Battery Storage with Manufacturing Solar: When It Pays

Most of our conversations about manufacturing solar reach the same fork. A plant manager has seen the rooftop array numbers stack up, the payback lands in the expected range, and then someone in the room asks the obvious follow up: do we add a battery? The honest answer is that for a large share of UK manufacturers, the battery is the wrong buy. For a specific and growing minority, it is the difference between a good project and a very good one. This guide sets out where the line sits, using the figures we model for real sites.

Start with the load profile, not the battery

Manufacturing PV is sized to daytime baseload, not to roof area. The working rule we apply is to install 70 to 90 percent of peak daytime demand, which maximises self consumption and avoids spilling cheap exports onto the Smart Export Guarantee. When a site follows that rule and runs a daytime heavy operation, almost every kilowatt the panels produce is consumed on site the instant it is generated. There is nothing left over for a battery to store.

That is why our standard position is blunt: for most sites with a daytime heavy load profile, where self consumption already maxes out, you do not need battery storage. Adding one simply parks capital in an asset that spends most of its life half empty. The money buys more return as additional panels, a supply upgrade, or EV charging infrastructure that soaks up midday generation directly.

Battery storage starts to pay above roughly 250 kW of PV, and only when one of three specific conditions applies. Work through them in order.

Condition one: you run meaningful night shifts

Solar generates during the day. If a large slice of your electricity demand falls after the sun sets, rooftop PV alone cannot reach it, and this is the cleanest case for a battery. The array charges the battery through the afternoon when generation exceeds live demand, then discharges it into the night shift when the panels have stopped.

Many manufacturing sites run 24/5 or 24/7 shift patterns. On a pure daytime single shift plant those hours align well with the PV generation curve and a battery adds little. But once a genuine night operation draws steady load in the dark hours, stored solar displaces grid electricity that would otherwise be imported at the full industrial rate. Self consumed solar replaces grid electricity at your full import rate, currently around 18 to 32p per kWh for industrial users, while surplus exported without a battery earns only 4 to 15p per kWh under the Smart Export Guarantee. The gap between those two numbers is the value a battery captures. Shifting a kilowatt hour from a 5p export to a 25p avoided import is a fivefold uplift on that unit.

The catch is that the arithmetic only works if there is genuine surplus to store. A site that already self consumes 85 to 88 percent of its generation on the daytime baseload has very little spare solar to divert into a battery, even with a night shift, because the daytime plant is eating almost everything the panels make. This is why we insist on modelling the load profile shift by shift rather than as an annual average. An annual average hides the exact hours where surplus and shortfall actually sit.

Condition two: heavy DUoS red band charges

Distribution Use of System charges are the network costs baked into your electricity bill, and they are banded by time of day. The red band, covering the early evening weekday peak, carries by far the highest unit rate. For a manufacturer drawing heavy load through that window, red band charges can be a material line on the bill in their own right.

A battery charged from solar during the day and discharged across the red band directly avoids those peak network charges. This is often the strongest financial case of the three, because it stacks with the energy saving rather than replacing it. You avoid the import unit cost and the punitive red band DUoS rate on the same kilowatt hour. For a site whose production runs hard into the early evening, that daily avoidance compounds across every working day of the year.

Whether it pays comes down to how much of your load genuinely sits inside the red band and how heavy your regional DUoS rates are, both of which vary by network operator. We pull at least 12 months of half hourly meter data before any final sizing precisely so the red band exposure shows up in the model rather than being assumed.

Condition three: you want to trade flexibility

The third case is different in kind. Here the battery is not only cutting your own bill, it is earning revenue by selling flexibility back to the grid. Assets can be enrolled in markets such as Dynamic Containment, where National Grid pays for fast frequency response, and other balancing services. A battery sized with headroom can stack behind the meter savings with front of meter flexibility income.

This is the most sophisticated route and the one most sensitive to getting the numbers right. Flexibility revenues move with market conditions, the enrolment and metering requirements are real, and the battery has to be specified and controlled to serve both jobs without cannibalising one for the other. The Energy Systems Catapult is a useful independent reference on how flexibility and system design interact. For a manufacturer with the load and the appetite, flexibility can turn a battery from a cost avoidance measure into a small profit centre, but it should never be the whole business case on its own.

Where a battery does not pay

The mirror image is just as important. If your load profile is daytime heavy, your shifts are single day, your red band exposure is light, and you have no interest in flexibility markets, a battery is close to dead capital. The classic profile here is a food and beverage or pharmaceutical site running refrigeration, chilling or cleanroom HVAC close to 24/7 with a very high, flat baseload. Those sites already achieve exceptional self consumption from the array alone, often in the mid to high eighties as a percentage of generation, which is why their solar payback is among the fastest in manufacturing. There is simply no meaningful surplus for a battery to bank.

In those cases the right move is to spend the battery budget on more panels or on absorbing daytime loads directly. Daytime EV charging is the standout, because it soaks up solar at 100 percent self consumption without any storage losses. Electric forklift fleets, plant vehicles and staff or visitor chargers all stack cleanly with on site PV, and we frequently design the charging infrastructure alongside the array. Every kilowatt hour a forklift charger pulls from the roof at noon is a kilowatt hour that never needed a battery.

Reading the decision at a glance

The table below summarises how the three conditions sit against a plain daytime heavy site. It is a directional guide, not a substitute for modelling your own half hourly data.

Site situation	Does a battery pay?	Why
Daytime heavy, single shift, high flat baseload	Rarely	Self consumption already maxes out; no surplus to store. Spend on more panels or EV charging
Meaningful night shift with genuine daytime surplus	Often	Stored solar displaces night import at 18 to 32p per kWh instead of exporting at 4 to 15p
Heavy DUoS red band exposure in the evening peak	Often	Battery discharge avoids the highest network charge band, stacking with the energy saving
Wants to trade flexibility (Dynamic Containment and balancing)	Sometimes	Adds a revenue stream on top of bill savings, but sensitive to market conditions and specification
PV system below roughly 250 kW	Usually not	Below this scale the storage economics rarely clear the capital cost

How we settle it for your site

We never sell a battery on a hunch, and we never rule one out on a hunch either. The method is the same as for the PV itself: we pull at least 12 months of half hourly meter data and model the load profile shift by shift. That model shows exactly how much solar surplus exists in each hour, where your night load actually falls, how much of your demand sits inside the red band, and therefore whether stored energy earns more than it costs. We model the battery business case alongside the PV so you can see, in your own numbers, whether it pays or whether the money is better spent elsewhere.

There is also a timing angle worth flagging. Where G99 grid connection lead times run long, and 6 to 18 months is now common for installs above 100 kW on constrained networks, we sometimes phase a battery into the design so you get immediate self consumption while waiting for the export agreement to land. In that scenario the battery earns its place partly as a grid connection workaround, letting the array deliver value on day one rather than sitting throttled.

If you are weighing storage against simply installing more panels, the full picture lives in our manufacturing solar cost breakdown, and the funding routes that shape the capital case sit on our grants and funding page. Sites with the strongest night shift storage case tend to be in automotive manufacturing, where paint shop, weld shop and compressed air loads run across shifts. When you are ready, send us your meter data through the quote page and we will model the battery case honestly, both ways.

The short version: a battery is a precision tool for night load, red band exposure and flexibility trading, not a default upgrade. For a daytime heavy manufacturer, the panels do the work and the battery just watches. Model it before you buy it, and let the numbers, not the sales pitch, decide.