ECARX heads into its third-quarter report with a problem that sounds like a good one on the surface. Revenue is growing, but each dollar is earning less. Last quarter the company delivered the volume rebound it had promised, then warned that rising memory chip costs would eat into margins for the rest of the year. The stock has fallen about 25% since then, so the market has clearly taken that warning seriously. The November 4 release, due before the open, will show how much of the warning has actually come true.

There is no usable Street consensus for earnings or revenue, so management's full-year targets are the best yardstick. ECARX reaffirmed 2026 revenue of $1.0 billion to $1.1 billion last quarter but did not raise it. That leaves a heavy load for the back half. Second-quarter revenue of $225.2 million was up 71% from the first quarter, which implies first-half sales of roughly $357 million. To reach even the low end of guidance, the company needs about $645 million over the final two quarters. That is well above the $219.9 million it booked in the year-ago third quarter. A third-quarter figure that steps up meaningfully from $225 million would keep the target believable. A flat quarter would make the guidance look like it relies on an improbable fourth-quarter surge.

Volume and mix are the drivers to watch. Shipments recovered to about 550,000 units in the second quarter from roughly 360,000 in the first. The higher-end Antora and Pykes platforms reached 42% of shipments, helped by nine new models entering production across four brands, four of them outside China. Continued unit growth and a rising premium share would support the story that ECARX is gaining content per vehicle even while Chinese auto demand stays soft. A stall in shipments would raise questions about how durable those launches really are.

Margins are where the narrative is most exposed. Gross margin has slipped every quarter, from 22% a year ago to 19.8% last quarter, and full-year guidance of 15% to 18% points to steeper declines ahead. Management also acknowledged that the roughly 15% hardware margin in the second quarter was helped by cheaper memory still in inventory. That cushion is expected to run out in the second half. Adjusted EBITDA has already shrunk to $0.5 million, and that streak of positive quarters now runs four deep. If cost discipline holds, operating expenses that fell 11% year over year while revenue grew could keep a fifth positive quarter within reach. A slide into negative territory would be a clear sign that the memory problem is overwhelming that discipline. Signs that higher memory costs are being passed through to customers in pricing, rather than absorbed, would be the most reassuring detail.

The strategic agenda adds a second layer. The planned roughly $266 million purchase of the FlyMe software business is meant to fix one of ECARX's weakest lines, a software segment that brought in just $0.7 million last quarter. Progress toward closing, along with new licensing customers beyond a recently nominated international luxury brand, would show that the deal can change the margin mix. Updates on the Qualcomm Ventures share exchange and the TPK LiDAR partnership matter less near term but bear on the long-range story.

Sentiment has swung from mildly bullish before the last report to firmly bearish now. At $0.82, the shares sit just above their post-earnings low of $0.78 and far below the 200-day average near $1.22. The trading range has also shifted lower than the prior quarter's. Expectations are therefore subdued, which lowers the bar somewhat. Even so, the core question is unchanged. ECARX must show that its revenue growth can pay for itself, with margins settling within guidance rather than falling through it.