ACCompany report
Archer Aviation Inc. ACHR
The bet you're really making is that Archer gets its electric air taxi, a four-seat aircraft called Midnight, approved by the FAA and carrying paying passengers before the money runs out. You're betting that a company with almost no sales, $5 million last quarter, can finish certifying a brand-new kind of aircraft while losing about $263 million every three months. Right now it is bleeding fast: the loss just grew to the largest in its history and cash fell to $853 million, a little over a year at the pace it burns, so it keeps selling new stock to refill the tank. You pay $4.3 billion today for that promise, near the bottom of a stock that has lost more than half its value in the past year.
Key data
ACHR · price with moving averages
Source: market data.
The business
Archer builds Midnight, a battery-powered aircraft that takes off straight up like a helicopter, then flies forward on wings, carrying a pilot and four passengers on short hops of 20 to 50 miles. The idea is to fly people over traffic, airport to downtown, at a price closer to a premium car ride than a helicopter charter. Almost none of that happens yet. There are no paying passengers in the United States, the aircraft is not certified, and revenue is only now appearing in trickles from early overseas and defense work. Launch partners include United Airlines, the city of Abu Dhabi, and defense work alongside Anduril, and Boeing holds a stake large enough to name a board nominee. The real asset is a lead in the FAA certification queue and a factory in Georgia. There is no moat until an aircraft is certified and someone pays to ride it. Everything else is a plan.
The numbers
Revenue only started to exist this year, so the honest sequence to read is the loss.
| Quarter | Revenue, $M | Net loss, $M | EPS |
|---|---|---|---|
| Q2 2025 | 0.0 | 206.0 | -0.36 |
| Q3 2025 | 0.0 | 129.9 | -0.20 |
| Q4 2025 | 0.3 | 188.9 | -0.26 |
| Q1 2026 | 1.6 | 217.7 | -0.28 |
| Q2 2026 | 5.0 | 263.2 | -0.34 |
The loss is not shrinking as revenue arrives, it is widening, because Archer is spending harder into building and testing aircraft, not scaling a product. Q4 2025 is derived from the filed full year less the first nine months. The five-year record shows the same shape, a company that has never earned a dollar and spends more every year.
| Period | Revenue, $M | Net loss, $M | EPS |
|---|---|---|---|
| FY2021 | 0.0 | 347.8 | -3.14 |
| FY2022 | 0.0 | 317.3 | -1.32 |
| FY2023 | 0.0 | 457.9 | -1.69 |
| FY2024 | 0.0 | 536.8 | -1.42 |
| FY2025 | 0.3 | 618.2 | -0.99 |
| 2026, 1H to June | 6.6 | 480.9 | -0.62 |
Cash is the whole game. Operating burn ran $149 million in the first quarter of 2026 alone, and capital spending adds roughly $33 million a quarter, so free cash goes out the door at about $180 million a quarter and rising. Against $853 million on hand, that is a little over a year before the next raise is not optional.
| Quarter | Cash, $M |
|---|---|
| Q2 2025 | 1,724.0 |
| Q3 2025 | 595.5 |
| Q4 2025 | 1,021.5 |
| Q1 2026 | 951.1 |
| Q2 2026 | 852.7 |
The sawtooth is the story: stock sales in 2025 lifted cash, burn pulled it back, and the pattern repeats. What this memo believes that the market does not is that the next two years are set less by the FAA than by how many times Archer must sell shares at these prices to stay alive. The print that settles it is the diluted share count and the terms of each new offering, not any single test flight.
Management
Insiders have sold and never bought. Over the past twelve months there were 20 sales worth $4.2 million and zero purchases, led by Eric Lentell at $631,000 in August, Thomas Muniz at $547,000 in May, and Tosha Perkins at $354,000 in March; plan status is not disclosed in the filings read this run. In a company this far from profit, the more telling number is the share count itself, near 781 million weighted and climbing with every capital raise, plus stock compensation that runs at a large multiple of revenue. Founder-CEO Adam Goldstein has guided timelines that, like every peer's, have moved right. No officer has bought stock at $5 to say the market is wrong.
How it fails or surprises you
Dilution outruns progress. Free cash goes out at about $180 million a quarter against $853 million of cash, so Archer must sell stock roughly once a year at whatever price it can get. Each raise near $5 issues far more shares than one at $12 would. The print that shows it first is the diluted share count on the next 10-Q and any fresh at-the-market offering.
Certification lands (right tail). FAA type certification of Midnight, followed by first paid passenger flights in Abu Dhabi or a US market, would convert a story into a business. The market pays little for it because every eVTOL timeline in the industry has slipped for years. The print that reveals it is a certification milestone or a booked commercial flight, not a demo.
Revenue is real, not staged. Sales went $0 to $1.6 million to $5.0 million in two quarters, with $11.3 million of contract liabilities sitting as deferred customer money. This is the fact the bearish read explains least. If defense and overseas orders compound from here, commercial traction arrives before the cash math forces the issue.
Closing thoughts
Nothing in the next four quarters resolves this. No print certifies the aircraft on a schedule you can underwrite, and no quarter turns the loss into profit. What matters is survival: whether Archer still has the cash, and a stock price high enough to raise more, when the FAA finally acts. The left tail is a company that dilutes shareholders down to a stub while it waits, and it is the fatter tail today, because the burn is certain and the certification date is not. The right tail, a certified aircraft with paying passengers and a first-mover lead, is worth multiples of today's price, but you cannot date it.
The bet is still that Archer gets Midnight approved by the FAA and carrying paying passengers before the money runs out. It breaks if the cash runway shortens faster than the certification clock, and the one pair of numbers that tells you first is cash on hand against diluted shares outstanding on each new quarterly filing. Watch those two together, and you will know whether this is a company being built or a balance being drained.
Methodology
Numbers from the 10-Q filed 2026-08-10 (period 2026-06-30) and prior filings; Q4 2025 net loss derived as full-year less nine months. Market data as of 2026-09-07. Consensus revenue and EPS are vendor-compiled. Insider activity from Form 4 filings, trailing twelve months. Valuation history is vendor year-end ratios, not a filing.
Bid Cap
Daily ideas, a 390-name database, and a model long/short book from an investor who mostly covers financials. $70 a month or $700 a year.
Subscribe on Substack


