Grid services, and the aggregator route into them

The system operator and the network operators pay for fast, controllable power to keep the grid balanced and at frequency. A battery is well suited to providing it, because it can charge and discharge in a fraction of a second. Access these markets and your battery can earn on top of the import cost it already avoids.

On its own, though, a single commercial battery is too small to get in. These markets are built for fleets, so the practical route is an aggregator, which pools your asset with many others into a virtual power plant and bids the fleet as one. This page covers both halves: the services a battery sells, and the intermediary that reaches them, including the contract terms that decide what the arrangement is actually worth.

We treat grid-services income as upside, never as the reason to buy. The on-site case, lifting solar self-consumption and shaving expensive peaks, has to stand on its own first. If the flexibility revenue then adds to it, that is a bonus you can plan around, not a number we will dress up as fixed.

Grid services are the jobs a battery does for the wider network in return for payment. The two broad families are frequency response, where the battery automatically charges or discharges within seconds to hold the grid at its target frequency, and demand-side flexibility, where you agree to shift or reduce your import at set times when the system is tight.

Fast frequency response is the headline product here. It rewards a quick, accurate response, and a battery can deliver it without you lifting a finger, because the control happens automatically. Demand-side flexibility schemes ask for a planned change in your draw from the grid, which a battery makes painless because it can cover the load instead of you switching equipment off.

Most commercial sites reach these markets through an aggregator. An aggregator pools many batteries and flexible sites into one fleet, often called a virtual power plant, that behaves to the grid like a single large controllable asset. That pooled scale is what gets you through the door, because the markets are designed around fleets that can deliver a guaranteed volume of power on command and a single commercial battery rarely clears the minimum alone. The aggregator registers and qualifies your battery against each market's technical rules, runs the telemetry and metering the system operator demands, places the bids, receives the dispatch instructions, controls your battery's charge and discharge to honour them within the limits your contract sets, then settles the income and passes you a share.

The second half of what a good aggregator does is stacking. No single market pays the most at every hour. Frequency response is valuable when the grid is volatile; the Balancing Mechanism pays when the system operator needs fast power to balance supply and demand; the Capacity Market pays a steadier amount for being available at peak; wholesale arbitrage rewards charging cheap and discharging dear. A capable aggregator co-optimises across all of them, moving the battery to whichever is worth most at that moment while holding back the capacity it has already promised elsewhere, and working around the on-site jobs you have set to come first. Two batteries of the same size, one with a sharp aggregator and one with a passive one, can earn very differently, which is why the route matters as much as the markets.

Qualification is a real gate rather than a formality. Your battery, its metering and its grid connection all have to meet the market's technical requirements, and the terms differ by product and by aggregator. We design the system so it can qualify where that makes sense, and we are honest where a site's connection or profile makes a given market a poor fit.

• The aggregator qualifies, bids and settles; you receive a share of what the battery earns.
• Entry depends on your battery spec, metering, telemetry and grid connection passing the market rules.
• Contract terms, the revenue split and notice periods vary by aggregator, so they are worth comparing.

Flexibility and balancing markets are competitive and they move. What a battery earns depends on the product it qualifies for, how often it is called, the prices clearing at the time and the share your aggregator keeps. None of that is fixed, so we will not quote you a per-megawatt figure as if it were. Anyone who does is guessing.

What we can say plainly is how it stacks. The first and most reliable return is on-site: stored solar offsets expensive import, and export is paid far less, so self-consumption is where the steady money is. Peak shaving trims your worst demand charges. Grid services then sit on top, as a variable third layer. Build the case on the first two and treat the third as upside you would welcome but have not banked.

It helps to name the markets rather than talk of grid services in the abstract, because each is run by a different body and each has its own door. The first family is frequency response, procured by the National Energy System Operator (NESO, the former National Grid ESO) through its suite of dynamic services. NESO defines three products that a fast battery is built for: Dynamic Containment, which arrests a sharp frequency deviation after a sudden loss of generation; Dynamic Moderation, which acts pre-fault to keep frequency inside a tighter band; and Dynamic Regulation, which corrects the slow, continuous drift around the 50 Hz target. A battery can pre-qualify for one or more of these and switch between them as their value changes through the day.

The second family is the Balancing Mechanism, the near-real-time tool NESO uses to match supply and demand minute by minute. Through it, a battery can submit bids and offers to turn up or down on instruction. Historically this was the preserve of large generators, but reforms under the Wider Access programme opened it to aggregated, smaller assets, which is how a single commercial battery can now reach it through a pool. Alongside both sits the Capacity Market, an availability payment secured at auction, and wholesale time-of-use arbitrage, which is a trading job rather than a grid service. A well-run battery is moved between these by its control system and aggregator to chase the most valuable opportunity at any moment, a practice the market calls revenue stacking. We name each mechanism so you can see what is on offer, but we will not attach a clearing price to any of them, because those prices are set in competitive auctions and live markets that move from one settlement period to the next.

Qualification is the gate that decides whether any of these markets is even open to your site, and it is governed by published rules rather than goodwill. NESO's dynamic services require the asset to meet defined technical and metering standards before it can pre-qualify, including a controllable response that tracks the frequency signal accurately and within the product's reaction and delivery times. The grid connection itself sits under the Distribution Code and Engineering Recommendation G99, which is the framework your network operator uses to agree how a battery may import and export. Settlement of any market income depends on metering that meets the Balancing and Settlement Code, typically half-hourly metering that the market can read and trust.

The battery and its inverter have to carry the right certification and behave predictably under the relevant standards, the same IEC 62619 cell-level safety standard and IEC 62933 system-level standard we hold the equipment to for the rest of the build. Aggregators will not enrol an asset that cannot evidence its response, so the telemetry and metering are as much a part of qualification as the cells. We design the system from the start so it can clear these gates where it makes sense to, and we are straight with you where a site's import or export limit, its connection agreement or its load profile makes a given market a poor fit. None of this is a formality you can assume your way past, which is exactly why the income has to be modelled rather than promised. The practical route through pre-qualification, bidding and settlement is almost always an aggregator, which carries the registration and dispatch burden a single site could not justify alone, and whose contract is covered further down this page.

Earning from the grid is not free of consequence for the battery, and an honest page has to say so. Frequency response and Balancing Mechanism activity add charge and discharge cycles beyond the predictable on-site work, and every cycle counts towards the gradual capacity fade that every lithium battery experiences over its life. That is one of the central reasons we treat grid-services income as upside rather than the core case: the steady on-site jobs are sized first, and market participation is layered on only where it fits inside the maker's warranted cycle and throughput limits.

Manufacturers warrant their commercial products against a stated number of full-equivalent cycles or a total energy throughput, and aggressive market cycling can move a battery towards those limits faster than a self-consumption-only duty would. We read the specific warranty and operating envelope for the chosen product from its current datasheet before committing it to any market, and we set the control rules so dispatch stays inside that envelope. Where a lucrative product would push the asset past its warranted duty or compromise the on-site savings, we leave it out, because a voided warranty is a far bigger number than a marginal frequency-response payment. This is also why we size the battery from your half-hourly load for the on-site case first; the model has to demonstrate the durability headroom for grid services before we count on it. The chemistry that makes this manageable, and the safety standards behind it, are covered on our peak shaving and fire-safety pages.

Not every commercial site that benefits from a battery is a strong grid-services prospect, and it is worth knowing which side of that line you fall before the topic shapes your business case. Three things weigh most heavily. The first is your grid connection: the import and export capacity in your connection agreement sets a hard ceiling on how much power the battery can offer to a market, and a constrained connection narrows what you can earn regardless of the battery's nameplate. The second is your metering: settlement-grade half-hourly metering under the Balancing and Settlement Code is a prerequisite for most market income, so a site still on a non-half-hourly supply has a step to take first. The third is duration and headroom: a battery sized tightly for solar self-consumption may have little spare capacity to dedicate to the grid without eating into the on-site jobs that pay most reliably.

This is why we work the question from the survey outwards rather than from a brochure inwards. An in-house insured drone survey and a half-hourly PV*SOL model show us your real load shape, your connection limits and the spare cycles a battery would have once the on-site work is covered. From that we can tell you honestly whether a grid-services layer is worth chasing for your site or whether the case stands perfectly well on solar self-consumption and peak shaving alone. How all of this folds into a single modelled return, with the markets fed in as variable layers rather than banked income, is set out on our battery costs and return page.

Almost all aggregator agreements are revenue-share. The aggregator takes a percentage of what your battery earns and passes the rest to you. That headline percentage is the obvious term, and several others decide what the deal is really worth. Read them before you sign, and compare more than one aggregator, because the terms vary widely and the market is competitive.

• The split, and what it is taken on. Is the share calculated before or after the aggregator's costs, and is it the same across all markets or only on the headline product?
• Term, exclusivity and notice. How long are you tied in, are you locked to one aggregator, and how much notice does it take to leave if a better route appears?
• Dispatch rights and cycling caps. How much of the battery's capacity can the aggregator call, how often, and is there a cap that keeps cycling inside the maker's warranty?
• Penalty exposure. If a called event is missed, who carries the penalty, you or the aggregator?
• Any floor. Some contracts offer a guaranteed minimum. Weigh it as a commercial term, not as proof the income is safe, because a floor is only as good as the party promising it.

Judge the aggregator itself as carefully as the hardware. Is the company licensed and active in the markets it claims, with a settlement track record under the Balancing and Settlement Code and a credible balance sheet behind any guarantee it offers? Ask which markets it actually trades for assets like yours rather than which it lists, and ask for the basis of any modelled figure it quotes, so you can see whether it rests on recent clearing prices or on an optimistic forward view. A number that arrives without a disclosed basis is marketing.

Then read the exit. The Capacity Market commits an asset across a defined delivery year under the Capacity Market Rules, so a battery promised into it is tied for that period regardless of the notice period in your wider agreement. Frequency and balancing participation is usually more flexible, but the contract still sets the term, the exclusivity and the notice to leave. The practical questions are whether you can switch aggregator if a sharper route appears, what happens to in-flight market commitments when you do, and whether the optimisation runs through control software you retain or software locked to that provider. Why keeping your own control layer protects the on-site jobs is covered on EMS software. None of this changes the honest position on the money: the income rests on auction outcomes, dispatch frequency, clearing prices and the share the aggregator keeps, so we build the case on the far steadier on-site savings and treat the market income as upside you would welcome but have not banked.