Concerned about CRREM?

I am - and if you’re involved in real estate investment, management or decarbonisation, perhaps you should be too.  

The Carbon Risk Real Estate Monitor (CRREM) has become a key part of commercial real estate decision-making, used across real estate in vendor due diligence, valuation discussions, asset management planning, and portfolio decarbonisation strategies.

To date CRREM’s value has been in providing a consistent assessment basis with a shared methodology across differing asset types and geographic locations, enabling comparison for instance, between a UK office assessed by one consultant, and a German office assessed by another.

This comparability and standardisation is now at risk.

Background

The CRREM foundation is moving away from supporting its in-house Excel-based calculation tool, from 1 July 2026, instead providing blueprints that allow service providers – like Cundall - to replicate and build their own equivalent functionality.

My concern is that not all aspects of a sound CRREM assessment are included within the provided blueprints, which can open the assessment up to increased levels of interpretation - specifically the treatment of vacancy and occupancy normalisation when calculating energy use intensity (EUI). By not prescribing a key variable, CRREM may inadvertently introduce inconsistency in the assessment and calculation of misalignment dates. This is the date by which the property is projected to exceed its carbon budget, which until now, has supported its widespread adoption.

The calculation methodology of normalisation matters: a partially vacant building may appear to perform better than it actually does because fewer people are using energy within it. Without normalisation (energy data adjusted for vacancy), reduced consumption may mask underlying operational inefficiency and reduce apparent CRREM misalignment risk. One consultant may choose to use a linear uplift to cover missing data, another may use baseload only adjustment, while another may opt not to adjust at all. While each method can be defensible on its own, the results are not truly comparable.

Why does this matter?

CRREM is no longer a niche sustainability tool as it influences the entire lifecycle of asset planning from vendor due diligence to capital expenditure strategy, investor reporting, and hold/sell decisions. A different misalignment year can change how an asset is perceived, the value it holds, and can affect whether a buyer sees a building as a manageable transition risk or a future liability.  

With the market's direction of travel towards standardisation – this is why this issue feels particularly significant, now. For instance, the UK Net Zero Carbon Buildings Standard (UKNZCBS) has been introduced to create a more consistent basis for assessing whether assets are aligned with net zero outcomes. CRREM has played a similar role for transition risk, so if vacancy and occupancy normalisation were to be left to individual judgement, there is potential to undermine that consistency in advice.

Technical Proposal

Vacancy normalisation is not a minor issue. Buildings have both base load and variable load. Base load is the minimum continuous energy use and may include security systems, emergency lighting, landlord systems, building management systems (BMS), metering, frost protection, and standby plant. Variable load (energy use that changes) on the other hand, may include tenant equipment, occupied lighting, domestic hot water, ventilation demand, and some heating and cooling demand. Assuming that both baseload and variable load energy use correlates exactly with occupancy is inaccurate; normalisation should only be applied to any variable load which changes with occupancy.

For vacancy normalisation, a sensible calculation hierarchy could be:

• Direct measured base load (using actual non-occupancy energy data)
• Measured vacant-period base load (data from when the building was empty)
• Guided benchmark default with disclosure and sensitivity

If the same building can produce different results depending on the consultant or software platform used, the market loses the comparability that made CRREM valuable in the first place. A good CRREM implementation should always show the unadjusted standing date, present any vacancy-adjusted result separately, and disclose the method used.  

CRREM’s next chapter must therefore go beyond preserving pathways and datasets.