IRadimed's 3870 Pump Is Winning the Demand Test—and Failing the Ramp Test

Record bookings and a 14% revenue jump couldn't stop a guidance trim, because the constraint is now the factory, not the market.

IRMD · Earnings Call · 2026-10-30

The record that wasn't enough

IRadimed reported a record September quarter on October 30: revenue of $24.2 million, up 14% year over year and 18% sequentially, with GAAP EPS of $0.51 and non-GAAP EPS of $0.55. The new 3870 pump is clearly being received well. CEO Roger Susi said the launch is “showing itself to be everything we had hoped for and more” — Roger Susi, CEO · 2026-10-30. Yet the full-year revenue guide was trimmed to $91.7 million-$93.2 million from $91 million-$96 million, and the fourth-quarter guide of $25 million-$26.5 million implies only a modest step-up from Q3. The stock had already voted: shares were down 18.3% over the prior 90 trading days and 25% from the July 6 peak of $102.30. At roughly 53 times trailing net income, this is not a stock that can afford a second consecutive "quality throttle." The story is company-specific: demand is not the problem; the factory and the recurring-revenue attach are.

Demand is not the bottleneck

The 3870 ramp is real. IRadimed produced over 400 systems in Q3 and shipped more than 300, after a production increase of over 125% from Q2. For Q4, management expects another 50%-plus increase in pump output, targeting roughly 500 units. That would put pump revenue at about $12.5 million in the quarter, or a $50 million annualized run rate, and total company revenue above a $100 million run rate. The problem is that the ramp rate is now the binding constraint. Susi was unusually candid about why the guide came down:

We can't get out over our skis pushing this ramp rate so hard that the quality slips. And so yes, you have to sort of -- you have to pay that piper and you have to be careful that the ramp rate is not so ridiculously high that you can't keep the quality up.
Roger Susi, CEO · 2026-10-30

This is not a brand-new issue. In July, Susi was already tempering manufacturing expectations: “The sales team would like us to ship over 200, but we just can't do it.” — Roger Susi, Chief Executive Officer · 2026-05-01 The difference now is that the demand signal has strengthened while the supply response remains deliberately throttled. Q3 bookings for both the 3870 pump and the 3880 monitor were records, and the company entered Q4 with a comfortable backlog. The replacement math is also getting better: roughly 70% of replacement orders for two older channels are converting into one new quad stack. That is up from the prior call's characterization that “more than half of these orders we've taken so far have been for this Quad systems” — Roger Susi, Chief Executive Officer · 2026-05-01. With nearly 6,400 old 3860 channels in the U.S. that are over five years old, the company translates that into a greater-than-10,000-unit 3870 opportunity, helped by a Quad Stak configuration that lifts ASP to nearly $25,000 per channel versus the old 3860's roughly $19,000. The truly new wrinkle is greenfield: facilities that rejected the old 3860 are now ordering the 3870 because its use model is more intuitive. That was upside management had not been counting on.

The recurring-revenue flywheel is late

The quarter's soft spot is disposables. Revenue grew 21% to $5 million, but none of that came from the 3870 installed base yet. Susi was blunt: “frankly, none yet, almost none” — Roger Susi, CEO · 2026-10-30. The reason is operational: newly shipped pumps need six to eight weeks of in-servicing before clinicians actually use them, so the disposable pull-through should begin late in Q4 and compound through 2027. That matters because pump hardware is a one-time sale, while disposables are the annuity. On the prior call, CFO John Glenn expected the quad configuration to help drive utilization; the current call pushes that benefit out. In the meantime, monitors grew 17% to $8 million, ferromagnetic detection grew 28% to $0.5 million, and international revenue jumped 45% to $4.5 million—though international is lower-margin and pinched the mix. Gross margin rose two points sequentially to 76% but remained below last year's 78%. The last filed gross margin was 74.1%; management expects a return to the high-70s early next year as volumes and ASP mix improve.

What 2027 needs to prove

The fourth-quarter guide implies a $100 million-plus annualized revenue run rate, and the 2027 setup is for “upper teens of growth in 2027” — Roger Susi, CEO · 2026-10-30—not the "modest" framing an analyst offered. At 14.2 times trailing revenue, the market is paying for that recovery. The bull case is straightforward: demand is validated, backlog is visible, the 3870 carries a higher ASP, and gross margin should recover as the learning curve flattens. The bear case is equally clear: the company has now trimmed guidance once because it chose quality over speed, disposables are lagging, and the valuation leaves little cushion if either issue persists. The next two quarters are the proof point—not whether hospitals want the 3870, but whether IRadimed can build it, install it, and monetize the consumables behind it.