Ctrl + Build: The Carbon Code – Why Embodied Emissions Are The Next Big Test For Data Centre Builders

Data centre operators are under pressure to decarbonise – and operational efficiency is no longer enough. The carbon locked into construction materials is now under the spotlight. For builders, that means new expectations, new workflows and a new kind of accountability on site.

Looking Beyond Operational Emissions

Carbon conversations in the data centre world tend to revolve around operations: cutting consumption, switching to renewables and improving performance.

And with good reason. In Australia, data centres already consume around 5% of total electricity. CBRE expects this to climb to 8% by 2030. 

But there’s a quieter carbon story playing out behind the scenes: the emissions locked in before the servers even arrive.

Steel and concrete – two of construction’s most carbon-intensive ingredients – dominate the emissions profile of a typical data centre. Globally, cement and steel account for roughly 15% of all CO₂ emissions. 

More than 60% of the world’s largest companies are now setting net zero targets. To achieve them, they must address all sources of emissions – including those embedded in the construction of buildings.

This puts new expectations on builders and contractors.

Visualisation of components to consider in WLCA for data centres and an approximate breakdown of lifecycle module A-C emissions

Source: Adapted from Arup, 2022. Illustrative only.

Why Embodied Carbon Matters

Embodied carbon emissions are generated during a building’s entire lifecycle – from raw materials extraction to transportation of building products, and from powering construction machinery right through to demolition and disposal.

The World Green Building Council estimates that 38% of global emissions come from our buildings. Operational emissions account for 29%. The remaining 11% are generated during design, delivery or demolition.

As Australia’s electricity grid decarbonises and more buildings are powered by 100% renewable electricity, the proportion of emissions created during construction will grow. The Green Building Council of Australia says as much as 85% of Australia’s built environment emissions will be “locked in” before anyone steps through a door – or switches on a server.

Embodied Carbon Sources Across the Building Lifecycle

Hidden Emissions, Big Impact

When it comes to data centre construction, one big source of emissions stands out: concrete. Concrete alone can represent up to 80% of a data centre’s embodied carbon emissions.

According to analysis by global architecture firm Gensler, reducing the amount of concrete used in a data centre through efficient structural design is the single most effective strategy. This is followed by material quality. High-performance aggregates and optimised mix designs can reduce the amount of cement needed.

But here’s the kicker. Gensler’s analysis suggests many of the emissions linked to supply, transport and rework aren’t captured – because they aren’t tracked.

Cracking the Carbon Code

Some project teams are already embedding embodied carbon tracking into digital workflows. Using Procore, they can:

• Benchmark embodied carbon for materials and major plant
• Coordinate delivery schedules to align with lower-carbon material availability
• Report embodied carbon metrics alongside traditional quality and cost data.

The goal? A single source of truth and a clear audit trail for data centre operators under pressure to prove their ESG performance.

Table 1: Who Holds the Key to Embodied Carbon?

Key Takeaways

• Embodied carbon is the next emissions frontier in data centre delivery
• Constructors and contractors will be expected to track and reduce construction-phase emissions
• Structured digital workflows are essential to capture and report embodied carbon
• Procore enables carbon visibility across materials, trades and timelines.

As data centre operators face growing pressure to limit embodied carbon, constructors and contractors who deliver clarity and accountability through data will earn trust – and win work.

Written by