2030: A Million UK Driveways, One Distributed Power Grid

A VPP is a software layer that aggregates thousands of small assets and presents them to the grid as a single dispatchable resource. National Grid ESO doesn't care whether 1 MW of frequency response comes from a gas turbine in Pembroke or 4,200 home batteries in Surrey — provided the response time, accuracy, and reliability meet the technical specification.

The economic prize is that a 1 MW gas plant providing dynamic containment earns roughly £130,000/year. The same response from aggregated home batteries earns £150,000-£180,000 (no fuel cost, faster response), of which the operator keeps a margin and passes the rest to participating households.

Every driveway.solar install runs an edge controller (ARM Cortex-M7, 1 GB RAM) executing local control loops at 50 ms intervals. The controller never relies on the cloud for safety-critical decisions — battery protection, anti-islanding, overload — but accepts dispatch signals from our control plane every 4 seconds.

The control plane runs on three geographically diverse AWS regions with active-active failover. Mean dispatch latency from National Grid ESO instruction to first physical response: 380 ms across the fleet. P99 latency: 1.8 seconds. Both comfortably inside the 2-second requirement for Dynamic Containment.

A VPP that inconveniences customers will fail. Our hard rule: the household always wins. If your battery is below 30%, we don't dispatch it. If your EV is plugged in and below 50% SoC, we never reverse the current. If your heat pump is heating the house, we don't curtail it. The VPP optimises within the constraints the customer sets, never against them.

What customers see in the app: a daily summary of how much money they earned, what service it came from, and how it changed their own energy use (usually: not at all). The median customer notices VPP activity twice a year — typically during a wholesale price spike when they get a notification celebrating an unusually large daily payout.

VPP revenue lets us price hardware below grid-tied competitors. Lower hardware prices accelerate adoption. Faster adoption gives the VPP more dispatchable capacity, unlocking better grid contracts. Better contracts increase revenue per kW, which funds further price cuts.

By 2028 we forecast S2V hardware below £4,200 per household — 38% below current price. By 2030, a fully integrated solar + storage + S2V + V2G system will cost less than a mid-spec EV and pay for itself in 28 months.

Three policy changes would accelerate this trajectory by 2-3 years: (1) reform the Smart Export Guarantee to a guaranteed minimum 15p/kWh floor; (2) require all new-build properties to ship with EV charging conduit and 16 mm² battery cable to a designated location; (3) align Building Regulations Part L with the 2050 net-zero pathway by mandating SAP 92+ from 2028 rather than 2032.

None of these cost the Treasury anything. All of them unlock £4-7 billion of private investment per year. The British solar industry made this exact policy argument in 2012, failed to win it, and watched the sector contract by 76% over five years. We have a second chance and we're not going to miss it.