Chief Operating Officer / China
Choosing the visible cost over the hidden one
- Vehicles delivered in 2025
- 138,000
- Including 24,000 exported to 19 countries
- Revenue, 2025
- RMB 19.6 bn
- Electric vans and light trucks, parts and service
- Colleagues in operations
- 9,800
- Two plants, procurement, quality, logistics and service
Decisions & consequences
01Idling a Chongqing line instead of matching a 12 per cent price cut
In March 2024 two competitors cut the list price of their 3.5-tonne electric vans by about 12 per cent, and others followed within weeks. Our sales leadership proposed matching across the range and moving both plants to two shifts, aiming for 165,000 deliveries that year against 131,000 in 2023. The controlling shareholder's nominee director supported them. Market share mattered to the group, and the Chongqing plant stood on land provided by a district government that expected it to run.
I opposed the plan. Dealers and fleet customers were already holding 2.6 months of our stock against a normal 1.2. Matching the cut would have taken roughly RMB 17,000 off each core van and gross margin from 15.1 per cent to about 4 per cent, for volume our own order data did not show. I proposed holding the core van's list price, offering fleet buyers an eight-year battery warranty and repairs at their own depots instead, and idling the second assembly line at Chongqing from July. The board's strategy committee backed the proposal, and the board approved it after a long discussion.
The cost was real. 640 people worked on that line. We transferred 410 of them to the Changsha pack line, a second shift on Chongqing's remaining line and the service network; 230 accepted voluntary severance above the statutory amount. Deliveries fell to 126,000 in 2024, and our share of the electric light-commercial market from 9.8 to 8.6 per cent. The district government withheld a RMB 26 million incentive instalment linked to output.
Gross margin for 2024 was 13.9 per cent and net profit RMB 420 million, while two larger rivals reported losses. Channel stock was back to 1.3 months by December. In 2025 deliveries recovered to 138,000 and our share to 9.1 per cent, with Changsha at 92 per cent utilisation. Our capacity plan now separates contracted demand from demand that only a price cut would create.
I would do two things differently. I would have visited the district government before the board paper was written; it learnt of the decision from our announcement and has been cautious with us since. And I would have funded the preservation of the idled line properly. We underspent, and the estimated cost of restarting it has risen by RMB 31 million.
Filing a recall after a depot fire, and keeping the supplier
In July 2023 a Tanzhou van caught fire while charging overnight at a customer's depot in Foshan. Nobody was hurt; three neighbouring vans were damaged. Within two weeks our engineers and our harness supplier traced the cause to high-voltage connector housings from one moulding batch, with internal voids that let moisture in and allowed arcing under load. The suspect parts were fitted to 7,900 vans built between November 2022 and April 2023. Our incoming inspection, five housings a lot checked by eye, could not have found them.
Two responses were proposed. Our sales leadership, and one of our largest fleet customers, wanted a service campaign: replace the connector at each van's next scheduled service over about nine months, without a formal recall in the middle of the autumn delivery season. I recommended a recall filed with the State Administration for Market Regulation, a written instruction to every owner to charge the affected vans only under supervision, and replacement within months rather than at routine services. The same fault could start a fire in a depot with people working in it. The president agreed.
Procurement wanted to delist the harness supplier, and I argued against that too. Its engineers had worked alongside ours throughout the investigation, and a replacement would take a year to qualify. We kept it on three conditions: a second moulding source, X-ray inspection of every high-voltage housing, and lot-level process data shared with us. The supplier bore 60 per cent of the cost.
The recall cost RMB 64 million, including 600 loan vans placed at customers' depots, and we estimate it cost us about 1,100 orders that autumn. Completion reached 98.7 per cent within five months. There has been no further incident from that failure, and connector failures in the field fell from 41 to 5 per million. The supplier quality agreement we rewrote afterwards rewards a supplier that reports an escape before we find it.
I would change one thing. We waited nine days for a confirmed root cause before telling owners to supervise charging. That advice cost nothing to give and should have gone out on the second day, while the engineers were still working.
Levelling the build plan against the parcel carriers' autumn
When I arrived, the rhythm of our Changsha plant was set by the parcel carriers. They placed most of their orders in early summer and wanted vans by the end of September for the November peak. The plant built 42 per cent of its annual output in the third quarter, on six-day weeks, with dispatched workers reaching 2,300. Our own data showed the cost. First-pass yield fell from 93 to 84 per cent in August and September, lost-time injuries in the third quarter ran at more than twice the rate of other quarters, and vans built in the autumn produced 1.6 times the early warranty claims of vans built in spring.
In late 2022 I proposed building at a steady daily rate from February to November and holding finished vans in a new 9,000-place yard at Changsha. Large customers would sign early allocation agreements in the first quarter, with title passing on delivery and price protection if our list price fell before then. Our chief financial officer disagreed, reasonably: working capital at the peak would rise by RMB 1.1 billion, and price protection exposed us to any price war. The sales team expected smaller carriers to refuse to commit early, and many did.
We began in 2023. We gave up about 3,400 orders from buyers who would not sign in advance, carried RMB 38 million a year of extra inventory cost and, in 2024, paid RMB 52 million of price protection when market prices fell. In the first winter, 140 stored vans suffered deep discharge because we had not planned maintenance charging, and 22 needed new packs.
The results justified the cost. The third quarter's share of output fell to 29 per cent. Dispatched workers at the peak fell to 640, first-pass yield averaged 94.1 per cent across 2023, lost-time injuries fell by 58 per cent and early-life warranty claims by 31 per cent. Overtime premiums fell by RMB 74 million. The yard is now a planned part of the plant, with its own charging rota and stock controls that our auditors examine.
If I did it again, I would pilot the agreements with our two largest customers before offering them to everyone, and I would design the yard's battery maintenance before parking the first van in it.
Putting vans in containers when the car carriers ran out
In October 2022 we signed our largest export order to that date: 2,600 left-hand-drive compact electric vans for a Chilean retail-distribution company's delivery fleet, delivered to the port of San Antonio between February and July 2023, with late-delivery penalties of up to 8 per cent of the contract's value. Weeks later our ro-ro carrier cancelled two sailings. Car-carrier space was scarce everywhere, spot rates had roughly tripled and the next space we could book was in June.
Our export sales head wanted to ask the customer to accept a delay. Our CFO preferred to rely on the contract's force majeure clause. I doubted that a shortage visible to the whole industry when we signed would support such a claim, and the customer had opened new depots that needed vans. I proposed shipping the first 1,400 vans in containers, two compact vans to a forty-foot high-cube box on steel racking, by rail to Shenzhen and on by container ship, and sending the remaining 1,200 by ro-ro in June and July if the customer agreed. It did.
Containers cost RMB 16,200 a van against RMB 7,400 at our contracted ro-ro rate, adding about RMB 12 million, and the order's margin fell from 12 to 6.5 per cent. The first sailing went badly: 3.8 per cent of its 180 vans arrived damaged, mostly mirrors and side panels, because our lashing let them shift in heavy seas. We redesigned the lashing with the forwarder, and damage on later sailings fell to 0.6 per cent. Across all containerised vans it averaged 1.0 per cent, against 0.4 per cent by ro-ro.
All 1,400 arrived by May and no penalties were paid. The customer has since ordered 1,800 more vans in 2024 and 2,100 in 2025. Export logistics moved into my organisation that year. We signed a three-year ro-ro agreement for up to 9,000 vehicles a year from 2024, kept containers as a qualified second route, and added a freight-adjustment clause to every export contract.
I would test the lashing on a single loaded box before committing 700 of them, and I would never again sign a fixed delivered price without a freight clause.