Abroad, we are selling the fifteenth year

When I began selling storage abroad, at Wenlan Power Cell in 2016, I assumed price would decide most export tenders, because it decided most tenders at home. Price matters abroad too, and Chinese suppliers win on it. But the customers who matter most, the ones who buy repeatedly and whose projects are financed by banks, read our proposals from the back. They begin with the warranty, the service agreement and the promised performance, and they ask who will stand behind them.

This year Qianyue's export sales will be close to RMB 1.6 billion, nearly 30 per cent of our revenue, from customers in Australia, the Gulf and Chile. Our growth abroad has come less from price than from learning to answer those questions before they are asked.

Why buyers abroad read the warranty first

A grid-scale battery abroad is usually financed with project debt. The lenders' engineers model the station's revenue over fifteen to twenty years, and that revenue depends on usable capacity, availability and round-trip efficiency declining no faster than the supplier promised. If they decline faster, the supplier is meant to pay, by adding capacity or by compensating the owner. A warranty is therefore a long-dated liability of the supplier, and a lender judges it as it would any other credit: by the balance sheet behind it and the supplier's record of honouring claims.

Many Chinese exporters treat the warranty as a sales term. They offer the longest period and the most generous degradation curve the competition offers, and they price the system as though claims will rarely arise. At the prices of the past two years, with some suppliers selling below cost, that is a promise many cannot keep. Buyers abroad know it. Several have told me that they now discount any warranty from a supplier whose financial statements they cannot read, and ask for parent guarantees, bank guarantees or insurance behind the rest.

We decided in 2023 to price the warranty explicitly. Every export proposal carries a reserve for expected warranty and service costs over the contract's life, calculated from our own fleet's degradation data, and we hold that reserve on our balance sheet. It makes our bids higher, typically by six to nine per cent, and we lose some tenders because of it. We also win repeat orders from owners who have watched our stations perform and whose lenders accept our warranty without a discount.

We also share our degradation data. Every export customer receives a quarterly report on each station's usable capacity, efficiency and availability measured against the warranty curve, produced from the same monitoring we use for our own fleet. When a station falls behind, the report says so before the owner's engineers discover it. A warranty is easier to believe when the supplier is the first to point out that it may soon be called upon.

Service engineers before sales offices

The second thing buyers read is the service plan. A storage station needs routine inspection, software updates, spare modules and an engineer who can reach the site quickly when something faults. Chinese suppliers have often opened a sales office in a new market and promised to serve it remotely from China. That works for software. It does not work for a fault at two in the morning in a desert substation, where an hour of downtime can cost the owner more than a year of spare parts.

Our rule now is that we do not sign a second project in a country until we have engineers and spare parts in it. In Australia we have 38 people, two-thirds of them hired locally, and spare-module stock in two cities. In Chile we work through a local service partner that we trained ourselves and audit twice a year. Each arrangement costs money before revenue arrives, and in the first year in each country our service business lost money. Availability across our export fleet has stayed above 98 per cent, and it is the first figure lenders ask for when we tender again.

Local people also change how we design. Engineers who live with a station in forty-five degrees of heat tell us things that test reports from Zibo do not. Two changes to our cooling design in 2024 came from our team in the Gulf, not from headquarters.

Trade barriers are a reason to localise, not to discount

Exporters from China face growing barriers: tariffs, local-content rules, security reviews of the software that controls grid assets, and new requirements on carbon footprint and traceability, such as those in the European Union's battery regulation. The tempting response is to absorb each barrier in price, cutting margin to stay competitive. I think that is the wrong response, for the same reason that selling below cost at home is wrong. It weakens the balance sheet on which the warranty depends.

The better response is to do more of the work in the market. That means local assembly where volumes justify it, local service and local construction partners, operating data stored and controlled where the customer and its regulator require, and openness about where every cell comes from. It also means choosing markets. We do not sell into the United States, because the tariff and policy risk there is too high for a company of our size to carry across a fifteen-year warranty. That decision has cost us opportunities some competitors have taken, and I accept that they may prove me wrong.

Software deserves particular care. A grid-scale battery is controlled by an energy management system that talks to the grid operator and, often, to the supplier's monitoring centre. Customers and governments are right to ask who can reach that system, and from where. We now offer every export customer local hosting of its operating data and the option to disconnect our remote access entirely, with service then delivered on site. Some have taken it. I would rather lose a little efficiency in support than leave a customer uncertain about who controls its asset.

Chinese storage has become the world's default supply in only a few years, mostly on cost and manufacturing scale. Keeping that position will take something else: the willingness to stand behind a station when it is old, in a country far from the factory. The overseas buyers who matter most are already pricing that willingness. Exporters should start pricing it too.