.png){kind=logo}
Load is growing again for the first time in two decades. Data centers, electrification, heat. The two traditional answers, more generation and more network, both arrive with a queue attached: the median wait from interconnection request to switch-on now runs beyond five years, and only 13% of the capacity that asked to connect between 2000 and 2020 is operating today.
Flexibility at the grid edge does not come with that queue. It is already installed, in driveways and on rooftops. ev.energy's research with The Brattle Group put the value of a single actively managed electric vehicle at up to $575 a year in avoided system cost, before solar, batteries, or bidirectional charging. Coordinate EVs, batteries, and solar on a residential feeder, and the peak falls by up to 44% with no new wires.
So why are utilities sitting on hundreds of MWs that they could unleash to manage demand, reduce rates, and speed up interconnection?
Because a virtual power plant is not a power plant, and the industry keeps procuring it as though it were.
A VPP is tens of thousands of devices belonging to tens of thousands of people who move house, sell cars, end leases, replace chargers, and unplug things without telling anyone. Identical megawatts on the contract, a completely different asset to run.
A VPP is closer to a garden. You do not commission a garden; you cultivate one, and the yield depends less on what you planted than on how well you tend it. Get the conditions right, and a VPP does something no power plant has ever done: it grows itself.
Tending would be far easier with default enrollment, set at the moment a charger, battery, or vehicle is first switched on. It is the largest opportunity available to this industry, and one we would back. But almost no jurisdiction has the machinery yet, and program managers have megawatts to deliver this year under opt-in rules. So three disciplines decide the harvest, and most programs fund the first, assume the second, and ignore the third.
Acquisition is a proposition problem, not a market problem
The household with rooftop solar wants midday charging, not a nudge to midnight. A single price-based offer, run well, reaches around 30% of eligible customers. Every increment past that comes from another offer and another route in: a fleet proposition, a multifamily resident charging solution, an installer handover, the automaker's app. Regulators typically forecast that 80% of eligible customers will enroll. Most programs reach under a third. That gap is a practice gap, not a market failure. Design specific propositions that solve problems for a varied and diverse customer group, and you can leap past 50% enrollment.
Activation is where the largest self-inflicted losses sit
A customer who signs up and never finishes verification is not a dispatchable asset. Ask for a meter number off a paper bill and the worst journeys convert fewer than one in ten. Verify automatically by leveraging implicit data from the customer (e.g. their location) and by preloading the utility's own data instead (e.g. their device territory), and the overwhelming majority activate.
Activation is not just about one device passively managed for one user. Each user can be moved up a flexibility ladder, through nudges that progressively encourage them to do the right thing for the grid and activate more devices. A customer on time-of-use pricing, with an EV responding manually, delivers about 1 kW of peak reduction. Orchestration across charger, solar, battery, and thermostat is 2 to 3 kW. Vehicle-to-home is 3 to 5 kW. Identical enrollment number, five times the dispatchable capacity from bottom rung to top. Programs like Avangrid’s and Clean Power Alliance's are already running this playbook on Eve™.
Retention is the discipline nobody budgets for
Almost nobody actively quits these programs. Devices are another matter: leases end, cars get sold, chargers reach end of life, connections lapse quietly. Every operator should plan for more than 30% device turnover a year. But that is not customer loss. A household that sells one EV and buys another has an open seat, and whether it gets filled is a design decision. Prompt them at the right moment and roughly 80% come back.
Get all three right, and the base starts adding capacity nobody bought: wider through referrals, which now outpace churn across our programs; denser as second and third devices attach to households already paid for; deeper as those households climb the ladder.
Every other asset on a utility's balance sheet does the opposite. A peaker's nameplate is the best it will ever be on the day it is commissioned.
Nameplate megawatts are a claim. Dispatchable megawatts sustained across years is where the value lies. Flexibility being cheaper is the easy part. The bigger takeaway is that a virtual power plant, tended properly, is the only capacity a utility can own that appreciates. Growing in value and size, year over year.
ev.energy builds and runs utility flexibility programs across North America and Europe, from planning through enrollment and dispatch to settlement, on its platform, Eve. Eve Programs is the operating layer where acquisition, activation, and retention run as one discipline. Explore Eve Programs.
