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

GEA Found a Product Cycle to Replace Its Trophy Order

Continuous Tableting turned a EUR 103 million order into a new flagship narrative, while the Baladna era and the tariff debate quietly left the stage
G1A.DE · Earnings Call · 2026-11-09

The batch plant's bad week

GEA's third quarter looked like another instalment of the same story — all KPIs up, guidance confirmed — but the cast changed. For roughly two years the company's earnings calls had a headline order: Baladna, the Algerian dairy megaproject that management kept promising would book "in Q2 or Q3" and then kept deferring. That order is now just a comparison base. What replaced it is a product, not a project: tablet pressing, or more precisely Continuous Tableting, which jumped to the top of GEA's extracted keyword list this quarter after barely registering a year ago. The mechanics are simple enough to explain to a non-engineer, which is exactly why management spent minutes on them. Traditional tableting runs eight sequential steps in eight separate rooms to avoid cross-contamination. GEA's system is powder-in, tablets-out, with in-line quality control. Stefan Klebert's pitch: more than 90% less production time, up to 40% lower cost per tablet, up to 30% less factory space, and more than 60% off the time from development to commercial scale. The order book agrees: GEA booked a large order volume of more than EUR 100 million for the technology in the quarter, part of a EUR 295 million total large-order haul against just EUR 64 million a year earlier. He called it, twice, "a game changer."

There are actually no peers yet because there is no other competitor. But this is definitely something where we will make sure that also a part of the savings will end in our home, which is not more than fair if you bring this kind of really, really great value to the customers.

Stefan Klebert, CEO · 2026-11-09
That is the rare thing in industrial machinery: a genuine technology monopoly, however temporary, in a market management believes is substitutable in full. "We have, at the moment, no indication that you could not change all the tablets to this continuous system," Klebert said when Akash Gupta of JPMorgan pushed on the addressable market. The company is already doubling capacity and building an additional factory in Belgium. And on the Mission 30 target of EUR 50 million to EUR 150 million of Continuous Tableting revenue by 2030, the CEO conceded: “From today's perspective, this might be too conservative.” — Stefan Klebert, CEO · 2026-11-09 CEOs say that about their own targets rarely, and almost never about a new product line.

The question management would not answer directly

Sebastian Kuenne of RBC asked the sharpest question of the call, twice: if GEA wins the tablet business, does it lose volume on the processed-food side as GLP-1 migrates from injection to a daily oral dose? Klebert's answer was a shrug dressed as optimism — customers do not always disclose what they will press, and anyway the tablet universe is vast. “It's not about a new thing like GLP-1 that might come on top and might have a positive impact. But once again, think about the huge amount of tablets which are in the market.” — Stefan Klebert, CEO · 2026-11-09 The honest read is that GEA has sold its investors an asymmetric bet on GLP 1 without owning up to the symmetry. If oral GLP-1 becomes mainstream, tablet volumes rise — good. If it cannibalises food processing throughput, GEA sells fewer dryers and mixers — bad, and unquantified. What the call does establish is that the pharma industry's conservatism, not engineering, was the historical bottleneck: quality consistency at the first processing step is what kept continuous tableting on the drawing board. That is now solved, and the first commercial lines are running.

What quietly left the stage

Two themes that dominated prior calls are now near-silent, and both are informative. First, data center exposure. In August, Citi's Klas Bergelind asked directly whether GEA could commercialise into the AI build-out. Klebert's reply was refreshingly blunt: “we don't expect any significant business from us for the -- in the data center.” — Stefan Klebert, CEO · 2026-08-10 Nine months on, that absence is not a problem. The global price tape shows data-center-adjacent equipment names among the heaviest 90-day decliners — co-packaged optics, high-bandwidth memory, data center revenue themes all down sharply. Sitting out the theme has cost GEA nothing but has cost its peers plenty. Second, tariffs. GEA's 2025 calls were dominated by the 15% tariff debate; the H1 2026 call disclosed mid-single-digit-million euro refunds, roughly half passed to customers. This quarter the word barely surfaces. That is remarkable given that tariff refund was the market's single hottest keyword in 20262 and remains a live earnings tailwind for a long list of industrials. GEA has essentially engineered the issue out of its P&L — contracts push duties to customers, and the CFO's guidance for 2026 was confirmed without a tariff asterisk. And Baladna? The Algerian project that once dominated the pipeline discussion is now only a comp: last year's Q4 contained a large order of EUR 140-170 million, which is why Klebert pre-emptively warned that Q4 2026 "will definitely not" repeat it. Baladna has served its purpose — the backlog is 20% higher year-over-year at EUR 3.7 billion, and management argued that alone makes 2027 growth "almost unavoidable."

The margin plumbing, and the one soft spot

The quarter's profitability was clean: EBITDA before restructuring up 12.6% to EUR 261 million, a record 17.9% margin, ROCE above 37%, net cash of EUR 115 million after swinging from net debt a year earlier, and trailing four-quarter free cash flow of EUR 566 million at a 61% cash conversion rate. The EUR 500 million buyback is 12% executed; 1.4 million shares were repurchased at an average EUR 66.35, already 4.5% below Friday's close. The blemish was Farm Technologies, where the margin fell to 16.4% from an exceptional 18%. The CFO attributed it to a mix effect and a tough prior-year comparison, and the service share actually rose 1.2 points to 49.6% — a good problem to have, but one that pushes the burden of proof onto the second half. There was also EUR 13 million of restructuring in Pharma & Food Applications tied to a footprint change, a reminder that from 2027 restructuring moves above the EBITDA line. On that, the prior call's guidance stands: “Next year, 0 because we don't exclude it any more next year.” — Stefan Klebert, CEO · 2026-08-10

The verdict

GEA is a mid-cap industrial that has spent years asking investors to trust a Mission 30 corridor, and this quarter it can finally say the quiet part: “we will be already in 2026 there” — Stefan Klebert, CEO · 2026-11-09 — hitting 2030 targets four years early, with 2027 set to be the second consecutive year inside the range. What makes this report more than a guidance confirmation is the narrative swap underneath it. A company that used to tell a story about winning one enormous project now tells a story about owning a manufacturing technology nobody else has, with better margins, a product cycle rather than a lumpy tender calendar, and a capacity expansion already underway. The gap between those two stories is where the risk lives: GLP-1's effect on food processing remains unquantified, the Continuous Tableting revenue base is still small enough that "too conservative" is cheap to say, and Q4 will not flatter on orders. But an industrial claiming a monopoly with no peers and a capacity build in Belgium is a different animal from the one that spent 2025 apologising for tariffs.