Karnell's Record Quarter Arrives With an Asterisk Taped to It

EBITA up 73%, a 20.6% margin, and a management team that spent the whole call explaining why none of it should be annualised — while EPS fell anyway.

KARNEL-B.ST · Earnings Call · 2026-10-23

The best quarter in the company's history, and the least repeatable

Karnell Group does not do macro. While the market's curated keyword board this quarter is dominated by Fed rate hike pricing, tariff refunds and the long unwind in AI data centers, the Swedish industrial roll-up's call contained almost none of that weather. Its story is entirely self-inflicted: net sales up 33% to SEK 582 million, EBITA up 73% to SEK 120 million, an EBITA margin of 20.6% against 15.9% a year earlier, and a rolling-twelve-month margin of 16.7% that management notes has climbed from roughly 11% at IPO. Nine-month EBITA of SEK 286 million was up 68%, with 41% of that organic. That is a genuine operating result. It is also, by the CEO's own repeated insistence, a mirage in part. “Q3 is also our seasonality's strongest quarter, so the results should not be seen as a new normal level.” — Petter Moldenius, CEO · 2026-10-23 The keyword new normal level was put in the deck specifically to be denied. Twice in prepared remarks, once in the takeaways, and once again in Q&A on seasonality, where Petter Moldenius conceded the historic pattern has flipped: Q4 used to be the peak, and for roughly three years Q3 has been, with Q1 the trough. What reads on the headline as an inflection is, in management's framing, a calendar effect layered on top of two temporary ones — larger deliveries and price adjustments that landed ahead of cost increases in the product companies.

One customer, one third of the organic growth

The most useful disclosure on the call was a number the company volunteered rather than one an analyst extracted. A single unnamed customer — flagged in prior quarters and still flagged now — drove about SEK 10 million of organic EBITA in the quarter, roughly a third of the group's organic EBITA increase. That is down from a larger share in the first half, which is the good news: “the growth is becoming more broad-based” — Petter Moldenius, CEO · 2026-10-23. The bad news is that the concentration has not gone away, only diluted, and Moldenius was explicit that “these orders have short lead times, and our visibility is limited, and we still don't consider these demand levels sustainable in the long term.” — Petter Moldenius, CEO · 2026-10-23 This is the crux of the quarter. The broad based growth language is real — both segments delivered, with product companies at 22.8% margin and the niche production companies at 23.6%, up from 18.4% — but the single-customer effect sits inside the niche production line, which grew organic sales 28% and organic EBITA 57%. Strip out the temporary delivery and pricing timing and the temporary effects and you have a good but far more ordinary business, which is precisely the point management kept making. Seasonally strongest quarter, one-off pricing timing, one concentrated demand pocket: three reasons not to extrapolate, all disclosed proactively. That is unusually candid housekeeping, and it is the reason to trust the number rather than distrust it.

Part of the result is temporary, and Q3 is seasonally our strongest quarter. This should not be seen as a new normal level, but the overall underlying development is good.
Petter Moldenius, CEO · 2026-10-23

What management quietly stopped doing

The more interesting absence is on the acquisition side. Karnell's entire model is buying small industrial technology companies — 20 of them, 860 people, across Sweden, Finland, the UK and Italy — with a decentralized model and a lean central team whose job is M&A and capital allocation. Yet no deal has closed since April, when OBA in Italy and flex7 in the UK were added. Acquisition costs in the quarter fell to SEK 0.9 million from SEK 1.9 million a year earlier, which Moldenius used as evidence of work done and not spent. Asked directly whether the market is too expensive, he declined the premise and instead described disciplined on valuation as the reason: “we had roughly SEK 1 million in acquisition cost during the quarter, which is the results of us doing quite some extensive work on a company and in the end choose not to execute.” — Petter Moldenius, CEO · 2026-10-23 For a serial acquirer, a six-month deal drought is the story the tape cares about, because the equity case is compounding by acquisition. The mitigant is balance sheet capacity: net debt excluding IFRS 16 fell to SEK 537 million, leverage dropped to 1.4x from 1.9x in the prior quarter, well inside the stated two-times target, and interest-bearing debt was paid down to SEK 726 million from SEK 771 million. Rolling-twelve-month operating cash flow of SEK 387 million, up 121%, and cash conversion of 111% of EBITA mean the gun is loaded. Sell-side scepticism about price is the tension; the firepower is the answer, unproven until it is used.

The accounting friction nobody led with

Two items will decide how this print reads on a screen. First, EBITA rose 73% but earnings per share fell. The bridge is entirely below the operating line: a roughly negative SEK 16 million currency revaluation on external euro-denominated loans, plus about SEK 9 million from revaluing put and call options — which, awkwardly, move against reported earnings precisely because the underlying companies are performing better than expected. Second, Karnell changed its return on capital employed definition this quarter to EBITA over average capital employed excluding cash and acquisition-related liabilities. The new definition produces 16.4%, up from 14.7% last quarter and 12.5% a year ago; on the old basis it would have been 16.0%. Management disclosed both, and the gap is small, but a definition change landing in the same quarter as a record margin and a 73% EBITA jump is the kind of coincidence a sceptical reader should price, however benign the explanation. Strip it all back and the honest read is this: Karnell had its best quarter, grew organically at a rate most European industrials would trade a year of guidance for, converted cash better than it ever has, and simultaneously told you not to believe the level, not to assume the customer base is stable, and not to expect a deal every quarter. The market's big themes — tariffs, rates, AI capex — barely graze a portfolio of niche industrial technology assets sold into mixed Nordic, UK and Italian demand. That isolation is the investment case and the risk in one: nothing external is lifting this, and nothing external is about to save it either.