AST SpaceMobile ($ASTS) reported the June quarter today, and the number the release leads with is a $1.30 billion revenue backlog. The number it does not put a denominator on is the one that decides whether the constellation gets finished.
Operating activities used $97.2 million in the three months to 30 June and purchases of property and equipment took $597.6 million, so free cash flow for the quarter was -$694.8 million. Revenue in the same quarter was $31.5 million. The burn is 22 times the revenue, and both sides of that ratio come from the company's own release.
That ratio is not a scandal on its own — a constellation is a capital asset built years before it earns, and every satellite operator's early cash flow statement looks like this. The question it does not answer, and the one this piece does, is how long the current balance sheet funds the current rate.
What is established
- Free cash flow of -$694.8 million in the June quarter. Derived, from the disclosed $97.2 million of operating cash used and $597.6 million of property and equipment purchased.
- Capital expenditure more than doubled sequentially, from $261.6 million in the March quarter to $597.6 million — a factor of 2.3 in one quarter.
- $2.72 billion of cash, cash equivalents and restricted cash at 30 June. Disclosed on the balance sheet.
- "Over $3.7 billion" pro forma, after the $1.15 billion of 1.625% convertible senior notes raised in July. That phrasing is the company's own, in this release.
- Revenue backlog of approximately $1.30 billion, described as aggregate contracted revenue across commercial partners and United States Government awards. It is a backlog, not cash, and the release gives no collection schedule.
- Thirteen spacecraft in orbit, with BlueBirds 14 to 16 ready to ship and production running through BlueBird 46 — thirty-three more spacecraft than are flying today.
- A $125.9 million loss on involuntary conversion sits in operating expenses, from the BlueBird 7 satellite stranded by a New Glenn second stage on 19 April. It is non-cash and is not in the burn figure above.
Two EPS numbers, and only one of them is the miss
The GAAP diluted loss attributable to common stockholders is $0.77. The figure the consensus was set against is $0.44, and the gap between them is almost exactly the stranded satellite.
The $125.9 million write-off is a consolidated charge, and non-controlling interests absorbed 23.0% of the quarter's loss. That leaves roughly $96.9 million attributable to common stockholders, or $0.32 on the 299.1 million weighted-average shares — which takes $0.77 back to about $0.45, within a cent of the adjusted figure. That reconciliation is our arithmetic on the release's own numbers, not a company disclosure.
So the write-off explains the distance between the two EPS lines. It does not explain the miss. Consensus sat at $0.28 of loss on the adjusted basis, and the adjusted result was $0.44 — a 57% shortfall that survives adding the satellite back. Revenue was similarly light: $31.5 million against a press-reported expectation of about $34.5 million, roughly 9% short. That expectation comes from third-party estimate feeds, is not a series this site stores, and the company itself gives no quarterly revenue guidance.
The conversion
Take the company's own pro forma cash figure and divide it by the quarter it just reported:
$3.70 billion ÷ $694.8 million = 5.3 quarters.
On the cash actually on the balance sheet at 30 June, before the July raise, it is 3.9 quarters. On the first half's average burn of $502.2 million a quarter — a fairer rate if you think the June quarter's capex was lumpy — it is 7.4 quarters.
So the honest range is four to seven quarters of runway at something like today's spending, and the July convertible bought roughly one and a half of them. That is the number a reader can carry away, and it is the reason the financing calendar matters more here than the revenue line.
The direction of travel argues for the low end rather than the high one. Capital expenditure went up 2.3 times in one quarter, BlueBirds 14 through 16 are described as ready to ship, production continues through BlueBird 46, and satellites in various stages of assembly are exactly what turns into cash out the door. Nothing in the release suggests the June quarter is the peak.
The denominator the backlog does not supply
The $1.30 billion backlog is a real number, up on the prior quarter and now including more than $125 million of aggregate United States Government awards. Against the last twelve months of revenue, $115.3 million, it is 11.3 times.
Against the burn, it is something else. $1.30 billion ÷ $694.8 million = 1.9 quarters. If every dollar of contracted revenue in that backlog arrived as cash tomorrow, with no cost of delivering it, it would fund under two quarters of this quarter's spending.
That is not an argument that the backlog is worthless. It is the denominator the headline number is missing: measured against revenue it looks transformational, and measured against the capital programme it looks like a rounding step. Both are true, and only the second tells you whether the constellation gets finished without another raise.
What a spacecraft appears to cost, bounded
Six spacecraft — BlueBirds 8 through 13 — reached orbit in the first half of 2026, and the company spent $859.2 million on property and equipment across those two quarters. The naive division is $143 million per spacecraft launched.
This is where most analysis quietly turns that into a unit cost. We are not going to. That $859.2 million also funds BlueBirds 14 through 46 in assembly, advance launch payments, the gateway build-out described as nearly fifty sites in various stages, and the Texas manufacturing expansion. The figure is a ceiling on the marginal cost of getting one satellite to orbit, not an estimate of it, and the true marginal number is not disclosed. The same arithmetic on cumulative gross capitalised property and equipment — approximately $2.3 billion against thirteen spacecraft flying, or $177 million each — is a ceiling for the same reason and a looser one.
What the ceiling is good for is the direction: at any figure inside it, thirty-three more spacecraft costs multiples of the cash on hand.
Against the tracked comparison
The only other direct-to-device operator tracked here is SpaceX ($SPCX), and the comparison runs one way on scale and the other on structure. SpaceX funds its satellite programme out of a launch business that already earns; AST funds it out of the capital markets. That difference, rather than any technical one, is what the runway arithmetic above is measuring.
There is no published R40 model for AST SpaceMobile, so nothing here is reconciled against one and no fair value is implied by any figure in this piece.
What to watch
- September-quarter capital expenditure. If it holds near $600 million rather than reverting toward $260 million, the low end of the runway range is the live one.
- Whether operating cash used stays near $97 million. The December 2025 quarter produced positive operating cash on customer prepayments; a repeat changes the arithmetic materially and a reversal changes it the other way.
- The next financing, and its form. Convertible notes at 1.625% with an effective conversion price of $149.20 are cheap money; the terms of the next raise are the market's own read on the runway.
- How much of the $1.30 billion backlog converts in the second half. Full-year revenue guidance of $150 to $200 million implies $103.7 million to $153.7 million in the back half against $46.3 million booked in the first, so the second half has to do two to three times the first.
- Launch cadence against BlueBird 46. Thirty-three spacecraft is the plan of record in this release; the rate at which they leave the factory is the capex forecast.
Sources and provenance. Quarterly revenue, operating cash flow, purchases of property and equipment, cash and restricted cash, the $125.9 million loss on involuntary conversion, the non-controlling-interest allocation, the 299.1 million weighted-average share count, the $0.77 GAAP diluted loss per share, the $1.30 billion backlog, the thirteen in-orbit spacecraft, the BlueBird 46 production statement, the "over $3.7 billion" pro forma cash figure and the $150–200 million full-year revenue guidance are disclosed in AST SpaceMobile's second-quarter 2026 business update and results, published 10 August 2026, and its quarterly report for the period ended 30 June 2026. The $1.0 billion of 1.625% convertible senior notes due 2034, the $150 million purchaser option and the $149.20 effective conversion price are disclosed in the company's July 2026 filings. The $0.28 consensus and $0.44 reported loss per share are Nasdaq's, on the Zacks adjusted basis; both sides of that comparison are press-reported from third-party estimate feeds and neither is a series this site stores or verifies, and the GAAP $0.77 must not be substituted into it. The roughly $34.5 million revenue expectation is press-reported from the same kind of feed. Free cash flow of -$694.8 million, the -$309.7 million March-quarter figure, the $502.2 million first-half average, the 22x burn-to-revenue ratio, the 5.3, 3.9 and 7.4 quarter runways, the 1.9-quarter backlog figure, the $0.32-per-share write-off reconciliation, the $143 million and $177 million per-spacecraft ceilings, the 11.3x backlog-to-revenue multiple and the second-half revenue implication are R40 arithmetic on those disclosed figures. Trailing-twelve-month revenue of $115.3 million sums the four quarters to 30 June 2026 as reported. No R40 model exists for this company, so no figure here is modelled.
