Why location defines the value of waste heat recovery

The European data centre market is approaching a pivotal moment. Energy demand from data centres is projected to reach 98.5 TWh by 2030, driven by continued growth in cloud services and the rapid adoption of AI workloads. At the same time, regulatory expectations are tightening. The recast Energy Efficiency Directive (EED) has moved sustainability from aspiration to obligation, placing waste heat recovery (WHR) firmly on the agenda for operators planning new facilities or expansions. The EED is a law setting targets for EU countries to cut energy consumption and improve efficiency by 2030.

Against this backdrop, waste heat recovery is often presented as a universal solution. A straightforward way to decarbonise data centres while delivering a social benefit. In practice, the reality is more complex. Variations in grid carbon intensity, district heating availability, and capital expenditure (CAPEX) are not uniform between countries. In fact, a data centre in Poland can prevent 2.5 times more carbon emissions than an identical facility in Denmark using the exact same waste heat recovery system. Whether waste heat recovery delivers meaningful environmental value depends greatly on the location.  

Moving beyond traditional efficiency metrics

For more than a decade, Power Usage Effectiveness (PUE) has shaped data centre design. This refers to finding a ratio between the total energy consumed and the energy used by computing equipment. It remains a useful indicator of operational efficiency, but it is increasingly inadequate as a decision-making tool for sustainability. PUE focuses narrowly on the relationship between IT load and total facility energy use, offering no insight into where that energy comes from, what form it takes, or how it might be reused.  

To properly assess the value of waste heat recovery, a broader framework is needed. This framework needs to consider not just energy consumption but also the full system's carbon outcomes. A more useful approach is to assess waste heat recovery through a holistic carbon balance. This considers three interrelated factors:  

• Embodied carbon: The upfront emissions associated with manufacturing and installing the additional infrastructure required for waste heat recovery.  
• Operational carbon: How waste heat recovery alters the data centre’s ongoing energy profile.  
• Displaced carbon: The emissions avoided when recovered heat replaces fossil-fuel-based heating in district heating networks or industrial processes.  

The carbon multiplier effect of location

Our analysis of comparable 30MW facilities across European locations reveals what we call, the ‘carbon multiplier’ effect.  

When modelling a water-cooled facility in Poland, the same waste recovery solution delivered over 2.5 times the carbon impact compared to Denmark. This is because the facility in Poland was where electricity grids are more carbon-intensive and district heating networks frequently rely on coal, whereas in Denmark both the grid and heating networks are already highly decarbonised.  

This has profound implications for site selection and investment strategy. In regions with carbon-intensive electricity grids and heating networks, such as Poland, waste heat recovery systems deliver huge benefits. District heating systems in these areas rely on coal and gas; therefore, using heat from a data centre displaces emissions that would otherwise be unavoidable. In this context, waste heat recovery becomes a powerful decarbonisation tool, enabling data centres to support the wider energy transition and improve the carbon profile of their communities.  

However, in a region like Denmark, where electricity grids and district heating networks are already highly decarbonised, the marginal carbon benefit from waste heat recovery systems is much lower. The same infrastructure could be used at almost a 1:1 level, but it would produce a smaller reduction in emissions because the offset heat is already relatively low-carbon.  

Tackling the energy-water trade-off

For air-cooled data centres, waste heat recovery introduces another important consideration: water use.  

Many air-cooled facilities use evaporative cooling to manage heat, particularly during warmer periods. While this is effective, it can increase water consumption, which has become a particularly sensitive issue in Europe. 

When waste heat recovery is integrated into the cooling strategy, evaporative cooling systems can be used less often during heat export. Our simulations in Ireland and Poland show that waste heat recovery can reduce annualised Water Usage Effectiveness (WUE) by approximately 57%, effectively addressing the long-standing trade-off between saving energy and saving water. WUE, in comparison to PUE looks at the amount of water that is consumed in relation to a unit of IT energy.

A location-led approach to waste heat recovery

The conclusion of these points is that data centres shouldn’t be seen solely as energy consumers. In the right context, they can act as stable, predictable sources of low-grade heat and function effectively within a wider ecosystem.  

In regions that still depend on fossil fuels, data centres with waste heat recovery systems can deliver disproportionately high benefits and accelerate decarbonisation. However, in more sustainably advanced markets, there needs to be greater strategy from operators, local municipalities, and designers to ensure the systems are worthwhile.  

With regulatory pressures continuing and sustainability metrics increasingly incorporated into legislation, operators and developers need to go beyond a one-size-fits-all approach. Waste heat recovery is a powerful tool, but only when operators understand that geography, infrastructure, and carbon context determine whether that power translates into meaningful environmental impact.