One Battery Earning From Four Markets at Once
A grid scale battery can charge cheaply and discharge expensively, provide frequency services, hold capacity, and defer network investment. No single one of those pays for it, and the combination might.
An Asset With No Single Business
A grid scale battery has an obvious use: buy electricity when it is cheap, store it, and sell when it is expensive. That is energy arbitrage, and on its own it rarely justifies the capital cost.
The price spread between the cheapest and most expensive hours in a day, multiplied by the number of cycles a battery can perform, generally produces a return well below what the equipment costs, particularly once degradation is accounted for.
So batteries are financed on revenue stacking, meaning the combination of several distinct products the same asset can supply.
The Products
| Revenue Stream | What It Pays For | Energy Throughput Required |
|---|---|---|
| Energy arbitrage | Price spread across hours | High |
| Frequency regulation | Fast response to imbalance | Very low, net near zero |
| Capacity or reliability | Being available at peak | Low, few events |
| Network deferral | Avoiding a substation upgrade | Low, targeted hours |
The third column is the key to why stacking works. Frequency regulation requires the battery to move small amounts of energy in both directions constantly, with net throughput near zero, which means it barely consumes cycles.
Capacity payments require availability rather than operation. A battery can hold capacity obligations and be called only a handful of times a year.
That leaves substantial capability unused, which can be sold as energy arbitrage.
Stacking works because the products consume different things. One pays for speed, one pays for availability, and one pays for throughput. A battery has all three and no single market pays for more than one of them.
Why It Is an Optimisation Problem
The constraint is that commitments conflict. A battery holding a capacity obligation must maintain enough state of charge to deliver if called, which limits how deeply it can cycle for arbitrage. A battery committed to frequency regulation must hold headroom in both directions.
So the operator solves a scheduling problem across markets with different timescales, different commitment periods, and financial penalties for failing to deliver on a commitment.
This is done with optimisation software rather than judgement, and the quality of that software is a genuine differentiator between operators of physically identical assets.
The Saturation Problem
The recurring commercial risk is that the highest value markets are small.
Frequency regulation is the clearest case. A power system requires a limited quantity of it, measured in a few hundred megawatts even in large markets. Batteries are extremely well suited to it and entered aggressively, and prices in several markets collapsed as capacity exceeded requirement.
Projects financed on frequency regulation revenue found that the market they were built for saturated within a few years, forcing a shift toward arbitrage and capacity at lower returns.
The general lesson is that a small high value market cannot support a large asset class, and any business plan resting on it should assume the price falls as competitors arrive.
Degradation Is a Real Cost
Lithium ion batteries lose capacity with cycling, and the rate depends on depth of discharge, temperature, and charge rate.
Every arbitrage cycle therefore consumes an asset with a finite number of cycles, which means the correct comparison is not the price spread against zero but the price spread against the marginal degradation cost of the cycle.
An operator that cycles aggressively into a modest spread is converting asset life into revenue at a loss, and this is why sophisticated dispatch models include a degradation cost term that raises the threshold spread required to trade.
Warranty terms typically guarantee capacity retention subject to operating limits, which constrains how the asset may be used and is a genuine contractual restriction on the revenue strategy.
The Duration Question
Batteries are specified by power and by duration, meaning how many hours they can discharge at full output. A two hour battery and a four hour battery with the same power rating are different assets.
Shorter duration suits frequency response and short peaks. Longer duration is required to shift substantial energy across a day and to qualify for capacity credit in markets that require sustained output.
As systems add more variable generation, the value shifts toward longer duration, and market rules requiring longer sustained delivery for capacity accreditation directly determine what gets built.
The Bottom Line
A grid battery is financed by layering several revenue streams that pay for different physical properties, because no single market pays enough to cover the asset. Stacking is an optimisation against conflicting commitments rather than a simple sum, degradation makes every arbitrage cycle a real cost, and the highest value markets are small enough to saturate quickly. Anyone evaluating one should ask which revenue stream dominates the model and what happens when the next hundred megawatts arrive in that same market.