Update, September 1, 2026: this preview says that NIO does not disclose brand-level average selling prices and that nobody outside the company knows them. On the earnings call that morning William Li gave one — RMB406,000 for the NIO brand in the June quarter, from vehicle-registration data, and over RMB430,000 in July. Our analysis of the reported quarter carries it, along with the cost, margin and cash-flow figures the release left out.
NIO reports the June quarter before the US open on Tuesday 1 September 2026, with the call at 8:00 a.m. Eastern. Consensus is revenue of about $4.80 billion and a loss of $0.05 to $0.07 per ADS, press-reported from third-party estimate feeds.
Unusually for a preview, the volume line is already public: NIO reports deliveries monthly, and it delivered 107,658 vehicles in the quarter, up 49.4% year over year and 2,342 below the floor of its own guidance of 110,000 to 115,000. So the revenue estimate is not a forecast of how many cars were sold. It is a forecast of what each one was worth.
Divide it out and the number is uncomfortable. On 21 May NIO guided second-quarter revenue of RMB32.78–34.44 billion — US$4,752–4,992 million at the RMB6.8980 rate the company itself used — on 110,000 to 115,000 deliveries. That is about RMB298,000 of total revenue per delivered vehicle, and NIO had just reported RMB305,909 in the March quarter. The guide assumed revenue per vehicle would fall about 2.6%.
To reach the $4.80 billion consensus on 107,658 cars, revenue per vehicle has to be RMB307,552 — 0.5% above the March quarter. The Street is not modelling the decline NIO guided to. It is modelling the opposite.
The points
- Consensus, and its range. Revenue of about $4.80 billion; a loss of $0.05 to $0.07 per ADS. The revenue figure sits 1.0% above the floor of NIO's own guided range and 3.9% below its ceiling.
- The EPS consensus is not on the basis this site stores. The year-ago comparator quoted alongside it is a loss of $0.28 per ADS; the June 2025 quarter this site carries is a GAAP loss of $0.32. A percentage improvement computed across the two is not one, and this site stores and verifies no consensus series.
- Deliveries are disclosed and they missed. 107,658 against a 110,000–115,000 guide: 2.1% below the floor, 4.3% below the midpoint. April 29,356, May 37,705, June 40,597.
- First-half deliveries were 191,123, up 67% — 83,465 in March and 107,658 in June.
- Mix moved in both directions at once. NIO brand 60,945 (56.6%), ONVO 29,124 (27.1%), FIREFLY 17,589 (16.3%). The third-generation ES8 contributed about 33,474 units and the ES9, launched at the end of May, 11,703.
- What each estimate implies per car. The guided floor needs RMB304,483; the $4.80 billion consensus needs RMB307,552; the guided ceiling needs RMB319,905, 4.6% above March. All three are our division on disclosed figures.
- Two thirds of the third quarter is already public, and it is flat. July deliveries were 35,934 and August 35,836, up 14.5% year over year — 71,770 across two months. September needs 35,888 for the third quarter to match the second, and the second quarter's own monthly average was 35,886.
- Margin is the real question. Vehicle margin was 18.8% in March and gross margin 19.0%, the best in four years. Management's full-year aim is a vehicle margin of about 17–18% and a non-GAAP operating profit for the year.
- NIO has no Rule of 40 score here, because as a foreign private issuer it publishes a cash-flow statement twice a year. We set that out yesterday; the June quarter does not change it.
The arithmetic, in one table
| Deliveries | Revenue | Per vehicle | |
|---|---|---|---|
| Q1 2026, reported | 83,465 | RMB25,532.7M | RMB305,909 |
| Q2 2026, guided | 110,000–115,000 | RMB32,780–34,440M | ~RMB298,000 |
| Q2 2026, delivered | 107,658 | — | — |
| At the guided rate | 107,658 | RMB32,077M · $4,651M | RMB298,000 |
| At the March rate | 107,658 | RMB32,935M · $4,774M | RMB305,909 |
| At the $4.80B consensus | 107,658 | RMB33,110M · $4,800M | RMB307,552 |
The reported quarter, the delivered volume and both guided ranges are NIO's. Every per-vehicle figure and the last three rows are our division and multiplication on them; the dollar translations use NIO's own RMB6.8980 rate.
Read the bottom three rows against the guided range of RMB32,780–34,440 million. Volume alone puts revenue below the guided floor, at RMB32,077 million if each car earned what the guide assumed. Holding March's revenue per vehicle gets to RMB32,935 million — just inside the range. And the consensus sits above both, which means the Street is underwriting a quarter in which price and mix rose while volume fell short.
There is a case for that, and it is the ES9. NIO's flagship SUV began deliveries on 28 May and shipped 11,703 units in five weeks — 10.9% of the quarter, at the top of the range. Against it, FIREFLY shipped 17,589 at the bottom of the range and ONVO 29,124 in the middle. NIO does not disclose brand-level average selling prices, and this is where an analysis would ordinarily supply one. We are not going to — nobody outside the company knows it, and a per-brand price invented to make the sum work would produce a revenue estimate far more precise than the disclosure supports. The three rows above are what the public record actually bounds.
The earnings line, and which basis it is on
Consensus of a $0.05 to $0.07 loss per ADS is a real range and worth quoting as one — the low end is 40% better than the high end. What it cannot be compared to directly is the figure on this site.
| Per ADS | This site stores | Quoted with the consensus |
|---|---|---|
| Q2 2025 | −$0.32 GAAP | −$0.28 |
| Q1 2026 | −$0.03 GAAP | — |
| Q2 2026 | — | −$0.05 to −$0.07 |
The four-cent gap in the top row is the whole warning. The comparator travelling with this consensus is not the GAAP loss NIO reported for the June 2025 quarter, so the estimates are not on the GAAP basis this site keeps, and a headline reading "loss narrows 80%" will have computed it across two different measures. On the basis we store, the useful comparison is the sequential one: NIO lost $0.03 per ADS in March.
The operating question underneath is cleaner than either number. NIO produced a non-GAAP operating profit of RMB66.8 million in the March quarter — about US$9.7 million on RMB25.5 billion of revenue, a rounding error in size and a genuine milestone in direction after years of losing billions a quarter. The GAAP net loss in the same quarter was RMB332 million, against RMB6.75 billion a year earlier. June is the third consecutive test of that, on 29% more cars, and management has said it expects a non-GAAP operating profit for the full year — which makes every remaining quarter of 2026 carry the same test.
The third quarter is mostly already known too
This is the part the guidance section of the call has to address, because the monthly data has run ahead of it.
| Month | Deliveries |
|---|---|
| June | 40,597 |
| July | 35,934 |
| August | 35,836 |
| July + August | 71,770 |
| September needed to match Q2 | 35,888 |
All NIO's own monthly figures. The subtraction in the last row is ours.
August was up 14.5% year over year, so the annual comparison still looks strong. Sequentially, though, the third quarter has printed two months at almost exactly the second quarter's monthly average of 35,886 — and needs a third one just like them to stand still. A third-quarter delivery guide materially above 108,000 is a statement that September steps up, and there are six weeks less room to be wrong about it than there was when the June guide was set.
What the price is paying for
Our NIO model was published this morning against $4.23 a share, and it already carries the delivery miss and July's 35,934 as its bear case's anchor.
| Case | Fair value | vs $4.23 |
|---|---|---|
| Bear | $1.06 | −75% |
| Base | $4.57 | +8% |
| Street | $6.85 | +62% |
| Bull | $9.95 | +135% |
All four are ours. The base case exits the horizon at 0.70 times terminal revenue at a 14% discount rate — just above what the market pays NIO on trailing sales today and below XPeng. Hold every other assumption fixed and solve for the exit multiple that makes the base case equal the price, and it is 0.645 times terminal-year revenue. That solve is ours.
NIO is the only one of today's five reporters trading below its base case, and it is trading below it by 7%. That is not a statement that the shares are cheap. It says the disagreement in this name is entirely about whether a volume carmaker that has never produced a positive year of free cash flow is worth two thirds of a turn of revenue or one third. Even the case built on what the sell side underwrites — twenty-four analysts averaging a $7.37 target — only reaches $6.85, and the last 7% of that target is multiple rather than volume.
Which puts Tuesday morning in proportion. Revenue is bounded within about US$150 million by figures already public, and two thirds of the next quarter's volume is public too. The disclosures that move a five-year model are the vehicle margin and the third-quarter guide, because those are statements about the rate, and the rate is what every case in the table disagrees about.
What to watch
- Total revenue against the guided RMB32.78–34.44 billion and the ~$4.80 billion consensus. The delivered volume brackets the outcome at RMB32.08–32.94 billion unless revenue per vehicle rose above March's level; consensus requires that it did.
- Vehicle margin against 18.8%, and the full-year 17–18% aim. The aim sits below the last print, so the question is how much of the March gain the company expects to keep — and the ES9 mix is the reason it might keep more.
- Non-GAAP operating result against RMB66.8 million, and the full-year non-GAAP operating profit target. A third consecutive positive quarter on 29% more volume is the operating claim the equity story rests on; the full-year target then puts the same test on the two quarters after it.
- The third-quarter delivery and revenue guide against 71,770 already delivered. Anything materially above 108,000 assumes a September step-up that July and August have not shown.
- Whether the June-quarter miss gets an explanation. The guide was six weeks old when the quarter closed. Supply, a launch ramp and demand are three very different reasons to be 2,342 cars short.
NIO reports the quarter ended 30 June 2026 before the US open on Tuesday 1 September, with the call at 8:00 a.m. Eastern; the date is from the company's own notice of 20 August 2026. Consensus of about $4.80 billion of revenue and a loss of $0.05 to $0.07 per ADS is press-reported from third-party estimate feeds as of 1 September 2026; it is not a series this site stores or verifies, and the $0.28 year-ago comparator quoted with it differs from the $0.32 GAAP loss per ADS this site carries for the June 2025 quarter, so the estimates are not on our stored basis. Second-quarter deliveries with their monthly, brand and model splits, the July and August figures, the first-half total, the first-quarter results, the second-quarter revenue and delivery guidance, the full-year vehicle-margin and non-GAAP operating-profit aims and the RMB6.8980 translation rate are all NIO's own, from its first-quarter release of 21 May 2026 and its monthly delivery updates. Ours rather than the company's: every revenue-per-vehicle figure, which is revenue divided by deliveries; the implied revenue rows built from them; the September figure needed to match the second quarter; and the fair values, exit multiples, discount rates and the multiple implied by the price, which are assumptions in our NIO model of 1 September 2026 and not company forecasts. The price of $4.23 is as of 1 September 2026; a live quote will differ.
