A solar farm sells electrons. A battery sells timing. It earns by buying power when it is cheap, selling it when it is dear, and getting paid to keep the grid steady in between. That makes it enormously useful and, for a lender, deeply uncomfortable. Revenues swing with the weather, the gas price, the number of rival batteries and the skill of whoever is pressing the buttons.
In Texas, Modo Energy’s ERCOT battery index fell from about $46,300 per MW-year in January 2026 to $15,300 in February, then recovered to $38,100 in April. Try sizing a ten-year loan against that.
The volatility hasn’t gone away. What has changed is that data and financial engineering now let investors measure it, price it and pass it to someone better placed to hold it.
A benchmark the market can agree on
Every investable asset class needs a shared yardstick. Until recently, batteries had a patchwork of private spreadsheets, so every negotiation began with the question of whose numbers to trust.
Great Britain now has an answer. Modo publishes FCA-regulated battery revenue benchmarks, tracking what operational batteries actually earn across wholesale trading, balancing, and ancillary services. In September 2026, SSE and Statkraft completed the first financial swap settled against one of them, the ME BESS GB two-hour index, under ISDA terms: Statkraft took the floating index, and SSE locked in fixed income on part of its fleet.
The data is what makes the deal work. Over the year to August 2026, the index averaged £69k/MW/year, roughly 50% more than a simple day-ahead price spread implies. Across 17 GB batteries, an index swap cut median monthly revenue volatility from £22k to £12k/MW/year. And because battery revenue tends to rise when wind and solar revenue falls, a 1,414 MW index swap could cut the monthly revenue volatility of a 500 MW offshore wind farm by 47%.
A swap needs no physical asset, no dispatch rights and no availability terms. It is pure financial plumbing, and it exists only because the underlying data is trusted enough to settle money against.
From merchant risk to contracted income
If benchmarks let investors measure battery revenue, contracts let them transfer it. Britain has moved from Capacity Market contracts (around 2018) to revenue floors (2021), portfolio tolls of 200 MW or more (2024) and the first index swap (2026). Each step shifted risk away from the owner and made the asset easier to lend against.
Great Britain: tolling goes long.
In a toll, a trader or utility pays the owner a fixed fee for full control of the asset; the owner gets infrastructure-style income, and the toller keeps the trading upside. Octopus Energy and Gresham House signed Britain’s first in June 2024, two years over 568 MW. Terms have stretched fast since. BW ESS raised debt from Westpac and UOB for its 100 MW Bramley battery on a seven-year toll with Shell Energy Europe. Drax signed a 10-year, CPI-indexed toll for Fidra Energy’s 250 MW West Burton project, then in February 2026 a 15-year toll for Zenobe’s 200 MW Coalburn battery in Scotland, giving it 710 MW of tolled and owned capacity without upfront capital. A 15-year fixed fee is, in effect, a bond-like cash flow a lender can size debt against.
Germany: subsidy-free and bankable.
Germany has gone from merchant-only to tolled in barely two years. Nofar Energy’s 104.5 MW Stendal battery raised €86.5 million from NORD/LB against a seven-year toll, described by its lawyers as the first of its kind in continental Europe. Eco Stor followed with NORD/LB financing for its 100 MW Schuby project, backed by a five-year toll with Alpiq. At Eco Stor’s 300 MW Forderstedt site, Germany’s largest, Next Kraftwerke has tolled 80% of capacity for five years, leaving 20% merchant, privately financed, with no public subsidy. Partial tolls like this let owners keep some upside while giving lenders a contracted floor.
Australia: tolls, government floors and portfolio debt.
The federal Capacity Investment Scheme underwrites project revenues between floor and ceiling prices through 10–15-year agreements. Private offtake is maturing alongside it: Akaysha Energy’s 415 MW Orana battery has 200 MW contracted to EnergyAustralia under a 12-year virtual toll. In October 2026, Akaysha rolled the debt on Orana and three other batteries into a single A$1.75bn portfolio facility backed by 11 banks. Moving from single-asset loans to portfolio debt shows lenders now trust revenue data across a fleet.
Underwriting with evidence, not optimism
Granular market data has changed how battery debt is sized. In Texas, Modo advises basing a project’s case on the trailing-year average, about $28,800/MW-year, rather than any headline month. The same data shows two-hour batteries out-earning one-hour systems in every month for two years, by 15% to 81%, which now feeds directly into how projects are designed and valued.
Data is shaping who lends, too. In Q1 2026, ERCOT accounted for 10 of 17 tracked US battery financings, led not by American banks but by MUFG (821 MW across four deals) and Société Générale (713 MW across three). Modo concludes that bankable Texas projects increasingly need offtake or tolling support. Meanwhile, 3.4 GW of projects withdrew after signing interconnection agreements in 2025–26, against 937 MW in all prior years combined; capital is concentrating on projects whose numbers stand up. Data doesn’t just help good projects raise money; it helps lenders say no to weak ones before they fail.
Insuring the algorithm
The most fintech-flavoured innovation may be insurance on software. In ERCOT, New Energy Risk and Ascend Analytics closed what they called an industry-first policy covering the performance of Ascend’s battery forecasting and bidding platform, enabling financing for a portfolio of grid-scale batteries. Unlike a toll, the cover guarantees minimum revenues while letting projects keep the upside from Texas’s price spikes, valuable in a market where a handful of volatile days can make a battery’s year. The insurer is, in effect, underwriting an algorithm’s track record, which is only possible because years of dispatch and price data now exist to test it against.
Optimisation platforms are becoming strategic assets in their own right. Drax bought Flexitricity in January 2026 before signing its tolls, securing the trading capability to run the batteries it pays for. Next Kraftwerke presents Forderstedt as proof that algorithms and market integration can deliver bankable revenue. Whoever has the best data and algorithms can afford to offer owners the most certain income, and that certainty is what unlocks the debt.
The fine print
None of these tools makes risk vanish; they move it. An index swap leaves basis risk with the owner: a battery that underperforms the benchmark still ends up below its fixed rate. Long tolls trade upside for certainty, and their value depends on the toller’s creditworthiness over 10 or 15 years. And no contract can fix a project that cannot connect: in Great Britain, connections reform has pushed 6.5 GW of Modo’s tracked battery pipeline back by at least six months, with only 37% holding a Gate 2 date.
Data does not remove uncertainty. It describes it precisely enough for someone to price it.
From volatile to investable
A decade ago, a grid battery was a technology bet. Today it is becoming an infrastructure asset class, for reasons as much financial as technical. Regulated benchmarks give investors a shared number. Tolls, floors and swaps turn that number into contracted income. Insurance puts a price on algorithmic performance. And granular revenue data lets lenders size debt on evidence rather than hope.
From a 15-year toll in Scotland to an insured revenue floor in Texas and an A$1.75bn facility in Australia, the lesson is the same. Capital flows to what it can measure. Data has made batteries measurable, and the money is following.
