the board sees it.

Tenure is the second test. If you occupy the building on a lease with only a few years left, or on terms that do not let you fix generating plant to the roof or take it with you, buying an asset with a 25-year life is the wrong shape for the commitment. The same applies where the roof covering is near the end of its own service life, because the honest sequence is to replace the roof first and mount the array on it afterwards rather than pay twice to remove and refit a system.

The third test is the cost of the capital itself. Where the money would otherwise earn or save more elsewhere in the business, or where borrowing to buy costs more than the system returns, spreading the cost or letting a funder carry it is the rational answer. That is what the alternatives to buying outright are for, and we will set them beside capex on your own modelled figures rather than steer you to the route that suits us. None of this is financial or tax advice; confirm your own position with a qualified adviser before you commit.

The trade-off with capex is plain: it ties up capital. You commit the full cost up front and the asset sits on your balance sheet rather than off it. For a business with the cash, or one that would otherwise leave it earning little, that is the point, because you keep the saving and the tax relief rather than handing a slice to a funder. For a business that needs to protect working capital, leasing or hire purchase spreads the cost so the system can be cash-flow neutral, and a power purchase agreement removes the outlay entirely, though a PPA largely benefits the funder who owns the asset.

Capex is the better call wherever the capital is available and the cost of that capital is lower than the return the system earns. Grants can in principle reduce the outlay, but they exist mainly for eligible public-sector bodies and specific schemes; most private commercial sites do not qualify, so we will tell you straight whether yours does rather than build a case around money that is not there. Whichever way you fund it, your own figures come from the on-site survey and the PV*SOL model, not from a rule of thumb.