Storage that earns its place, sized to your load.

Commercial storage pays in more than one way, and the value stacks: a lower metered peak, cheaper energy shifted from off-peak, more of your own solar used on site, and sometimes income from grid services. Which of those matters most depends entirely on your load and your tariff, so the design starts with your half-hourly data, not a per-kWh rule of thumb.

This page orients you before the survey: what the equipment is, how it is coupled to your supply and where it sits on your electrical infrastructure, then how it is sized and costed, the chemistry and safety engineering behind it, and the jobs a battery can do. The tax treatment of standalone storage is not settled, so we are careful with the numbers and point you to your own adviser.

A commercial battery is plant, and no single job pays for it

A commercial battery energy storage system, or BESS, is a block of lithium cells, power-conversion electronics and control software that sits between your incoming supply, your solar array and your loads. It charges when power is cheap or your panels are generating more than the building can use, and it discharges when power is expensive or your demand spikes. On a site over 50 kWp the unit is rarely a single wall box. It is a rack, a cabinet or a row of cabinets, with its own protection, metering and an energy-management layer deciding moment by moment what to do.

The reason a commercial battery earns its keep is that its value stacks: no single one of those jobs usually justifies the capital on its own, but together, against the right load and tariff, they can. The starting point is always how the equipment behaves, which the next three sections set out plainly. Everything after that is a question of which parts of the value stack your specific site can actually reach.

A commercial battery is four building blocks working together

A battery energy storage system is best understood as four parts, each with a clear role, rather than as a single box. Treat it as a stack.

• The battery modules and racks. The cells that actually hold the charge, grouped into modules, then into racks or a cabinet. Commercial systems are almost universally lithium iron phosphate, chosen for thermal stability and long cycle life. This is the part that scales: more modules means more stored energy.
• The battery management system. The electronics that watch every cell, keeping each one inside its safe voltage and temperature window, balancing them against each other, and shutting the system down if anything goes out of range. It is the safety and longevity layer, and it does serious work at this scale.
• The power conversion system, or battery inverter. Batteries store and release direct current; your building runs on alternating current. The power conversion system converts between the two in both directions and controls how fast the battery charges and discharges. Its rating in kilowatts sets how hard the battery can push or pull at any instant, which is a separate question from how much energy it holds.
• The energy management system. The controller and software that decide what the battery should do and when. It reads your site metering, your tariff and your solar generation, then instructs the power conversion system to follow whatever strategy you are running, from self-consumption to avoiding peak-rate import to shaving demand peaks. The control layer is covered in depth on EMS software.

It charges from your own solar or from cheap import, and it sits behind your meter

The battery has two ways to fill up and one job when it empties. It charges either from your own solar, soaking up generation the building is not using at that moment, or from the grid when import is cheap, typically off-peak overnight on a commercial tariff. The energy management system chooses the source from price and from the strategy you have set. When the site needs power and importing is expensive, or when demand is peaking, the battery discharges through the conversion system to supply the building. On a site with solar that is what self-consumption means in practice: stored daytime generation offsets expensive evening import instead of being exported at a fraction of what you pay to buy electricity back. The economic logic is the gap between your import rate and your export rate, and the battery is the tool that captures it.

No battery returns every unit it takes in, and a case that ignores that reads high. Losses sit in three places: the conversion between direct and alternating current in both directions, the cells warming slightly as they charge and discharge, and the always-on draw of the controls, the battery management system and any thermal management. Together they give the round-trip efficiency, the share of energy in that you get back out. It is a measured property of the specified kit and the duty rather than a brand boast, so we read it from the current datasheet and size on the usable energy that survives the round trip rather than the nameplate figure.

Physically the system is installed behind the meter, on your side of the supply point, connecting into the site distribution board or a dedicated panel alongside any solar and the building's normal loads. That position is what lets it offset import and support self-consumption directly, because the energy it cycles is energy you are managing for your own consumption rather than power you are trading as a generator. The installation is engineered to the same standards as the rest of your electrical infrastructure: wired to BS 7671, verified to IEC 62446-1, declared to your distribution network operator under G99 where required, and delivered under CDM 2015 on a JCT or NEC contract. There is no MCS involved, because MCS is the domestic scheme; assurance at this size is the engineering stack, signed off by an AM2-trained qualifying supervisor.

Most commercial projects end up mixing AC and DC coupling

The coupling decides where the battery sits relative to your inverters. A DC-coupled system shares an inverter with the solar array and stores energy on the DC side before it is ever converted to AC. That keeps round-trip losses down when the battery is mostly soaking up surplus solar, and it suits a new-build where panels and storage are designed together. An AC-coupled system has its own inverter and connects on the AC side, alongside the existing supply. It is the natural choice for retrofit, for storage that has to work independently of the solar, and for sites where the battery's main job is tariff arbitrage or peak shaving rather than catching PV surplus.

Most commercial projects end up as a considered mix. A large site may run DC-coupled storage on a new PV block and AC-coupled storage to handle the wider load, governed by one control system. We choose the topology from what the site has to do, how much solar is involved, whether it is retrofit or new, and how the connection is configured, rather than from whichever arrangement a brochure happens to show. The hardware that ties it together is set out in the two sections above, and the brain that coordinates it is the subject of EMS software.

We size from your half-hourly data because kW and kWh answer different jobs

Sizing follows the job the battery has to do, and a per-kWh rule of thumb cannot capture that. A battery is described by two figures that are easy to confuse: power, measured in kW, which is how hard it can push or pull at any instant, and energy, measured in kWh, which is how long it can sustain that. The right ratio between them depends entirely on what you are asking the battery to do. Peak shaving is a power problem first: you need enough kW to flatten the spike and enough kWh to cover how long the spike lasts. Arbitrage and self-consumption are energy problems: you need the capacity to hold a meaningful block of cheap or solar-generated power and release it later. Backup is sized on the critical load you must keep alive and the hours you need to ride through.

That is why we will not put a figure on a battery from a phone call. We size from your half-hourly consumption data and your solar generation profile, read against the specific outcome you are buying. A unit sized for backup may be wrong for arbitrage, and a unit sized to shave a short sharp peak may be far smaller than one sized to shift energy across a whole evening. The method, and the data we ask for, is set out in full on battery sizing.

Bankable means scale, track record, warranty and IEC certification

We are brand-agnostic, and bankable here is an engineering test rather than a marketing tier. Solar modules have a recognised tiering scheme. Batteries do not, so we never describe a BESS as Tier 1, because the phrase belongs to PV modules and means nothing for storage. For a battery, bankable comes down to four hard questions. Is the manufacturer large enough and established enough that the warranty will still be honoured in ten years? Does it have a real track record at the scale you are buying? Is the warranty written in terms you can actually rely on, with a stated cycle life and capacity retention? And is the product certified to the standards that matter, IEC 62619 for the cells and modules and IEC 62933 for the system?

Judged that way, we specify from the makers we supply through UK distribution, led by Sigenergy. Each of them has its own page on the manufacturers we specify, where the directory filters to storage and every maker's ownership, manufacturing and warranty terms are set out in one place. The one we choose follows the project, its scale and your appetite for risk, with the warranty terms confirmed before contract rather than assumed. Where each brand sits relative to its capital is part of the picture on battery costs.

Seven makers supply the storage we specify, and the cabinets have names

A battery on a page is an abstraction. On a site it is a named cabinet from a named maker, with a part number, an enclosure grade, a cooling arrangement and a configuration that either fits your load model or does not. These are the makers whose storage we supply through UK distribution, with the lines we would actually specify and, where we can name them from a dated listing, the cabinet configurations behind them.

• JA Solar. Liquid-cooled commercial and industrial battery cabinet. The cabinet series is BluePlanet C&I ESS. Cabinets listed at 125 kW / 261 kWh.
• SolarEdge. AC-coupled commercial battery cabinet and battery inverter. The cabinet series is Commercial Storage System. Cabinets listed at 50 kW / 102.4 kWh and 50 kW / 204 kWh.
• Solis. All-in-one and packaged battery cabinets on the S6 hybrid platform. The cabinet series we specify are EverCore and PrimePower. Cabinets listed at 50 kW / 100 kWh, 60 kW / 120 kWh, 125 kW / 261 kWh and 50 kW / 109 kWh.
• SolaX. A modular high-voltage stack and a packaged C&I cabinet. The cabinet series is Trene. Cabinets listed at 100 kW / 215 kWh.
• Fox ESS. Hybrids, modular high-voltage batteries and factory-built cabinets. The cabinet series we specify are G-Max and G-Max Plus. Cabinets listed at 100 kW / 215 kWh, 100 kW / 241 kWh and 125 kW / 241 kWh.
• Sigenergy. Modular LFP stacks and a grid-islanding gateway. The series we specify are SigenStack, SigenStor, Sigen Gateway C120 to C1200 and Sigen Energy Controller.
• Sunsynk. Modular high-voltage LFP battery stacks. The series we specify are Sunsynk 5.12 kWh HV rack module and Sunsynk high-voltage battery series.

The configurations named above are read from our UK distributor's commercial cabinet listing as at 15 August 2026. That is one filtered view of one catalogue on one date, not a supply list: a maker with no configuration beside its name is no less procurable, the same maker will build configurations that view did not show, and we re-confirm the part number and the configuration against the current listing and the manufacturer datasheet before either reaches a proposal. What the list is for is to make the choice concrete before the survey. Each maker's ownership, manufacturing footprint, warranty terms and honest placement sit on the manufacturers we specify, and how power, energy and duration decide which of these cabinets suits your load is on battery sizing.

Peak shaving, time-of-use arbitrage, self-consumption and backup: most designs combine them

Four jobs cover most of the load-side value, and the right design usually combines them. Peak shaving targets the half-hours when your demand spikes and your capacity or demand charges are set, holding the metered peak down by discharging through it; we cover the mechanics on peak shaving. Time-of-use arbitrage charges the battery when energy is cheap and discharges it when it is dear, turning a tariff spread into avoided cost, which is the subject of time-of-use arbitrage.

Solar self-consumption is often the quiet workhorse on a site with panels: instead of exporting surplus generation for a low rate, the battery stores it and feeds it back into the building later, so more of your own clean power offsets an expensive import. We set that out on solar self-consumption. Backup power is the job that has nothing to do with saving money and everything to do with continuity, keeping a defined critical load running through an outage, which we cover on backup power. Which of these dominates is a property of your load and tariff, and it is exactly what the survey is there to find.

Four conditions take the storage case apart, all visible in your own data

There are sites where the right storage recommendation is none, and we would rather say that at the modelling stage than sell you a cabinet that sits idle. A battery earns from a gap: between what your array makes and what the building is using at that moment, between a cheap half-hour and an expensive one, between your average demand and your metered peak. Where those gaps are narrow, there is nothing for the asset to work on, and no amount of clever control will manufacture a return the load shape does not contain. Four conditions in particular take the case apart, and all four are visible in your own data before anyone spends anything.

• A flat load that already absorbs the array. A three-shift plant, a cold store or a site with heavy round-the-clock process load draws steadily through the day and the night, so very nearly every unit the panels make is consumed the moment it is made. There is almost no surplus left to store. A battery on that site spends its life charging from the grid at one price to release at another, which is a much thinner trade than the solar is already delivering.
• A tariff with no spread worth arbitraging. Shifting energy in time only pays if the gap between your off-peak and on-peak unit rates is wide enough to cover the round-trip losses and the share of cycle life each cycle consumes. On a flat contract, or one whose spread is thin, the battery works hard for very little. If your capacity and demand charges are modest as well, peak shaving has nothing much to shave, and two of the four value streams are gone at once.
• No connection headroom for the extra capacity. Storage adds connected capacity at the same point as everything else on the site, so the network operator has to accept it. Where the connection is already close to its limit, the answer that comes back can be a reinforcement cost, a restrictive export condition or a wait long enough to undo the business case. That is worth establishing before a battery is designed rather than after, and it is why we look at the connection early; the realities are set out on the grid connection queue.
• Capital that earns more elsewhere. A battery competes with every other call on the same money: more modules on a roof that is not yet full, a compressor or refrigeration plant that is quietly wasting more than the battery would ever save, or holding the cash. Where the storage return is thinner than the next-best use of that capital, storage is the wrong purchase this year even on a site where it would technically work.

None of that makes a battery a poor asset. It makes it a conditional one, and the conditions are testable. We model storage against your half-hourly load and your actual tariff, and where the case is thin we show you the working and say so, including where the sensible answer is to build the solar now and look at storage again when your shift pattern, your tariff or your connection changes. You get the model either way, so the decision is yours to take on your own figures.

Grid revenue is genuine, and modelled rather than promised

This is the part of the value stack where we are most careful with language. Beyond cutting your own costs, a battery can be paid to support the network: providing fast frequency response and other balancing services, holding capacity available for the system operator, or being pooled with other assets into a virtual power plant and dispatched by a third party. We describe each route on grid services, which also covers the aggregator that carries a single site into those markets, and on the Capacity Market.

The income from these routes is modelled and market-dependent, and we never present it as promised. The rates move with auctions and with a fast-changing market, eligibility carries technical and contractual conditions, and a route that pays well this year may not next. So we put no fabricated per-kW-year or per-MWh figure on a page. Where revenue stacking is part of a business case, we model it against current market conditions, flag it as a variable rather than a guarantee, and let your own financial adviser sign off the assumptions. Treated that way it can be a genuine part of the return, and we will not write it up as a fixed promise.

We specify LFP by default, with safety engineered in layers

For commercial storage we specify LFP chemistry as the default. Lithium iron phosphate trades a little energy density for a markedly better thermal and safety profile and a long cycle life, which is the right balance for a fixed asset that has to cycle daily for many years. NMC has its place where space is tight, but for most yard and plant-room installations LFP is the sensible choice, and we explain the trade in full on LFP vs NMC.

Safety is engineered in layers. The cells and modules should meet IEC 62619 and the system IEC 62933, with the battery management and energy-management software watching cell voltages and temperatures continuously and shutting down before a fault propagates; the control layer is covered on EMS software. Around the hardware sit fire detection, separation distances, ventilation and the early-warning monitoring we set out on fire safety. For larger or outdoor deployments a containerised BESS brings its own fire suppression, thermal management and enclosure rating as an engineered package, which simplifies siting on a constrained yard.

The grid connection is usually the longest lead; planning runs beside it

The connection is usually the longest pole in the tent, so we plan it at the start of the project. A storage system that can export to the network needs a G99 connection agreement with the distribution network operator, which across Yorkshire is Northern Powergrid and across Greater Manchester and Lancashire is SP Electricity North West, formerly Electricity North West. The application sets out how much the site can import and export and under what conditions, and the operator's response can shape the size and even the viability of the scheme. On a constrained part of the network the offer can carry a cost or a curtailment condition, and the timescale is measured in months, not weeks. We cover the realities of the queue in the guide on the grid connection queue.

Planning is the other gate. A wall-mounted or indoor battery often sits within permitted development, but a sizeable outdoor or containerised installation can need full planning permission, with fire-service consultation, environmental and acoustic considerations and siting constraints to satisfy. We set out what to expect on planning. Both processes run in parallel with the design, and we sequence them deliberately so the connection offer and any planning condition are known before you commit capital.

Cost is survey-led, and the return is modelled from your own data

We do not publish a from-price, because an honest battery figure cannot come from a brochure. The capital depends on the power and energy you actually need, the coupling and connection, the brand specified, the civils and the connection conditions, and those only become real after a survey. So the cost is survey-led, and the number you receive is yours rather than a headline designed to win a click. How the capital breaks down, and how brand and scale move it, is set out on battery costs and return.

The return is modelled the same way. We take your half-hourly load and your solar profile, apply the jobs the battery will do on your site, and produce a business case grounded in your own data. Any saving, payback or ROI we show is modelled and market-dependent rather than guaranteed, because tariffs and grid-service rates move. The same page sets out how the stacked value is built and how the size you install moves it. One further caveat we will not paper over: the capital-allowances treatment of standalone storage is not settled in public HMRC guidance, so we give you the engineering and the indicative figures and leave the tax position to your accountant.