Grid Flexibility
The cheapest storage is a demand that can wait. Most of what we do with electricity does not actually need to happen right now.
The effect compounds within years. Put it in place and it keeps working.
Origins
For a century, the grid worked one way: demand was sacred and supply obeyed it.
Whenever you flicked a switch, somewhere a generator was ramped up to meet you. Your convenience was the fixed point of the entire system, and an enormous, expensive machine was built around the assumption that it must never be questioned. Power plants sat idle for most of the year, kept warm and paid handsomely, purely to be available for a handful of cold evenings when everybody turned on a heater at once.
This was not stupid. It was the correct design given the technology of 1920, and it made electricity feel like a law of nature rather than a service.
But it has been quietly wasteful the entire time, and the numbers are almost comic: a large share of grid capacity exists for a tiny fraction of the hours in a year. We built a whole second electricity system that spends most of its life switched off, waiting.
The realisation, arriving properly only in the last decade or so, is that demand does not actually have to be sacred. A great many things we do with electricity are indifferent to timing. A water heater does not care whether it heats at 3am or 6pm, provided there is hot water in the morning. A fridge can drift a degree. An electric car sitting on a driveway for fourteen hours does not care which four of them it charges in.
The cheapest power plant is the one you do not build because the demand moved.
What it actually is
Grid flexibility is the practice of making demand follow supply, instead of the other way around — and, where necessary, moving power around the system rather than storing it.
Demand response. Shift consumption to when power is abundant and cheap. Water heaters, EV charging, industrial processes, air conditioning pre-cooling, data centre workloads: all can move, and most of it is invisible to the user. This is the single largest untapped resource in the electricity system and it costs almost nothing, because the assets already exist.
Transmission. The wind is blowing hard somewhere; it is calm here. A big, well-connected grid smooths both supply and demand across a continent, and it is the cheapest form of storage there is — you are not storing energy, you are borrowing somebody else’s weather.
Smart charging and vehicle-to-grid. Every electric car is a battery on wheels, parked and idle roughly 95% of the time. A national EV fleet is, in aggregate, a colossal storage resource that has already been paid for by somebody else.
Forecasting. Better weather prediction means a grid operator can plan for the wind dropping in six hours rather than reacting when it does. Boring, cheap, extremely effective.
None of this requires new hardware. It requires information, incentives and market rules — which is why it is the cheapest thing on this list and why it is so consistently neglected.
The numbers
The idle capacity. A substantial share of grid infrastructure exists to serve peak demand that occurs in a very small number of hours per year. Flattening the peak avoids building generation, transmission and distribution that would otherwise sit idle for the overwhelming majority of its life.
The car fleet. Private cars are parked roughly 95% of the time. An electrified fleet is a distributed battery of extraordinary size, already purchased, already connected, and currently doing nothing while it sits.
The cheapest resource. Demand flexibility requires no new generation, no land, no mining and no construction. It is, per unit of value delivered, the cheapest thing available to a grid operator, and it is systematically underused because market rules were written for a system where demand could not respond.
Transmission as storage. Interconnecting regions with different weather is functionally equivalent to storage and is frequently cheaper. Europe’s interconnectors let Danish wind meet Norwegian hydro and German demand, and the whole system needs less storage than any one country would alone.
The constraint is not technical. Every element of this exists and works. The barrier is market design, tariff structure and regulatory inertia — which means the cost of not doing it is entirely self-inflicted.
Why it matters
This is the least glamorous solution on this entire website and it may be the most quietly infuriating, because it is essentially free and we are mostly not doing it.
There is no invention required. No mine. No land. No new material. Nothing has to be built. The wires, the water heaters, the fridges, the cars, the factories all already exist. What is missing is a price signal that tells them when the electricity is abundant and when it is scarce, and a set of market rules that lets them respond.
We are, right now, keeping enormous fossil-fuelled power plants on standby, burning fuel to stay warm, for a handful of hours a year — while millions of water heaters sit heating water at exactly the wrong moment, because nobody ever told them otherwise.
There is something almost embarrassingly simple about it. Your grandmother did the washing when the electricity was cheap, because she had an off-peak meter and she paid attention, and she was doing sophisticated grid balancing without ever hearing the phrase. We took that away, made everything flat-rate and invisible, and are now spending billions to rebuild the intelligence we deliberately removed.
The cheapest power station is the one nobody had to build. That is not an environmental argument. It is a plain statement about money, and it should be the easiest sell in the entire climate portfolio.
What it actually takes
Market rules, which is where this lives and dies. Most electricity markets were designed when demand could not respond and storage did not exist. They pay for megawatts, not for flexibility, and a resource that cannot be paid does not appear. This is the entire bottleneck and it is not technical.
Price signals people can actually see. Flat-rate electricity tariffs are a decision to hide the truth from consumers. Time-of-use pricing, done fairly, lets households and businesses respond, and most of them will — particularly if the response is automated and they never have to think about it.
Automation, not virtue. The winning design does not ask people to change their behaviour. It asks their water heater to. A thermostat, a car charger and a hot water cylinder that quietly optimise in the background are far more effective than any amount of public exhortation.
Protecting the vulnerable. Time-of-use pricing can hurt people who cannot shift their demand — the elderly, the sick, those on medical equipment, shift workers. This is a real risk and it is manageable through tariff design, but it must be designed in rather than apologised for afterwards.
Transmission, which everyone hates. Nobody wants a power line, and lines are harder to permit than the generation they carry. This is the most consistent physical bottleneck in the whole energy transition.
Where it matters most
Europe is the world’s most interconnected grid and the proof that transmission substitutes for storage. Danish wind, Norwegian hydro, French nuclear and German demand balance each other across borders, and the whole is far more stable than any part.
China has built ultra-high-voltage transmission across thousands of kilometres, moving power from windy, sunny, empty western provinces to the crowded east. It is the largest deliberate act of grid engineering in history.
California and the Southwest have the sharpest duck curve on Earth and are therefore the most advanced laboratory for demand response and time-of-use pricing.
Australia, with the world’s highest rooftop solar penetration, is discovering the opposite problem: too much power at midday, and a grid that must learn to absorb rather than merely deliver.
The East African Rift and off-grid regions are building minigrids from scratch, which means they can build them flexible from the start rather than retrofitting intelligence onto a century-old machine. They may end up with better grids than we have.
How to tell it’s being done well
Can flexibility get paid? If the market only pays for megawatts, flexibility does not exist as far as the system is concerned. This is the whole question.
Is it automated? Programmes that rely on people paying attention fail. Programmes that quietly optimise the water heater succeed.
Who bears the risk? Time-of-use pricing can penalise people who cannot shift, including the sick and the elderly. Good design protects them. Bad design calls it their problem.
Is the transmission being built? Flexibility and interconnection are substitutes for storage, and both are cheaper. Neither gets built without political will.
What you can do
Anyone
- If your utility offers time-of-use pricing and you can shift laundry, dishwashing and car charging, you will pay less and help the grid at the same time.
- Your grandmother had an off-peak meter and did the washing at night. She was doing grid balancing and nobody called it that.
Homeowners
- A smart EV charger that fills the car when power is cheap and abundant is the single highest-value flexibility device most households can install.
- Hot water is a battery. A timed cylinder heats when electricity is cheap and holds it for when you need it.
Policymakers
- Reform the market rules. They were written when demand could not respond, they pay for megawatts rather than flexibility, and that single fact is the largest self-inflicted cost in the electricity system.
- Time-of-use pricing, designed to protect people who cannot shift.
- Build the transmission. It is the most consistent physical bottleneck in the entire energy transition and the least popular thing to permit.
Business and investors
- Aggregating distributed flexibility, water heaters, EV chargers and industrial loads, into a dispatchable resource is a real and growing business, and the assets are already installed and paid for.
- Data centres are enormous, flexible, controllable loads. Most are not being used as such, which is a substantial missed opportunity.
Who is working on this
We are researching which organizations in our directory of 8,493 actively work on this solution, and we only list an organization once we have verified it. That research is ongoing. In the meantime, search the directory yourself:
Questions
What is grid flexibility?
Making demand follow supply instead of the other way around. Shifting consumption to when power is abundant, moving power across regions with different weather, and using assets that already exist, water heaters, EV chargers, industrial processes, as a resource. It requires almost no new hardware, only information, incentives and market rules.
Why is it the cheapest solution?
Because nothing has to be built. No generation, no land, no mining, no construction. The wires, water heaters, fridges and cars already exist. What is missing is a price signal telling them when electricity is abundant and a market rule letting them respond.
How much grid capacity is wasted?
A substantial share of generation and network infrastructure exists solely to serve peak demand occurring in a very small number of hours per year. We built an entire second electricity system that spends most of its life switched off, waiting for a cold evening.
Can electric cars really help the grid?
Yes, and it is one of the largest untapped resources anywhere. Private cars are parked roughly 95% of the time. An electrified fleet is a colossal distributed battery that has already been purchased by somebody else and is currently doing nothing while it sits on a driveway.
Doesn't time-of-use pricing hurt poor people?
It can, and this must be designed for rather than apologised for. People who cannot shift their demand, the elderly, the sick, those on medical equipment, shift workers, can be penalised by naive time-of-use tariffs. It is manageable through tariff design, but it has to be built in from the start.
Why isn't this already happening?
Market rules. Most electricity markets were designed when demand could not respond and storage did not exist, so they pay for megawatts rather than for flexibility. A resource that cannot be paid does not appear on anybody's balance sheet. The barrier is entirely regulatory, which means the cost of not doing it is self-inflicted.
Sources
- Project Drawdown - Improve Grid Flexibility (Drawdown Explorer) Framework and classification. Cited, not reproduced.
- IEA - Grid-scale storage and system flexibility
- IRENA (2026) - 24/7 Renewables: The economics of firm solar and wind
- IPCC (2022), AR6 Working Group III - Energy Systems
- Regulatory Assistance Project - power market design
The solution taxonomy follows the framework popularised by Project Drawdown. The analysis above is our own; for their carbon modeling and rankings, visit them directly.