Munters' Order Book Is Winning the AI Buildout; Its P&L Is Still in Startup

A 70% order surge and SEK 28bn backlog meet negative cash flow, a flattered AirTech margin and a new CEO softening the old 30% script.

MTRS.ST · Earnings Call · 2026-10-23

The commercial machine is running hot

Munters' Q3 2026 report opens with a number that sounds like a different company: order intake up 70%, book-to-bill 1.6, and a backlog up 178% to SEK 28 billion. “We had a significant order intake of plus 70%, a book-to-bill of 1.6. The backlog, which resulted in plus 178%, so a staggering value of SEK 28 billion now in the backlog.” — Stefan Aspman, CEO · 2026-10-23 New CEO Stefan Aspman, taking over from Klas Forsström, can point to a genuinely broad commercial surge: Data Center Technologies orders more than doubled, with a 2.7x book-to-bill, and sales rose 30% sequentially to the highest level ever for DCT. AirTech grew orders in all regions, helped by component sales for battery production in APAC, and FoodTech held steady while software ARR rose 18%. This is where the company meets the global AI data centers theme. Munters is no longer a small supplier hoping for a slice of the buildout; its data center backlog is now large enough that management says 2027 is essentially filled and 2028 is opening up. Peers in the recent reporter set—ABB talking about DC architecture, Bloom Energy pushing 800V DC, CBRE flagging data-center land sales—confirm the same demand pillar. But the global tape is more ambivalent: the AI data centers cluster is up strongly over 360 days but down about 5.6% over the last 90 days, with breadth sharply negative. Munters' order book is running against a cooling tape, which makes execution, not demand, the swing factor.

The conversion problem is the story

Aspman's own framing is unusually candid for a first call. He describes Munters as a project business that must complete four conversions: market potential into orders, orders into revenue, revenue into profit, and profit into cash.

How do you go from market potential and turning that into orders? ... And how do you make sure that, that revenue turns into profit and that profit turns into cash?
Stefan Aspman, CEO · 2026-10-23

On his own traffic-light reading, the first conversion is green, the second is mixed, the third is challenged, and the fourth is negative. That is the quarter in one sentence. The profit conversion in DCT is the weak link. Sales rose 30% sequentially, but DCT profit declined as component constraints forced rework and disrupted the lean flow at the new Virginia factory, while product mix and tariff impact weighed on margins. The tariff hit was 2 percentage points in the quarter, down from 4 points previously, and management expects it to wash out further; Aspman even says next year he will stop talking about it. That is real progress. But the high-teens DCT margin ambition is now explicitly a midterm play with no timing, and the CEO concedes there is more to wish for on profit. Cash is the sharper edge. CFO Katharina Fischer reported negative operating cash flow, driven mainly by the inventory buildup in DCT, only partly offset by customer advances. “We had a negative cash flow operating activities, mainly related to the inventory buildup in DCT, somewhat mitigated by advances from customers.” — Katharina Fischer, CFO · 2026-10-23 Leverage rose to 3.4x from 3.2x, and working capital to sales fell to 4.2%, below the target range. The inventory is not necessarily bad—it is partly a deliberate bet to secure components and be ready for Q4—but it means the record backlog has not yet become cash. Until it does, the 1.5-2.5x leverage ambition remains a promise rather than a trajectory.

The old script versus the new CEO

The most interesting tension is not in the order numbers; it is in the language. On the prior call, then-CEO Klas Forsström was explicit: “The guidance on 30% is still valid, and that is what we hold. It is tilted more towards the end of the year.” — Klas Forsström, Chief Executive Officer (CEO) · 2026-07-17 He also said the DCT ramp was roughly a quarter behind and that the real revenue booster would come in Q4. Earlier, in January, he promised that chiller production would be fully up after Q1 and that the big U.S. delivery increase would start from Q2. “we will have the chiller production fully up and running in U.S. after Q1. So the big increase of deliveries in U.S. will, of course, start to come from Q2 and forward.” — Klas Forsström · 2026-01-29 In Q3, the new CEO still calls the delay roughly a quarter but immediately adds that it is no exact science, and when asked about the 30% growth guide he says: “I would like to get away from the exact numbers here. There is a range, right?” — Stefan Aspman, CEO · 2026-10-23 He then calls 30% highly reasonable with caveats. The formal outlook is unchanged, but the rhetorical hedge is new. That matters because the prior calls already used supply-chain language for two quarters; the market has heard the excuse before. Aspman's defense is that “the absolute vast majority of the limiting factor is external” — Stefan Aspman, CEO · 2026-10-23, and there is evidence for that—component shortages are industry-wide. But the same sentence could have been lifted from July. The burden is now on Q4 deliveries, where analysts estimate roughly SEK 2.5 billion of DCT revenue versus 1.9 billion in Q3; Aspman effectively confirms that is the plan. AirTech adds a subtler quality question. Its 12.9% margin looks like a breakthrough, and it is: the business has reached double digits faster than management expected. But 2.5 percentage points came from a cancellation fee worth about SEK 50 million, and the underlying margin was 10.4%. The order cancellation was the first of its kind in AirTech, according to management, and they argue it is not a sign of elevated risk. Still, investors should separate the operating improvement from the one-off. The company chose not to treat it as an item affecting comparability, which is defensible but also flattering.

What to watch

Munters is not a broken story; it is a company whose commercial engine has outrun its industrial and cash-conversion engine. The order book says the AI and digitalization demand cycle is real. The P&L says the Virginia ramp is still in startup mode. The cash flow says the working-capital cost of that ramp is real, too. The Q&A reflected that tension: analysts returned to DCT margins and supply bottlenecks, and management had to deny double orders while insisting prepayments remain standard. The next catalysts are concrete: Q4 DCT deliveries, the FoodTech divestment process, and the Capital Markets Day on 11 November. If Q4 delivers the step-up and inventory stabilizes, the record backlog starts to look like earnings rather than a warehouse. If it slips again, the market will stop treating supply chain as external and start treating it as a Munters problem. The new CEO has the right diagnosis; the next quarter is the proof.