Behind-the-Meter Storage: Business Case

Executive Summary

1. The Arbitrage Opportunity

Octopus Agile pricing creates predictable daily arbitrage windows. Wholesale pass-through means prices track real-time grid conditions, with consistent patterns driven by demand cycles and renewable generation. The chart below shows the average price by hour of day across the analysis period, with P25 and P75 bands indicating typical variation. The clear diurnal pattern—cheap overnight, expensive at peak demand—provides the arbitrage foundation.

The default operating strategy is greedy arbitrage: charge whenever the current price falls below the historical median (-) and discharge whenever above. Backtesting shows this simple threshold rule outperforms more complex strategies—including time-based scheduling and percentile triggers—by responding to actual price rather than predicted patterns. It captures the majority of available spread without requiring forecasting or parameter tuning. Critically, the system architecture supports remote override: the fleet controller can push explicit commands (charge, discharge, hold) when external signals warrant deviation from the default. This enables coordinated response to predicted events—such as pre-charging ahead of forecast high-demand periods or holding discharge during grid stress events—while maintaining a robust autonomous baseline.

The economics presented here are deliberately conservative: a vanilla trading strategy backtested across a full seasonal cycle, with no optimisation for fleet coordination, demand response revenue, or predictive dispatch. If the investment case is compelling under these baseline assumptions, intelligent fleet management can only improve returns. This approach ensures the core proposition stands on its own merits, with upside from operational sophistication rather than dependence on it.

2. The Load Constraint

Behind-the-meter discharge is limited by two factors: household consumption and inverter capacity. The proposed design intentionally does not export to grid, operating purely to offset household consumption, and uses an 800W inverter aligned with expected balcony solar regulations. While G98 permits up to 3.68kW per phase export, we target the 800W limit anticipated for plug-and-play systems—simplifying installation and positioning for the likely regulatory framework. The effective discharge rate is the lower of inverter capacity (800W) and household consumption.

Methodology: Values derived from 365-day backtest on Octopus Agile Yorkshire pricing data. For each capacity increment, the model simulates daily charge/discharge cycles constrained by 0.8kW base load discharge rate and 90% round-trip efficiency. Marginal value represents the additional annual savings from each incremental kWh of storage capacity.

The marginal value curve shows steep diminishing returns beyond 3 kWh. Each additional kWh of capacity captures progressively less arbitrage value, while the cumulative curve shows the knee point where additional capacity stops adding meaningful returns. The arbitrage capture distribution reveals a powerful insight: most days require only 2-3 kWh to capture 50-75% of available arbitrage value, while 100% capture often demands 4+ kWh—demonstrating why oversizing delivers poor capital efficiency. At 0.8 kW discharge rate, 3 kWh represents the optimal balance.

3. Sizing Comparison: 3 kWh vs 2 kWh

Direct comparison of unit economics with 0.8 kW base consumption and greedy strategy:

At 0.8 kW base load, the 3 kWh system generates higher absolute savings and delivers superior return on capital—the discharge constraint now allows full utilization of the larger capacity during typical 3-4 hour peak windows.

4. Fleet Economics: £100M Deployment

With fixed capital investment, the superior unit economics of 3 kWh systems at 0.8 kW discharge rate translate directly to fleet-level returns:

5. Scale Path

Phased deployment from pilot validation to national rollout:

6. Regulatory Context

7. Proposed Solution

Compact LFP battery units optimised for behind-the-meter arbitrage, designed for single-person self-installation. The optimal configuration balances capacity against the discharge constraint—the analysis presented here uses 3 kWh as the reference case based on 0.8 kW typical base load, but the platform architecture supports flexible sizing (1-3 kWh) to match evolving tariff structures and regulatory requirements.

The core value proposition is technology flexibility: capacity, charge rate, and form factor can be tuned to optimise for specific deployment contexts. What remains constant is the focus on low cost, light weight, and installation simplicity—removing the installer bottleneck that constrains conventional home storage rollout.

A note on the effect of import duties and VAT on unit economics

Imported batteries under the correct HS code attract zero UK import duty, and import VAT is fully recoverable through postponed VAT accounting, meaning the only unavoidable tax cost in the supply chain is the small net China VAT of around 4%. Because an energy supplier can reclaim all UK VAT on purchase, the effective landed cost of the asset reflects only the true manufacturing and freight costs, without any additional tax burden.

If the battery is sold directly to a customer, VAT must be charged at 20% and cannot be reclaimed by the end user, immediately increasing the real cost of the asset and damaging payback. Keeping the battery owned by the energy supplier avoids this VAT leakage, preserves the underlying unit economics, and ensures the system remains as low-cost and financially efficient as possible for both supplier and customer.

8. Discussion & Next Steps

9. Why This, Why Now, Why Us

Structural disruption: GivEnergy, Powervault, myenergi, and traditional installer networks optimise for high-margin, installer-dependent products—they cannot pursue self-install without destroying channel economics. This creates a segment they are unable to serve.

Value capture: The arbitrage is gated by the Agile tariff—only available to Octopus customers. The billing relationship becomes distribution channel and competitive moat. Vendors can manufacture hardware; they cannot manufacture the tariff.

Hardware capability: We bring a decade of custom energy storage development through Chinese manufacturing, enabling £140/kWh landed cost that conventional supply chains cannot match.

You provide customer access and the tariff moat; we deliver hardware at unmatched cost; self-install attacks where incumbents cannot defend. Each partner contributes what the other lacks.