The urgency of decarbonization needs to be matched by rapid decision-making frameworks that set realistic performance targets and identify optimal pathways towards meeting these targets.

If you’re responsible for a large portfolio, different buildings in your portfolio are likely designed for different programs and operating characteristics, have different shapes and sizes, operate in different climates, and are connected to different utility grids. You already know that there isn’t a one-size-fits-all solution to determine which strategies will have the biggest impact.

Most building energy consultants will treat your portfolio as an extrapolation of a few typologies – each formed by grouping buildings based on their program types, location, or date of construction. They manually develop energy models that represent each typology and then simulate future energy scenarios to identify target performance levels. This method doesn’t account for individual building nuances, and thus cannot inform actionable building-level implementation plans or offer realistic performance targets for your portfolio. The method is simply a misguided attempt to apply to an incredibly time intensive process developed for one-building-at-a-time analysis to an entire portfolio.

There are also benchmarking systems, such as Energy Star Portfolio Manager, which use regression analysis to rate energy performance based on historical energy use and other variables such as building type, size, location, and date of construction. Some of these benchmarking systems also provide high-level guidance on the impact of efficiency measures - also based on regression analysis. This approach can work sometimes – usually when you’re your building is close to what might be considered “average” for a given typology - but it doesn’t capture the nuances of individual buildings and isn’t flexible enough to modify if needed.

Our old office building near Bryant Park, in New York, was a perfect example of the pitfalls of conventional benchmarking systems. The building had steam heating, drafty windows, outdated lighting systems, and inefficient ventilation. However, it received an Energy Star score of 100 (the highest score possible),and an “A” for efficiency under Local Law 33 - New York’s program for assigning energy grades to buildings. The most likely explanation is that the building was only partially occupied during the period when data was collected, and the facilities management team had curtailed energy use (rightly so) in unoccupied portions of the building. One could argue that, compared to its peers, it was in fact more efficient, but only because of operational efficiency and not because of the underlying characteristics of the building. This erroneously leads one to believe that there are limited opportunities for improvement, when in fact there are tons of way to improve the building and reduce energy use. Traditional benchmarking systems can’t capture this type of nuance, but Carbon Signal can, by allowing a wider set of inputs, such as operational characteristics, that can be adjusted if needed to accommodate real-world conditions and constraints.

How Carbon Signal is different

Carbon Signal is founded on the hypothesis that we can bridge the gap between assessment and implementation by assessing every building based on its own individual characteristics and its own specific performance targets, with its own set of optimal strategies in a fraction of the time, and at a fraction of the cost, of traditional methods. In other words, rather than benchmarking a building against its peers, we believe that buildings should be benchmarked against the best version of themselves, taking into consideration the operational nuances and programmatic constraints that are unique to each individual building. Furthermore, all the variables at play – the underlying physical and operational characteristics of the building – should be fully customizable to create a higher level of fidelity in the analysis and more detailed recommendations on how to improve building performance.

The engine that powers Carbon Signal works by quickly creating auto-calibrated energy models for each building, rapidly evaluating a full suite of energy and carbon interventions, and intelligently determining high-priority strategies for each building to inform portfolio-wide decarbonization targets and pathways. Carbon Signal offers a decision-making framework that relies on a fast, directionally accurate, and scalable approach that combines traditional building energy modeling with machine learning techniques, to help rapidly establish performance targets and prioritize efficiency strategies for any building, in any location, on an ongoing basis. Traditional benchmarking methods are fine as a first pass, but they can only take you so far. With Carbon Signal, you can quickly get a targeted set of high-impact strategies that you can carry forward through to implementation.

‍