The building blocks
A residual value model starts with the pack's specification: energy capacity, chemistry, module count and format. It then applies a forecast of state of health at the expected retirement date, derived from telematics trends or degradation curves for similar vehicles. The result is an estimate of usable energy remaining when the pack leaves the vehicle.
That energy is valued through two routes. The second-life route multiplies reusable capacity by a price per kilowatt-hour for the relevant grade, less repurposing costs. The material route multiplies recoverable metal content by payable metal prices, less processing costs. The pack's value is typically the best combination of the two across its modules.
Key sensitivities
State of health is usually the largest driver, because it determines how much of the pack qualifies for second life and at which grade. Chemistry comes next: nickel-rich packs carry more material value, while LFP packs depend more on second-life demand. Standardised formats that are easy to disassemble and integrate also command better prices.
Market conditions add volatility. Metal prices move material values, and second-life prices depend on demand for stationary storage and the cost of new batteries. Models should run scenarios rather than a single point estimate, particularly for retirements several years away.
| Driver | Effect on value |
|---|---|
| Higher measured SOH | More modules qualify for higher grades |
| Nickel- and cobalt-rich chemistry | Higher material recovery value |
| Standard, serviceable format | Lower repurposing cost |
| Falling new battery prices | Downward pressure on second-life prices |
Keeping the model honest
Models are only useful if they are calibrated. After each retirement cycle, compare forecast SOH and value with module-level test results and actual settlement. Differences reveal whether degradation assumptions, price inputs or cost assumptions need to change.
Transparency with finance and procurement teams matters too. A residual value estimate often feeds leasing rates and total cost of ownership calculations, so the assumptions, scenarios and uncertainty ranges should be documented and reviewed regularly.
- Compare forecasts with module test results
- Run price scenarios for metals and second-life demand
- Document assumptions and update them each cycle
Field note