What drives the cost of a commercial battery, and what it returns

The first question a finance director asks about battery storage is what it costs. The honest answer is that there is no single number, because a commercial battery energy storage system is engineered to a specific load rather than bought off a shelf. Two sites with the same floor area can need very different systems once you look at how they actually draw power across the day.

This page is capex-first, and it runs to both ends of the case. It sets out what goes into the capital cost, why the credible figure is survey-led, where the tax position sits, and then what that capital is judged against: the stacked value a battery earns, net of the losses and the degradation that are part of how the asset works. It is a plain-English orientation for a commercial buyer, not formal financial or tax advice. We confirm the numbers for your site.

The capital cost of a commercial battery is the sum of several engineered parts, not a single line item. Each one moves with the design, which is why a credible price needs the design to exist first.

• Usable energy, measured in kWh. The store of energy the system actually delivers, after the headroom the chemistry keeps in reserve. This is the largest single cost driver, and it is set by how much energy you need to shift across the day.
• Power rating, measured in kW. How fast the system can charge and discharge. A site that shaves a sharp half-hour peak needs more power for the same energy than one gently soaking up solar, and power and energy are priced separately.
• Cell chemistry. The bankable commercial systems use lithium iron phosphate (LFP) cells for their safety and cycle life. Chemistry sets the cell cost, the usable depth, the thermal management and the warranted life.
• Power conversion and EMS integration. The power-conversion system and the energy management system that decides when to charge and discharge. On a site that already has solar this has to be integrated with the existing inverters and metering, which is engineering, not a plug-in.
• Installation, switchgear and any grid works. Civils and mounting, the protection and switchgear that ties the system into your distribution board, and any reinforcement or metering the network operator requires under the G99 connection. Grid works in particular vary widely by site.

Two systems with the same power rating are not the same purchase if one holds twice the energy. Of the drivers listed above, the cells are the largest single line, and cells scale with the energy you are buying. The power-conversion system, the switchgear, the protection and much of the installation labour scale with the power instead, and they change little when the store behind them grows.

That is why duration is the axis a board should watch. Doubling the duration at the same power rating doubles the energy the cabinet has to hold, and cells are what that energy is made of, so the capital follows it while the conversion stage stays where it is. Adding power at the same energy moves a different and generally smaller part of the bill, and it is the change your grid connection is more likely to object to.

One maker in our UK distributor's commercial cabinet listing as at 15 August 2026 sells a 50 kW cabinet at 102.4 kWh with a 2-hour designation and the same 50 kW at 204 kWh with a 4-hour one. Same power, same conversion stage, roughly twice the store. We put no price on either here, because the capital only becomes real once your load model has settled the duration, and because a configuration is a specification while a distributor's commercial terms are not ours to publish. What the pair does show, plainly, is which axis the money follows. That listing is one dated view of one catalogue rather than the whole of what we can specify, so a maker or a configuration it does not happen to show is not thereby unavailable, and we re-confirm the part number and the configuration against the current listing and the manufacturer datasheet before either reaches a proposal.

The practical consequence is that "how much is a battery" cannot be answered before "for how long". A two-hour system and a four-hour system on the same peak-shaving duty are different assets with different capital, and choosing between them is a modelling decision rather than a budget one. That choice is set out on battery sizing.

You will see batteries advertised at a headline rate per kilowatt-hour. For a commercial system that figure is close to meaningless, because it captures only the cell cost and hides everything around it. The power rating, the conversion and control electronics, the switchgear, the install and the grid connection can together rival the cells, and they do not scale with energy in a straight line.

The ratio of power to energy is the clearest example. A system sized to discharge hard for a short peak carries more conversion and power hardware per stored unit than one sized to discharge slowly for hours, so the same kWh costs more on the first site. Quoting a single rate across both would be dishonest. We would rather model your actual half-hourly load and price the system that fits it than print a from-price that no real project matches.

The credible number comes from the survey, not the brochure. We run an in-house insured drone survey of the roof or site and model your half-hourly consumption in PV*SOL, so the system is sized from how you really use power rather than from a rule of thumb. That sizing then sets the energy, the power and the integration work, which is what the capital cost is built from.

Because the system is specified to your load, the figure you get is yours, and it carries the assumptions that produced it. We are brand-agnostic on the hardware. The cells and the conversion electronics are specified from the bankable makers we lead with, and the system is engineered and assured to the non-MCS commercial standard: the component standards for the install, the G99 connection with your network operator, and CDM 2015 on site.

Capital cost and the after-tax cost are not the same thing, and the gap is where capital allowances sit. For a commercial buyer the tax treatment of the spend changes the real cost of ownership, so it belongs in any honest costing conversation.

This is also where we have to be careful. The capital-allowances treatment of standalone battery storage is not settled in public HMRC guidance, so we will not assert a specific allowance against a battery on this page. Where storage forms part of a wider solar project the position can differ again. The right move is to size and price the system against your real load, then confirm the allowance position with your own tax adviser before you model the return.

The capital figure is the start of the cost story, not the whole of it. A commercial battery is an asset that runs for many years, so the honest way to cost it is over its working life: the capex, plus what it costs to keep running, plus what it costs to renew or retire at the end. Engineers call this the total cost of ownership, and it is the figure that should sit in a board paper rather than the headline capex alone.

Three running costs matter. The first is operations and maintenance: the inspection regime, the firmware and energy-management updates, and the upkeep of the thermal management and protection that keep the system safe and warranted. The second is the warranty itself. A bankable maker warrants both a capacity-retention curve and a throughput or cycle count, and reading those terms tells you how hard the battery can be worked before the warranty is the maker's problem rather than yours. The third is end-of-life. Lithium iron phosphate cells fade gradually rather than failing at a cliff, so a system can be augmented with added capacity or have modules renewed rather than wholesale replaced. We design the enclosure and the controls so that path is open, and we factor the cycle life the maker warrants into the lifetime cost rather than the nameplate. How that warranted life feeds the return is set out further down this page, where the stacked value is netted against degradation.

There is also a disposal and recycling obligation at the very end. Industrial batteries fall under the UK batteries regulations and the producer-responsibility regime the Environment Agency oversees, so end-of-life handling is a regulated cost, not a free disposal. We design for it, and we set out where that sits in your costing rather than leaving it to surface later.

A large share of what separates a cheap headline from a real quote is the work that keeps the system safe, compliant and connected. None of it is optional on a commercial site, and all of it lands in the capital figure.

Safety engineering is the clearest example. A commercial battery is designed, tested and certified against the stationary-storage standards, principally IEC 62933 for the system and IEC 62619 for the cells, with the fire-protection design framed against guidance such as NFPA 855 and the relevant BS EN provisions. The fire strategy, the separation distances, the detection and the ventilation are engineered into the build, and the Health and Safety Executive's expectations for a workplace installation sit over all of it. We set out how that is engineered on our battery fire safety page, and it is a real line of cost rather than a label.

The grid connection is the other big variable. A system above the small-scale threshold connects under an Engineering Recommendation G99 application to your network operator. What that application returns drives the cost: a site with spare capacity at the point of connection needs little, while a constrained site can need protection upgrades, metering, or network reinforcement before it can export or even charge at the rate you want. The network operator sets that, not us, which is why we confirm it from the connection process before pricing rather than guessing. The same survey work that sizes the system also scopes this, which is why the figure we give you is whole and not a base price with surprises bolted on later.

It helps to know what a credible commercial quote already contains, so you are comparing like with like. A complete figure from us covers the engineered hardware, the usable energy and the power rating, the power-conversion and energy-management integration with any existing solar, the switchgear and protection, the civils and mounting, the safety engineering, and the G99 connection works your network operator requires. It is built from the survey and the half-hourly model, so it reflects your real load rather than a representative system. The wider context for what a commercial scheme costs sits on our commercial solar cost guide.

It is worth being equally clear about what a capital figure does not settle. It is not the lifetime cost, because the running and end-of-life costs above sit on top of it. It is not the after-tax cost, because the capital-allowances position has to be confirmed with your own adviser. And it is not the return, which is the value the battery stacks against your own tariff and load, set out in the two sections below. A cheaper headline that omits the safety engineering, the connection works or the integration buys a different and smaller scope, priced lower. The honest comparison is whole-scope against whole-scope, which is what the survey-led figure gives you.

A capital figure means little on its own, so the second half of the question is what the money buys back. A commercial battery does not earn from one thing. Its return is stacked from several jobs running in the same day, and no single one of them usually pays for the asset alone.

• Peak shaving. The battery discharges through your highest-demand half-hours so the meter sees a lower peak, which cuts the capacity and demand charges billed on the size of that peak rather than on the energy used. It is the largest single driver on most commercial sites, and it is set out on peak shaving.
• Time-of-use arbitrage. The battery charges when import is cheap and discharges when it is dear, banking the difference across the day. It only pays where the tariff has a genuine spread, which is covered on time-of-use arbitrage.
• Solar self-consumption. Surplus generation that would otherwise be exported at a low rate is stored and used on site, where it offsets an expensive import. A unit you avoid buying is worth far more than a unit you export, as set out on solar self-consumption.
• Grid-services income. Where the site and the asset qualify, the battery can earn a variable income from balancing and flexibility markets. It is a layer of upside on top of the stack rather than the foundation of it, and eligibility is site-specific. The mechanisms and the route into them are on grid services and the Capacity Market.

Against that value you net two costs that are physical and unavoidable. Round-trip efficiency means every cycle loses some energy to conversion and heat, so a stored unit is always worth slightly less than the unit you bought, and an arbitrage gain has to clear that loss before it counts. Degradation means usable capacity fades gradually across the working life, so the value the asset stacks in the later years is lower than the value it stacks in the first. We model both against the maker's warranted capacity and cycle life rather than the nameplate, and across the whole life of the asset rather than its best year.

Which drivers actually pay depends on your load shape and your tariff. A site with a sharp afternoon peak leans on shaving; a site on a wide day-to-night price spread leans on arbitrage; a site with a large array leans on self-consumption. Most strong cases use all of them, and the mix comes out of the half-hourly model rather than a rule of thumb.

The return moves with the size you install, and not in a straight line, which is why a single figure means little without the size attached to it. A battery sized to clip only your sharpest demand peaks is small, cheap to install and spends most of its capacity on the highest-value job, so the value it stacks per kWh is dense. Scale it up to ride longer arbitrage windows or to carry a backup load and the later kWh do lower-value work, so the average return per kWh softens even as the total saving rises. The honest question is never what the return is, but what the return is at this size, for this load, on this tariff.

So we run several candidate sizes against your half-hourly data and show how the stacked value and the payback shift as the battery grows. That lets a board pick the point where the marginal kWh still earns its keep rather than the largest system a supplier can sell. The sizing logic that feeds it is on battery sizing, and the capital that anchors the payback half of the calculation is the survey-led figure this page is built from; we will not pretend a return without a real cost beneath it.

The assumptions under that model are named rather than buried. Your import price, your export arrangement, the demand-charge bands and the tariff windows come from your own bills and your supplier rather than a national average. The generation and dispatch profile comes from the same in-house insured drone survey and half-hourly PV*SOL run that sizes the system. Degradation and cycle life come from the maker's warranted terms for the exact product. All of it is disclosed so your finance team can rerun the case on its own numbers instead of taking ours on trust, and any saving, payback or spread we show is modelled and market-dependent rather than promised. An interactive calculator is coming and will let you explore the drivers yourself; it does not exist yet, so we will not point you at a number it has not produced. How payback itself is constructed is unpacked in the commercial solar ROI and payback guide, and the funding routes sit on commercial finance.