Intuitive Machines (LUNR $9.86 -7.59%) reports Q3 2026 results on Thursday, November 12, with the call at 8:30 a.m. ET, before the market opens. The stock goes in at $9.86, down 39% from $16.23 at the end of 2025 and 78% below its $43.83 close at the end of May. It fell 33% during the quarter itself, from $21.39 on June 30 to $14.24 on September 30.
This is the first full year with Lanteris, the former Maxar satellite factory Intuitive Machines bought on January 13. Revenue jumped from $52.4 million in Q3 2025 to $206.2 million in Q2 2026, and management guides $900 million to $1 billion for 2026 with positive adjusted EBITDA. The first half brought $392.9 million, so the second half has to deliver at least $507.1 million. Our answer is that the low end of that guide is reachable only if contracts that are still authorizations become definitive contracts in time, and that Q3 comes in below every analyst's estimate.
The data behind this piece
Explore the primary data Equibles carries for LUNR:
Why Q3 will miss
The quarter has to grow twice as fast as Q2 did. Consensus of $247.6 million is 20.1% above Q2's $206.2 million. Q2 grew 10.4% over Q1, but Lanteris joined on January 13, so Q1 held only 78 days of it. Scale Q1's $141.6 million of product revenue, the line that did not exist before Lanteris, to a full quarter and it becomes about $163.4 million. Q2's $166.7 million was only 2.1% above that. On a like-for-like basis, the satellite business was flat in Q2.
Several things push Q3 up. The three geostationary satellites ordered in Q2 started work under their authority to proceed. The Space Development Agency's Tranche 2 and Tranche 3 tracking-layer buses are in production, and work is starting on 18 more buses for L3Harris's AMDT-3 Golden Dome program, an award the CFO said was booked in the third quarter. Goonhilly and COMSAT, the ground-station businesses bought on August 3, add two months. Others pull it down. All 16 Tranche 1 satellites were delivered in Q2, and SiriusXM-11 was due to be handed over in August, so two production programs are winding down.
The 10-Q gives a harder check. On June 30, fixed-price orders not yet performed, which the filing calls remaining performance obligations, were $814.7 million, and the company expects to recognize "approximately 30-35%" of them "over the remaining 6 months". That is $244.4 million to $285.1 million for Q3 and Q4 together. Fixed-price work brought $162.1 million in Q1 and $180.0 million in Q2. Existing firm orders alone therefore point to less fixed-price revenue per quarter in the second half than in the first, not more.
We model $182.0 million of product revenue in Q3, up 9.2%, $43.0 million of service revenue and $1.0 million of grants, for $226.0 million. Our Q3 is still 9.6% above Q2. It is below consensus because we give new work time to ramp, and because the larger awards are not yet contracts.
Our forecast for Q3 and Q4
| Our forecast | Consensus | Guidance | |
|---|---|---|---|
| Q3 revenue | $226.0M | $247.6M (10 analysts, range $235.0M to $260.5M) | None for the quarter |
| Q3 adjusted EBITDA | $2.2M | Not available | None for the quarter |
| Q4 revenue | $270.0M | $280.6M (5 analysts, range $274.3M to $288.4M) | None for the quarter |
| FY 2026 revenue | $888.9M | $919.8M (5 analysts, range $903.9M to $929.1M) | $900M to $1B |
| FY 2026 adjusted EBITDA | About $4M | Not available | Positive |
| FY 2027 revenue | Not modelled | $1,167.6M (4 analysts) | None |
Consensus was captured on October 20, 2026. Q3 EPS consensus is -$0.11 from 4 analysts. GAAP earnings will move with the warrant liability, which rises and falls with the share price: it cost $11.6 million in Q2, when the stock rose, and the 33% fall in Q3 points to a gain this time, so the GAAP loss can look better than the operations.
A third of the backlog is one authorization
Backlog was $1.76 billion on June 30, up from $213.1 million at the end of 2025. The 10-Q explains why it is more than twice the $814.7 million of firm fixed-price orders. After the Lanteris deal, the company "reassessed" its backlog policy, and it now counts the full expected value of work a customer has authorized, not only signed contracts. The $947.3 million difference is made of:
- $587.0 million for a multi-satellite program of three commercial geostationary satellites, "for which we received a $45.0 million authority to proceed and recorded backlog reflecting an estimated total program value of more than $600.0 million". The customer is undisclosed; on the call, the CEO would not say whether it is tied to the C-band spectrum reallocation.
- $316.0 million of funded cost-type and time-and-materials work, where revenue is billed as services are performed: the OMES III and Near Space Network contracts and others.
- $44.3 million of variable consideration the company has not yet counted as revenue, mainly mission-success payments.
That matters for the guide. Management said 25% to 30% of June backlog becomes revenue in the rest of 2026. At 25%, the year lands at $833.4 million; at 27.5%, $877.4 million; at 30%, $921.5 million. The guide's low end needs 28.8%. Consensus, at $528.1 million for the second half, needs exactly 30.0%, the top of the range.
The two filings in a row show the same pattern. On March 31, the fixed-price orders were $792.3 million, with 60% to 65% expected over the following nine months, which implied $295.4 million to $335.0 million for Q3 and Q4 once Q2's $180.0 million is taken out. By June 30, after $1.34 billion of new awards in the half, the schedule for the same two quarters had fallen to $244.4 million to $285.1 million. The new awards are large but long, and the firm work due this year shrank. The second half rests on authorizations turning into contracts, which is how the CFO described it.
Margins
Gross profit before depreciation and amortization was $30.1 million in Q1 (16.1% of revenue) and $35.9 million in Q2 (17.4%). The two halves of the business move apart. Product revenue earned 19.5% in Q1 and 28.4% in Q2. Service revenue, mostly the lunar missions, went from a 5.9% margin to -31.5%, largely because of a $14.7 million increase in the estimated cost to complete IM-4, the fourth lunar mission.
Adjusted EBITDA was $2.7 million in Q1 and -$13.8 million in Q2, so the first half closed at -$11.1 million. To finish the year positive, the second half needs more than $11.1 million. General and administrative spending excluding share-based pay and deal costs nearly doubled, from about $22 million in Q1 to about $42 million in Q2, with more people, software licenses and retention awards from the Lanteris deal; the CFO called the rest of the year "steady state" once those are removed.
We assume a 22% gross margin in Q3 and 23% in Q4, without another IM-4 charge, $40 million to $41 million of cash overhead and $7.5 million to $8 million of research and development. That gives $2.2 million in Q3 and $13.1 million in Q4, about $4 million for the year. A repeat of Q2's mission charge would erase it.
Balance sheet and dilution
Cash was $367.4 million on June 30. The company spent $60 million on operations and $24 million on capital projects in Q2, mainly the lunar data satellites it owns, and the CFO said capital spending stays "at these elevated levels in the coming quarters". It raised money three ways this year:
- 11,574,069 shares at $15.12 to institutions in February, for $175.0 million.
- An at-the-market program of up to $500.0 million, opened June 2. By the August call it had raised $291 million gross at an average of $26.81 a share, leaving about $209 million. At $9.86, each $100 million now costs about 10.1 million shares, against about 3.7 million at the earlier average.
- 22,991,028 shares to Advent International as part of the $853.3 million Lanteris price. Advent sold 11,495,514 of them on July 28 at $12.84, for $147.6 million, and still holds the same number.
The $345.0 million of 2.5% convertible notes due 2030 convert at about $13.11, and a capped call covers the dilution up to $20.98. With the stock below the conversion price, they act as debt. Total shares were 228.9 million on August 6: 173.2 million Class A and 55.7 million Class C, held by the founders' entities.
We expect about $313 million at September 30, with the cash half of Goonhilly's £37.0 million price and COMSAT's $10 million base price paid in August.
What management said
On how much backlog turns into revenue:
“Approximately 25 to 30% of our Q2 backlog is expected to be revenue in 2026, 35 to 40% in 2027, and the remaining thereafter.” Pete McGrath, CFO, Q2 2026 earnings call
On what decides the range:
“The primary variable determining where we land within the range is the timing of contract definitization and revenue conversion, not customer demand.” Pete McGrath, CFO, Q2 2026 earnings call
When B. Riley's Griffin Boss asked what gets the company to the middle of the guide:
“There's also those ATPs as we definitize them and the timing of that definitization will pull revenue into this year versus if they roll later, it would defer it to next year.” Pete McGrath, CFO, Q2 2026 earnings call
On cash:
“Free cash flow is expected to improve throughout the second half of the year as investments stabilize and milestone receivables come in following our recent awards.” Pete McGrath, CFO, Q2 2026 earnings call
On the next landing:
“We are scheduled in the launch window of January through March for Mission 3 of next year on a SpaceX Falcon 9.” Steve Altemus, CEO, Q2 2026 earnings call
The full Q2 call covers the four CLPS bids due this year and the 10-year CLPS 2.0 contract, which the CEO put at over $10 billion across several winners.
Positioning
- Short interest: 38,042,831 shares on October 15, up from 34,036,834 on June 30, with 4.4 days to cover (short interest history). That is 22.0% of Class A shares, a level worth comparing with the most-shorted US stocks.
- Options: the November 13 straddle at the $10 strike costs about $1.69, a move of 17.1% either way by that date (option chain).
- Insiders sell, nobody buys. Co-founder and director Kamal Ghaffarian has sold on 11 dates under a 10b5-1 plan since May 4, for $31.1 million. Chief technology officer Tim Crain sold $5.4 million in June and September. The last open-market purchase was 241,080 shares by director Michael Blitzer at about $9 in November 2025; he resigned from the board on October 21 (insider trades).
- After the last report, the stock rose 3.6% on August 13 and another 8.3% the day after, on the backlog.
Our assumptions
| Assumption | Our value | Basis |
|---|---|---|
| Q3 product revenue | $182.0M | $166.7M in Q2, flat like-for-like; new geostationary and Golden Dome work, Tranche 1 and SiriusXM-11 finished |
| Q3 service revenue | $43.0M | $36.7M in Q2 after the IM-4 charge; Goonhilly and COMSAT from August 3 |
| Q3 grant revenue | $1.0M | $5.9M in H1; the Texas Space Commission grant's budget periods ran to June 30 |
| Geostationary program contract | Signed in Q4 | $45.0M authorization on a $600M+ program |
| Second-half revenue | $496.0M | 28.2% of June backlog, inside the 25% to 30% range |
| Q3 gross margin | 22% | 17.4% in Q2, 24.5% without the $14.7M IM-4 charge |
| Q3 cash overhead | $40.0M | $41.9M in Q2; "steady state" per the CFO |
| Cash at September 30 | About $313M | $367.4M at June 30, ATM sales, acquisitions, Q2-like spending |
We also ran two cases. If the geostationary contract and the other authorizations are signed in Q3 with catch-up revenue, Q3 reaches about $255 million and the year about $935 million, a beat. If they slip into 2027, the second half falls to the 25% end of management's range, Q3 lands near $215 million and the year near $835 million, and the guide is cut.
Risks to our call
- Definitization can come fast. A signed geostationary contract in Q3 would bring revenue for work already done under the authorization, and that alone can close most of our gap to consensus.
- The backlog schedule is approximate. The 30% to 35% figure is a range the company rounds; a few points more on $814.7 million is $20 million to $40 million.
- New awards. CLPS task orders, the lunar orbiter surveyor and national security programs bid this year can add revenue quickly if work starts under authorization.
- Mission charges. IM-4 cost estimates moved by $14.7 million in Q2; another change in either direction moves service revenue and margins.
- A lower share price raises the cost of funding. With $209 million left on the ATM at $9.86, more dilution per dollar raised is likely if cash use stays near $84 million a quarter.
What to watch on the call
- Whether the geostationary program is a definitive contract, and how much backlog still sits on authorizations.
- The full-year revenue guide: kept at $900 million to $1 billion, narrowed, or cut.
- The remaining performance obligations figure and its schedule in the Q3 10-Q, against $814.7 million in June.
- Bookings in Q3, including the AMDT-3 value and any CLPS awards.
- IM-3 readiness for its January to March 2027 launch window, and the IM-4 cost estimate.
- Free cash flow, capital spending on the data satellites, and ATM sales since August.
Sources
- Intuitive Machines Q2 2026 10-Q, filed 13 August 2026 (accession 0001628280-26-056821): document.
- Intuitive Machines Q2 2026 results release, 8-K filed 13 August 2026 (accession 0001628280-26-056476): document.
- Intuitive Machines Q2 2026 earnings call transcript, 13 August 2026: transcript.
- Intuitive Machines Q1 2026 10-Q, filed 15 May 2026 (accession 0001628280-26-035236): document.
- Goonhilly and COMSAT acquisition, 8-K filed 3 August 2026 (accession 0001193125-26-330778): document.
- Board change, 8-K filed 22 October 2026 (accession 0001628280-26-066926): document.
- Quarterly revenue from Intuitive Machines' financials; consensus captured on its estimates page on 20 October 2026; short interest from FINRA; prices to the 2 November 2026 close.
This article is our analysis of public filings and our own financial model. It is not investment advice, and forecasts can be wrong. Equibles staff may hold positions in the securities discussed.