Danfoss expects data-center sales to double amid AI boom

2 hours ago 16

Danfoss, the Danish industrial conglomerate, is quietly becoming one of the biggest beneficiaries of the AI infrastructure buildout. The company expects its data-center business to at least double its share of total sales in 2026, driven by surging demand for the cooling systems that keep power-hungry AI chips from melting down.

In 2025, data-center sales made up roughly 7% of Danfoss’s EUR 9.4 billion in total revenue. That figure nearly doubled year-over-year, and the company now projects it could reach around 15% of total sales by the end of 2026.

The cooling problem nobody talks about

Danfoss manufactures oil-free Turbocor compressors and liquid-cooling components designed for both air-cooled and liquid-cooled data centers. One of its flagship products, the Turbocor TGS380 compressor, is used in waterless chiller systems that achieve notably low Power Usage Effectiveness, a key metric for data-center energy efficiency.

PUE is the ratio between total energy a data center consumes and the energy that actually powers its computing equipment. A PUE of 1.0 would mean perfect efficiency. In practice, anything below 1.3 is considered excellent.

The company’s client base includes hyperscalers, co-location providers, and semiconductor manufacturers.

Strong financials backing the pivot

First-half 2026 results showed 15% organic growth, with total sales reaching EUR 5.3 billion. The operational EBITA margin hit 15.3%, reflecting improved profitability alongside the revenue expansion.

North America has been a particularly bright spot. In 2025, the company’s Climate Solutions segment posted 26% organic growth in the region.

Danish business newspaper Børsen reported that data-center sales could scale from approximately 7% to around 15% of total sales as Danfoss invests in capacity and growth opportunities.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.

Read Entire Article