Open in interactive viewer → charts, metric popovers & call review

Helios Technologies: A Record Quarter, a Halved Stock, and the 2027 Confession

Q3 was the best third quarter in company history and management raised the full-year sales guide — then the shares fell 30% in four days. The transcript explains why.
HLIO · Earnings Call · 2026-11-06

The record nobody wanted

Helios Technologies delivered a genuinely good quarter on paper. Sales of $230 million were a third-quarter record, up 4% as reported and 13% pro forma; adjusted diluted EPS rose 11% to $0.80, three cents above the top of guidance. Management raised full-year sales to $895–902 million — the low end alone eclipses the prior record of $885 million — and net debt fell $109 million year-over-year to $255 million, or 1.4x trailing adjusted EBITDA versus 2.4x a year ago. Cash from operations of $40 million was a third-quarter record. And the stock? Down roughly 42% over ninety trading days, with a brutal 30% slide in just four sessions into the print, sitting 43% below its August 7 high of $83.39. The tape is not arguing with the quarter. It is arguing with what came after it.

We are not providing specific 2027 guidance today, but the priorities remain the same... we are focused on delivering sustainable compounding growth over the long term and would not assume that this year's double-digit pro forma growth rate is indicative of our expected annual growth profile going forward.

Sean Bagan, Executive (likely CEO) · 2026-11-06
That is the confession. Five consecutive quarters of double-digit pro forma growth — the CEO even leans on the Yankees, Canadiens and Real Madrid five-peat analogy — and then an explicit instruction not to extrapolate it. The fourth-quarter guide does the rest of the damage: $205–212 million, which Jeremy Evans concedes is “down 3% to slightly up” — Jeremy Evans, Executive (likely CFO or similar financial role) · 2026-11-06 year-over-year once you strip out last year's extra shipping week (roughly $10 million) and $4–5 million of orders pulled forward ahead of announced price increases.

A green shoot pointing into a rolling tape

Buried in the prepared remarks is the genuine new thing: Faster secured its first thermal-management order for data centers, the milestone management had been telegraphing for three quarters. The addressable prize, per Evans: “roughly $1 billion from a market opportunity, and that does represent new build-outs as well as replacement within existing data centers” — Jeremy Evans, Executive (likely CFO or similar financial role) · 2026-11-06. Sean Bagan describes a deliberately incremental go-to-market — “we're going to look to stack a lot of small wins together” — Sean Bagan, Executive (likely CEO) · 2026-11-06, with 30-odd prospects sampling product and NDAs signed with a couple of hyperscalers. This is a company-unique theme. Existing data centers and the coupling/quick-disconnect vocabulary are fresh in Helios' keyword history; a quarter ago the emphasis was simply that this represented the "single largest opportunity across Helios." That prior framing now reads a touch promotional next to the discipline of this call, where the order is explicitly immaterial to 2026 and contributes nothing to the guide. Here is the contrast that matters. Helios is entering a market that the public tape has already begun to doubt — the broad AI data center complex is one of the market's heaviest decliners over the last 30 and 90 days, with dozens of negative constituents across optics, memory and power. Bagan even concedes the midterm-election spotlight on build-outs could slow penetration. Helios may be right on a multi-year view, but it is selling a 2028 story into a tape that spent this quarter de-rating exactly that story.

What management stopped saying

Perhaps the quietest signal is an absence. The IEEPA tariff refund theme, which ranked near the top of Helios' keyword set last quarter and got a full paragraph in May, has vanished entirely from this transcript. The May language was careful and conditional: “We continue to monitor that situation closely. We will pursue refunds, although it is not extremely clear yet how that process will play out” — Jeremy Evans, Chief Financial Officer · 2026-05-12. By November, not a word — even as the refund theme runs loudly through other reporters this week (industrial and consumer names alike). Dropping a contingent asset after quarters of emphasis is its own disclosure. What replaced it is cost reality. Evans flags metals and tooling inflation on the hydraulics side, memory chips and circuit boards in electronics, and freight where “fuel surcharges are as high as they've been since COVID” — Jeremy Evans, Executive (likely CFO or similar financial role) · 2026-11-06 — a more-than-100-basis-point year-over-year drag on gross margin, over 50 sequentially. That is the old electronic component worry from March and May finally showing up in the P&L rather than in a risk paragraph. The margin math shows the strain. Gross margin of 34.6% was up 2.8 points year-over-year in the last reported quarter, yet sits inside a decade-long downtrend of roughly six points from its 2017 peak of 39.9%. Adjusted EBITDA margin this quarter slipped to 20.3% from 20.5% despite $1.4 million more EBITDA.

Cheap for a reason?

The bull case is not silly. Hydraulics grew 15% pro forma against low-single-digit NFPA market data — Bagan is emphatic that this is share gain, not restocking — and Sun Hydraulics is winning back distributors after years of delivery problems. The Radio Sound tuck-in is a disciplined template: $1.7 million deployed, $1–2 million of revenue, margins matching Electronics once synergy capture lands in early 2027. Deleveraging has been real; Effective net cash improved to about -$159 million from -$256 million, tracking the 1.4x leverage figure management cites. But the setup is a $1.56 billion market cap on ~$900 million of sales and a 20% EBITDA margin, against a stock that has just been violently re-rated. The investing question is no longer whether Helios can grow — it can — but whether a 5%-plus organic compounder with an unproven data center option deserves to trade at anything resembling the multiple it carried in August. The market answered 30% in four days. The transcript's own honesty about 2027 is what gave it permission.