India Has Tendered 260 GWh of Storage. It Has Built 2 GWh of Cells.

Written 26 August 2026, from a Wood Mackenzie assessment published 20 August. Pipeline and capacity figures are analyst estimates, and announced capacity in particular is a statement of intent rather than a measurement.

India’s competitive storage tenders in 2026 add up to a demand pipeline of roughly 260 GWh. India’s commissioned domestic cell manufacturing capacity is about 2 GWh. Domestic cells are therefore covering under one per cent of what the country has already asked the market to supply.

Wood Mackenzie’s director Ankita Chauhan put it in one sentence: “The gap between policy intent and operational capacity is wide.”

• •226 GWh announced through 2035, 2 GWh built. Four players have commissioned gigafactories. Announced capacity and capacity are separated by a decade of execution.
• •China holds about 2,695 GWh of cumulative capacity. India’s entire announced 2035 ambition is smaller than a rounding error against the incumbent’s existing base.
• •Locally made cells are expected to cost 25 to 40 per cent more than imports. Not because Indian manufacturing is intrinsically expensive — India is assessed as second only to China on cost, with a 9 per cent advantage over South Korea — but because scale is missing.
• •A 5 GWh plant runs at about −10 per cent EBITDA. Breakeven needs roughly 10 GWh and healthy margins need 20 GWh. Most announced Indian plants are sized below the point where the arithmetic works.
• •10 to 15 years to a globally competitive, self-sufficient cell industry. That is the headline finding, and it is a longer horizon than most of the policy is written against.

A 20 per cent domestic content requirement now applies to grid-scale BESS projects. Wood Mackenzie estimates that scaling that to 100 per cent would add roughly 30 per cent to total capital cost on a 100 MW, two-hour project.

That figure is the whole policy argument in one number. Thirty per cent more capital for every megawatt-hour deployed in the interim, in exchange for an industry that exists at the end of it. Whether that is a good trade depends entirely on whether the industry actually materialises — which is precisely what the 10-to-15-year finding calls into question.

It also has an immediate effect on the system prices we wrote about in our note on the three-fold spread in turnkey storage costs: India is deliberately choosing not to buy at the cheapest available price.

• •A tender pipeline is not demand. Tendered capacity includes projects that will be re-bid, deferred or cancelled, and comparing it against built factories overstates the gap somewhat.
• •Cells are not the whole battery. Module and pack assembly, BMS, thermal design, enclosures, testing and installation are already done in India at real volume, and they are a large share of the value.
• •Import dependence is not the same as vulnerability, but it is not nothing either. It is a currency exposure, a freight exposure and a policy exposure, all at once.
• •Europe tried this and largely failed. We wrote separately on what happened there, and the lesson was that money and political support did not substitute for process yield.

For anyone specifying a battery in India over the next few years, the honest reading is that the cells in it will be imported, and that this is not a defect in the supplier you are talking to. What separates a good pack from a bad one here is not the flag on the cell — it is the pack engineering, the BMS, the thermal design, the testing and the warranty behind it, all of which are done locally and vary enormously.

For anyone reading the manufacturing announcements as a reason to wait: do not. A decade is a long time to run on a worn-out lead-acid set. We have written on what catching up would actually require and on why Europe’s attempt did not work.