Battery Market Intelligence · Last reviewed 2026-09-17
| Geography | Global |
|---|---|
| Year | 2025 (2024 comparison) |
| Battery scope | Grid-scale + commercial + residential storage (all chemistries) |
| Unit | GWh energy / GW power — both, stated per figure |
| Data source | Benchmark Mineral Intelligence, IEA Global Energy Review (see Sources) |
| Metric | 2024 | 2025 | Change |
|---|---|---|---|
| Installations (energy) | ~210 GWh (implied) | ~315 GWh | ~+50% |
| Installations (power) | ~73 GW | ~106 GW | ~+43% |
| China share of additions | — | ~60% | Leader by a wide margin |
Two sources, two metrics — the GWh figure counts energy capacity, the GW figure counts power; they move together because duration changes slowly (see the calculation below).
The average duration, derived: 315 GWh ÷ 106 GW ≈ 3.0 hours — the average storage system installed in 2025 carries about three hours of energy at rated power, up from the 1–2 hour systems that dominated earlier years as longer-duration use cases entered the mix.
The share of the lithium market: storage at ~315 GWh against roughly 1,500 GWh of total lithium cell output in 2025 means storage now absorbs about a fifth of lithium production — a structural shift from the days when EV traction took effectively all of it.
Three engines turn at once: the cost curve (cells near $108/kWh make storage bankable without subsidy in more markets each year), grid needs (renewables' variability creates the arbitrage and capacity markets storage serves), and policy (China's mandates, the US IRA, EU flexibility markets). The combination explains why storage grew through a year when EV demand wobbled — the two segments are finally running on different cycles.
| Layer | Character | Growth engine |
|---|---|---|
| Utility-scale | GWh-class projects, long-duration creeping up | Grid arbitrage + capacity markets |
| Commercial & industrial | Peak shaving, backup | Electricity price differentials |
| Residential | Solar-paired, kWh-class | Retail tariffs + resilience |
The same two-speed shape as the wider industry — utility-scale carries the volume, residential carries the breadth, and the commercial layer sits between them on cost.
Storage market sizes conflict because the boundary is genuinely fuzzy: some counts include only grid-connected projects, others add residential and C&I; some report power (GW), others energy (GWh); and the conversion between them depends on assumed duration. The honest reading: treat the ~300 GWh / ~100 GW / ~50% growth figures as the reliable layer — the direction and the rate — rather than any single vendor's dollar total (see methodology).
My read: Storage just became the battery industry's second engine — a fifth of lithium demand, growing at 50% while EVs grow at 20% — and the structural consequence is that battery economics can no longer be read through the car window alone.
Why: Two engines mean two cycles: an EV slowdown no longer means a battery slowdown, and a storage boom no longer needs an EV boom to carry it. For buyers, suppliers and forecasters, the practical shift is the same — storage's demand, cost and pricing must be modelled as its own market now, because it has earned one.
My editorial view, not investment or purchasing advice.
Return to Battery Market Intelligence · World Battery Hub. Market data carries explicit sourcing and is not investment or purchasing advice.