My position: I own ONDS at an average cost of about $9, and I'm holding it. Nothing here is investment advice.
Key points
- Ondas posted record Q2 2026 revenue of $83.8 million, up 67% from Q1 and more than 13 times the $6.3 million it did a year ago, and raised its full-year target to $525 to $550 million.
- Management guided Q3 revenue to $140 to $155 million, nearly double Q2, and the full-year math leaves roughly $250 million to land in Q4.
- Ondas still lost money: a net loss of $89.7 million, or $0.19 per share, on heavy acquisition and ramp spending. The stock fell about 9% in midday trading.
- The company added 560 employees, 155 of them engineers, brought six facilities and 230,000 square feet of unused capacity online, and closed its DZYNE and Cyberhawk deals. Backlog is $613 million reported, $757 million pro forma.
A week ago I wrote that Ondas (ONDS) earnings on August 13 would tell me one thing: whether July's run was a real re-rating or short covering with good timing. I own the stock at about $9, so I wanted a clean answer. The report gave me one, and it wasn't the answer the tape traded on.
Revenue came up 67% from the first quarter at a record $83.8 million which was more than 13 times the $6.3 million Ondas did a year ago. Management raised the full-year target to the $525 to $550 million range, up from the "at least $390 million" it guided back in May. The stock fell hard, down about 9% in midday trading. If you only read the headline and the price, you'd think something went wrong. Nothing did. This was never the quarter that mattered.
The quarter everyone is looking at is the small one
Here's the part that got lost in the red. Ondas guided third-quarter revenue to $140 to $155 million. That's not a typo, it's almost double what the company just printed. Run the full-year math and it back-loads even harder. Q1 did $50.1 million, Q2 did $83.8 million, and a Q3 midpoint near $147 million leaves something north of $250 million to land in the fourth quarter to reach the middle of the new range. So the record quarter everyone is reacting to is actually the smallest of the back half of the year.
CEO Eric Brock put it in the release without much hedging: "We expect to sustain this momentum and deliver another significant revenue ramp during the second half of 2026, increasing our full-year 2026 revenue target to a range of $525 million to $550 million." He added that the business should "continue to accelerate in the second half of 2026 as volume deliveries ramp on key programs, particularly across our counter-drone, multi-domain ISR and precision strike verticals."
That's the whole thing, and it lines up with why I bought this in the first place. The big revenue number was always going to show up in Q4 and roll into 2027 as the backlog converts, not in a summer quarter. Management just put a range on it.
The loss is the price of building this fast
Ondas lost money, and a lot of it on paper: a net loss of $89.7 million, or $0.19 per share, with an adjusted EBITDA loss of $50.6 million. That was the part I expected, and it's worth being clear about why. This is what it costs to bolt five businesses together in under two years and staff up for three times the revenue it's doing now. Gross margin was actually 43.1%, so even with the big loss the company isn't selling its product below cost.
I flagged the dilution last time and I'm not going to soften it now. Share count is up to about 530 million, and Ondas keeps paying for acquisitions partly in stock. That's the real knock on this name, and it's why short interest has been so heavy. But the balance sheet isn't the thing to worry about. Ondas ended the quarter with $1.4 billion in cash, cash equivalents, restricted cash and short-term investments, including $657.9 million in straight cash. A company spending into a ramp is a lot less frightening when it's sitting on that kind of pile.
What tells me the ramp is real
Guidance is just a promise until a company builds the capacity to deliver it, so that's where I looked. In and around the quarter, Ondas added 560 employees, 155 of them engineers, and brought six manufacturing facilities online with 230,000 square feet of what it calls underutilized capacity, plus another 50,000 square feet of upgrades planned. You don't hire 155 engineers and light up six plants to keep doing $83 million quarters. You do it for the $250 million ones.
The backlog backs it up. Reported backlog was $613 million as of June 30, up from $68 million at the end of last year, and $757 million pro forma once you count DZYNE, which closed in July, and Cyberhawk, which closed August 10. On top of that, Ondas booked $175 million in new orders during Q2 and another $105 million so far in Q3. My own rough backlog count last week was around $624 million, so the reported $613 million told me my math wasn't far off, and the acquisitions push it well past that. We laid out the government spending shift driving these orders in an earlier piece, and broke down the DZYNE deal here.
Where I stand: I didn't sell into the drop and I didn't add. Nothing in this report changed the reason I own it, and the one thing I asked management for last week, guidance that holds or moves higher, and it moved a lot higher. The company just guided the next quarter to nearly double this one, raised the year, and is hiring and building like it means it. I care about the Q4 print and what 2027 looks like once this backlog turns into revenue, not the number that made the stock red today.
Sources
- Ondas Inc. second-quarter 2026 earnings release and financial statements, August 13, 2026: revenue, net loss, adjusted EBITDA, cash, backlog, bookings, Q3 and full-year guidance, and Eric Brock quotes
- Ondas Inc. first-quarter 2026 earnings release for prior guidance and Q1 revenue
- Ondas Inc. DZYNE and Cyberhawk acquisition announcements
- Market data and trading volume as of midday August 13, 2026
