Ten Thousand Home Batteries Behaving Like One Power Station
Aggregating household batteries, water heaters, and thermostats produces a resource a grid operator can dispatch. The technology works and the difficulty is entirely in the contracts and the market rules.
The Grid Problem It Addresses
Electricity supply must match demand instantaneously. Grid operators traditionally manage that by dispatching generators, and the expensive part is the small number of hours each year when demand peaks and the least efficient plants must run.
A substantial amount of flexibility already exists on the other side of the meter. Home batteries can discharge, electric water heaters can shift when they heat, thermostats can precool a house before a peak, and electric vehicle chargers can pause.
Individually each is far too small to matter. A virtual power plant aggregates thousands of them into a single controllable resource that a grid operator can call on.
What It Actually Sells
The aggregator does not sell electricity in the ordinary sense. It sells several distinct products, and understanding which one matters commercially.
| Product | What Is Delivered | |
|---|---|---|
| Capacity | A commitment to be available at peak | |
| Energy | Actual discharge or reduction during an event | |
| Ancillary services | Fast response to frequency deviations | |
| Distribution deferral | Avoiding a local network upgrade |
The ancillary services category is where distributed resources have a genuine technical advantage. A battery responds in milliseconds, far faster than a thermal generator, which makes it better suited to frequency regulation than the machines it competes against.
The value is not in the electricity, which is a small quantity. It is in the speed and the certainty of the response, and in avoiding the physical infrastructure that would otherwise be built for a handful of hours a year.
The Regulatory Prerequisite
None of this works unless wholesale market rules permit aggregated small resources to participate, and for a long time they did not. Markets were designed around generators measured in megawatts with defined minimum bid sizes, telemetry requirements, and dispatch protocols.
A regulatory order in the United States in 2020 required organised wholesale markets to establish participation models for distributed energy resource aggregations, removing minimum size barriers that had excluded them.
Implementation has been slow and uneven across markets, and the details matter enormously: minimum aggregation size, metering and telemetry requirements, how baseline consumption is measured for demand reduction, and whether a resource can participate in a wholesale market and a utility programme simultaneously.
That last question, described as dual participation, determines whether the same battery can earn from more than one revenue stream, which frequently decides whether the economics work at all.
The Behavioural Problem
The technical challenge is smaller than the human one. A grid operator dispatching a generator knows it will respond. An aggregator dispatching ten thousand households is relying on devices remaining connected, customers not overriding the control, and participation not decaying over time.
Programmes address this with contractual commitments, override limits, and payment structures rewarding availability rather than only performance. Attrition is a real operating metric, and customers who override repeatedly during hot afternoons are exactly the ones the programme needed.
Battery based programmes are considerably more reliable than thermostat based ones, because a battery discharging is invisible to the household while a thermostat adjustment is not.
The Business Models
Three approaches have emerged and they differ in who owns the hardware.
Utility owned programmes have the utility install and control devices, recovering the cost in rates. Simple and slow.
Customer owned with an aggregator has households buy their own equipment and enrol, receiving a payment for participation. The aggregator has no hardware capital at risk and is exposed to enrolment and retention.
Aggregator owned has the company install equipment at no upfront cost to the household in exchange for control rights and the resulting market revenue, which is capital intensive and captures more of the value.
The second model has scaled fastest because it requires the least capital, and it produces the weakest control over the resource, which is the recurring tension.
Why It Matters More Now
Two developments have made the concept more valuable than it was.
The generation mix has shifted toward resources whose output depends on weather, which increases the need for flexibility that can be called at short notice.
And connecting new generation has become slow, with interconnection queues extending for years, while transmission and distribution upgrades face permitting delays. A resource assembled from equipment already installed behind meters requires no new interconnection at all, which is a substantial practical advantage over building anything.
How to Read the Sector
For anyone assessing one of these businesses, the questions are about durability rather than technology. How many revenue streams the resource can access, since single stream economics are usually marginal. Contract length and termination rights with participating households. Attrition rates and override frequency. And regulatory exposure, since the entire revenue base depends on market rules that are still being written and can change.
The Bottom Line
A virtual power plant turns flexibility that already exists behind household meters into a resource a grid operator can dispatch, and its advantage over building generation is that nothing needs to be built or connected. The technology is not the constraint. The constraints are whether market rules let small aggregated resources bid, whether a resource can earn from more than one programme, and whether households keep participating on the afternoons when it matters.