The HVAC company hiding a semiconductor secret

AAON Is Growing Revenue 42% But Bleeding Margins — Here's Why That Matters More Than You Think

In Tulsa, Oklahoma, a company best known for making commercial heating and cooling equipment is now one of the most interesting reads in the data center boom. AAON, Inc. (NASDAQ:AAON) just posted 42.5% revenue growth year-over-year. By most measures, that's a blowout number. Companies in the S&P 500 would kill for it. But here's the part that should make you stop and think: their earnings per share grew only 29% while revenue grew 42%, and operating margins have collapsed nearly five full percentage points from their five-year average.

That gap between revenue growth and profit growth is the whole story right now.

When a company works harder to generate each dollar of profit, one of a few things is happening. Either they're scaling faster than their cost structure can absorb (hiring, building, buying capacity ahead of demand). Or competition is forcing prices lower even as volumes rise. Or the new business they're chasing (in this case, hyper-scale data center cooling) comes with worse unit economics than their legacy commercial HVAC work. Probably some combination of all three.

The data center cooling angle is real and it's worth understanding. As Nvidia (NASDAQ:NVDA) GPUs and custom AI chips pack more compute into tighter spaces, the heat they generate has become a genuine engineering problem. Microsoft (NASDAQ:MSFT), Amazon (NASDAQ:AMZN), and Google (NASDAQ:GOOGL) are all building at a pace that requires purpose-built cooling infrastructure. AAON makes precision climate control systems suited exactly for that environment. When a hyperscaler breaks ground on a new campus, AAON is in the room.

So the revenue story is legitimate. The market is real. The demand is there.

The problem is valuation. AAON trades at 40.6 times forward earnings with a price-to-earnings-growth ratio of 2.54. That means the market is paying a premium for consistent, high-quality execution, and over the last four quarters, AAON has delivered a 54% earnings beat, then a 35% miss, then a 13% beat, then a 15% miss. That pattern is not the earnings profile of a company in control of its own operations.

The bear case is simple: margins keep compressing, the market re-rates at a lower multiple, and the stock gives back a chunk of its data center premium. At 40x earnings, there's no margin of safety if the growth story stutters.

The bull case is equally simple: this is a capacity-building phase, the cost overruns are temporary, and in 18 months AAON prints the margin recovery quarter that makes today's price look like a gift.

Both scenarios are plausible. What isn't plausible is paying 40x earnings for a company that can't predict its own quarterly results within 30%. The next 12 months will tell you which version of AAON you actually bought. Until margins start moving in the right direction, the story is a watch, not a buy.

✦

Quick Takes

Three more stories worth knowing about

Becton Dickinson Beat the Number and Lost the Plot

Becton, Dickinson and Company (NYSE:BDX) posted a 28% EPS beat while revenue grew just 1.6% against a trailing twelve-month pace of 6.2%. That deceleration is the headline the headline won't say. Net margins compressed from 8.8% to 8.0%, which means the earnings pop came from somewhere other than the core business running better. At 11.5x forward P/E it looks cheap, but a PEG ratio of 10.4x tells you the market has already done this math and decided the growth isn't coming back fast enough to justify excitement. BDX is a show-me story now.

Arrow Electronics Has No CFO and No CEO — But Has a Genuinely Interesting Idea

ARW (NYSE:ARW) held what was essentially an investor pitch dressed up as an earnings call, delivered by an interim board member with no CFO on the line and no actual Q1 numbers. Bill Austin's bet on becoming an exclusive go-to-market operator for software vendors across Europe is a legitimately interesting thesis. He compared the IP&E consolidation opportunity to semiconductors a decade ago, and that framing deserves a serious look. But without permanent leadership, without financials on the table, and in a components cycle only just starting to recover, this is a story for patient capital. Watch the CEO hire.

Installed Building Products Is Proof That Execution Matters Even When the Market Doesn't Cooperate

Installed Building Products, Inc. (NYSE:IBP) just posted its first slightly negative revenue quarter in years, down 0.4% year-over-year. Housing starts are the whole game for this company and right now they're soft. What IBP's team under Jeff Edwards has done is keep operating margins at 13.0% against a five-year average of 12.3%, a genuine improvement in business quality, not just volume riding a hot housing market. The problem is the 25x forward P/E and a PEG of 3.79 are pricing in a housing recovery that hasn't arrived yet. If starts move, IBP's cost structure means earnings leverage will be dramatic. If they don't, this valuation needs a rethink.

✦

Overlooked Stock

SKYT

SkyWater Technology, Inc.

SkyWater Technology (NASDAQ:SKYT) is something you almost never see: a pure-play semiconductor foundry based entirely in the United States, operating out of Bloomington, Minnesota. Most of the world's chip manufacturing runs through Taiwan and South Korea. SkyWater's entire pitch is that it doesn't. It makes analog, mixed-signal, and rad-hardened chips for aerospace, defense, biohealth, and advanced computing customers, sectors where supply chain geography actually matters to the buyer. The U.S. CHIPS Act directed real money toward exactly this kind of domestic foundry capability, and SkyWater has been inside that conversation from the start. It's not trying to compete with TSMC (NYSE:TSM) on cutting-edge logic. It's building the only viable onshore option for customers who cannot, for security or regulatory reasons, send their designs offshore. For a contrarian looking at the reshoring trend from a specific, defensible angle rather than a broad macro bet, SkyWater is worth an afternoon of serious reading.

The Contrarian Take

The SpaceX IPO Won't Be the Rising Tide Everyone Expects

The conventional view is that a SpaceX IPO at $1.75 trillion lifts every space-adjacent stock: Rocket Lab (NASDAQ:RKLB), AST SpaceMobile (NASDAQ:ASTS), Intuitive Machines (NASDAQ:LUNR), all of them. The theory is that institutional money floods into the sector and reprices every public proxy higher. Here's what that view gets wrong. A SpaceX IPO at $1.75 trillion is so large that it will pull capital out of smaller space stocks, not push capital into them. Fund managers who want space exposure will buy SpaceX directly and trim their speculative positions in smaller names to fund it. RKLB at roughly 20x revenue and ASTS with no meaningful revenue yet are priced for a world where they're the only ways to access the theme. The day SpaceX trades publicly, that premium disappears. The smarter play before a SpaceX IPO may be to trim the proxies, not chase them.