Asset-level datasets form the building blocks for almost any part of the investment lifecycle, particularly in sustainability integration. Most financial market participants agree the more granular the data the better. This is because the accuracy and reliability of financial analysis increases significantly with better data transparency and traceability, which makes asset level sustainability data one of the most sought after and invaluable insights for investors. However, it is unrealistic to expect perfectly traceable asset level sustainability data across 100% of the coverage universe. There are various reasons some of the asset level sustainability data is incomplete, disparate and sometimes unavailable, including:
Sustainable investing considerations are constantly evolving. There are industry and geographic differences in not only which ESG metrics are relevant, but also if they should be considered relevant today. This means companies in different sectors and geographies look at ESG data in significantly different ways.
Governments, regulators, and investors often prioritize different ESG metrics. This often leads to companies disclosing a minimum set of ESG data that can satisfy demands from all parties.
Specifically for smaller companies, there simply aren’t enough resources available to correctly measure their ESG data so some report partial ESG data which includes only what they have been able to measure or interpret.
ESG guidelines evolve over time as standards organizations and regulators try to make them more relevant and accurate. However, this can introduce inconsistencies over time or between companies using different guidelines. (E.g., location-based vs market-based Scope 2 carbon emissions measurements.
Some companies interpret sustainability differently or deliberately obfuscate sustainability information to appear more sustainable than they are. This is referred to as 'greenwashing'.
The range of demands and the worry of not meeting sustainable targets has led some companies to delay reporting ESG data in the hopes of buying time to make amends. This is commonly known as 'green hushing’
Reported data is difficult to find and often contains gaps. Sustainability practitioners have realized that industry benchmarks and estimations are the next best alternative until reporting improves. In addition to improving coverage, data analytics using reported ESG data vs benchmarks often uncovers anomalies in reported ESG data.
To help investors navigate their way through the abundance of information, the lack of adherence to standards, and the disparities in asset-level ESG data, Preqin provides a range of solutions. The diagram below summarizes the various asset-level solutions - ranging from reported metrics to model-based solutions – enabling clients to utilize the data they find most relevant to their use cases. The below summarizes the various asset-level solutions that Preqin provides. What follows is a detailed description of each of the solutions and the way Preqin has interpreted, applied, and validated the approach.
Asset-level solutions
Fund inclusions: Identify assets that certain types of sustainability fund managers (e.g. impact investors) prefer.
UN SDG alignments: Identify companies that perform activities that could potentially be aligned with UN Sustainable Development Goals.
Controversial industries: Understand if the company exists within an industry or performs an activity that is deemed controversial with respect to sustainability aspects.
Investors sustainability management indicators: Gauge the sustainability disclosure requirements of the company based on the sustainability strategies and priorities of its investors.
Risk and impact - SASB/industry materiality: Identify the most material sustainability risk factors of a company based on the industries it operates in.
Sustainability metrics: Gain insight into 20+ sustainability metrics of a company split across different areas of its value chain.
Real estate metrics: Obtain valuable insight into carbon and energy metrics as well as green building certifications across 80,000+ properties.
Asset-level metrics
In an effort to provide more actionable data at the asset level, we have chosen 15 commonly disclosed metrics for portfolio companies in private markets. With regulation progressing toward more required reporting on certain ESG metrics, we expect these data points to be more readily available in the public domain due to regulation such as the SFDR PAIs. Moving away from estimations will empower Preqin clients to make informed investment decisions based on real, tangible information rather than estimated metrics.
The 15 metrics will be used to develop Sustainability Benchmarks – enabling funds, fund managers, and investors to easily compare sustainability-related information across different locations and sectors, in addition to viewing metrics as normalized vs. unnormalized. The metrics are collected across a five-year reporting period, with metrics being taken from 2020 onward. This aids in developing broader, more comprehensive benchmarks over time, as well as giving our clients insight into the progress portfolio companies are making in their sustainability integration-focused efforts and how they are potentially improving their impact. Having access to readily available ESG metrics, standardized across various industries and locations, will aid our clients in setting sustainable investment goals, tracking competitors, and improving their own sustainability practices across the entire investment lifecycle. Greater transparency into the landscape of these metrics will also facilitate a detailed look at individual portfolio company data and the effect on company performance. Being able to pinpoint genuine peers and alter for data variability will enable highly accurate like-for-like comparisons, therefore minimizing legal risk and ensuring compliance to avoid regulatory penalties.
The 17 metrics we have begun tracking are listed as follows:
Carbon emissions - Scope 1
Carbon emissions - Scope 2
Carbon emissions - Scope 3
GHG Intensity
Total Energy Consumption
Renewable Energy Consumption
Water Consumption/Withdrawal/Use
Total Waste
Hazardous Waste
Total Number of Board Members
Number of Women on the Board
Number of Women in Leadership Positions
Total Number of Full-Time Employees
Total New Hire Employees
Number of Female Employees
Number of Work-Related Injuries
Number of Work-Related Fatalities
Sustainability Benchmarks
Our first edition of Sustainability Benchmarks includes Scope 1, 2, and 3 carbon emissions of portfolio companies within Preqin’s universe. As we continue to collect more data, the benchmarks will expand to 17 quantitative data points. These benchmarks will aid investors, fund managers, and funds in market mapping – giving them a better understanding of where they are overperforming or underperforming in relation to their peers.
Normalized and unnormalized benchmarks are available within Sustainability Benchmarks Solutions to develop benchmarks that are scalable and easily comparable within different types of portfolio companies. The unnormalized benchmarks standardize units across all metrics to enable straightforward comparisons. We standardize the unnormalized benchmarks to tCO2e from many different reported units, such as: tCO2e, MTCO2e, ktCO2e, and million tCO2e. The normalized benchmarks enable relative comparison across assets with large variance in size and revenue. We normalize the benchmarks for all three carbon emissions metrics: Scope 1, 2, and 3 (both market-based and location-based for Scope 2).
Once the units are converted to tCO2e, emissions data is combined with revenue and employee data to create the normalized benchmarks. When dealing with multi-year data, it is crucial to maintain alignment across reporting periods, as revenue and employee figures must correspond to the relevant year to ensure accurate benchmarking and comparison across portfolio companies. Dividing the standardized emissions data with the revenue and employee data enables a comparable look into GHG emissions across an incredibly wide range of assets. This comprehensive approach ensures that the benchmarks remain consistent, relevant, and insightful for all portfolio companies.
Fund inclusions
Our research teams regularly collect and update fund managers’ strategies as well as sustainability fund labels. We are also able to capture the details of deals that are associated with a fund. The fund inclusions criterion simply shows the number and types of funds that are associated with or include this asset in their portfolios.
SDG alignment model
Sustainable development goals (SDGs) are commonly associated with impact investing. Impact investing is broadly used to describe the investment strategy/philosophy of investing with the objective of having a significant and (ideally) measurable impact on various ESG metrics while not overlooking the financial gains.
The most common frameworks for identifying and measuring impact investing in both private and public markets are: the UN sustainable development goals (SDGs), MandG III, and Global Impact Investing Network (GIIN). We utilize the UN SDGs framework in our models.
UN Sustainable Development Goals
The 2030 Agenda for Sustainable Development, adopted by all United Nations Member States in 2015, provides a shared blueprint for peace and prosperity for people and the planet, now and into the future. At its heart are the 17 sustainable development goals (SDGs), which are an urgent call for action by all countries – developed and developing – in a global partnership. They recognize that ending poverty and other deprivations must go hand-in-hand with strategies that improve health and education, reduce inequality, and spur economic growth – all while tackling climate change and working to preserve our oceans and forests.
These are further split into 17 specific goals that all member states will aim to achieve by 2030. The diagram below summarizes these goals.
Preqin’s model is based on the first 16 goals. Our current model does not capture SDG 17 since it is based on the partnerships and initiatives a company is taking. However, we are looking at capturing details of a range of asset level sustainability-related partnerships and might make this available to users in future releases.
Preqin’s UN SDG model endeavors to help investors understand the potential alignments of a company’s business activities with the UN SDGs.
UN SDGs alignment model
This model looks at a company’s industry classification as well as its description (usually from its website) to identify which specific SDGs it aligns to. The diagram below summarizes this in a flowchart. It can be broken down into the following steps:
Step 1: We first check if the company has a primary industry/ business activity that is aligned with an SDG. This is true only for some specific industries such as renewables.
Step 2: Where the company does not have an SDG-aligned primary industry, we further check for sub-industries which reflect the various business activities that the company performs. If a company is found to have a relevant sub-industry, we do not directly label the SDG alignment but proceed to Step 3.
Step 3: Since sub-industries are still quite broad, we further finetune the SDG alignment using a list of keywords that we have carefully curated.
Known shortcomings we are looking to resolve in future versions of the models:
The purpose of these models is to provide a starting point for identifying company alignments. However, this does not currently evaluate the intent or gather evidence on the positive impact of the alignments.
Negative screen – Currently the model tags a company as SDGs aligned based on alignments that omit the other business activities the company performs. In future versions, we are looking at adding a negative screen to exclude companies that are potentially controversial or known to have a negative impact. (E.g. Oil and gas exploration companies with a renewables business division are currently tagged as aligned but will be excluded in future versions of the model).
There might be instances where a company is part of an impact fund, but our models have not been able to identify SDG alignments. This would most likely be due to insufficient information about the company for the model to make a prediction.
Controversial industries mapping
Our research team is constantly up to date with the sustainable investment landscape. Recently, we have come across industries that many ESG practitioners and regulators look at as having a significant negative impact on various ESG parameters. For example, the contribution of the oil and gas industry to climate change. We have carefully curated a list of these industries and mapped them to Preqin’s industry classification system to come up with a controversial industries taxonomy. You can view the taxonomy in the Appendix.
For every company, we try to display all the activities that it performs that might be deemed as controversial by various ESG practitioners.
Investor sustainability management indicators
This is a list of the sustainability management indicators of all the (known) investors in the company. More details on the methodology for management indicators can be found in the Appendix.
Risk and impact solutions overview
What is sustainability-related risk and why is it important?
Sustainability-related risk is an area of the market that has typically been difficult to standardize, because there are many diverse factors that can lead to financial or societal loss. For example, impact risk reflects the impact an event, like the Deepwater Horizon oil spill, has on society, while physical risk manifests in adverse climate events like the recurrence of devastating Malaysian floods. The effect of climate activism or greenwashing on corporate reputation is an example of both regulatory and reputational risk, and even the transition to climate-friendly technologies comes with its own risk to industries that may face loss of business as EVs and renewables gain traction.
The concept of ‘materiality’ in sustainability
Sustainability-related risk can arise from many factors and apply to a wide range of business considerations. However, not all types of risk are relevant for all companies and industries. For example, a company that has no association with any country subject to sanctions might not consider sanctions a relevant risk. One of the more popular ESG-related frameworks that attempts to standardize risk has been built by the Sustainability Accounting Standards Board (SASB). SASB is a non-profit organization that has been part of International Sustainability Standards Board (ISSB) since 2022 and is tasked with building sustainable accounting standards. At the core of the definition of standards by SASB is the concept of ‘materiality’. The concept is influenced by the general concept of ‘materiality’ in finance.
However, the concept of 'materiality' in finance - IFRS - Amendment issued: IASB clarifies its definition of 'material' cannot be directly applied to sustainability-related considerations. Therefore, the use of this term by SASB caused a lot of confusion in the sustainability world as SASB has acknowledged in this blog - Materiality: The Word that Launched a Thousand Debates - SASB (ifrs.org)
Preqin’s interpretation of ‘materiality’
After looking at various research papers and ongoing debates among sustainability practitioners, we realized that materiality could broadly be looked at in two ways:
Financial materiality – This is primarily about understanding the external risks of various sustainability aspects (e.g., floods) and how they are likely to affect a company’s operations (e.g., mining operations). This is where we have attempted to apply the SASB materiality map which provides a very good framework for classifying and understanding financial materiality. More details about the SASB materiality map can be found here – Exploring materiality - SASB (ifrs.org)
Impact materiality – This is primarily about understanding the internal risks that the various operations of a company (e.g., carbon monoxide emissions) have on the people, planet, and economy (e.g., respiratory health).
Financial materiality
Impact materiality
General assumptions used by Preqin while implementing risk models:
We utilize the data that is currently readily available to us, and information about the company that we don’t currently have could potentially include/exclude some of the risks we identify
All the risk assessments are based on these factors relating to the company:
The industries that the company operates in
Where applicable, the latest revenue figures of the company available to us
Industry materiality (SASB)
We utilize the SASB materiality map which divides materiality into 5 dimensions which are further broken down into 26 factors. The materiality map provides a mapping between various industries and the factors that are material for each industry.
We’ve attempted to bring together all the factors from SASB into a graphic that provides a complete picture of the five key dimensions (in green) and the 26 factors as they relate to the five dimensions.
The purpose of the materiality map is to enable users to identify which of the 26 factors would be material based on the industry that the company operates in.
SASB dimensions and factors
To apply these to Preqin’s universe of companies, we have mapped SASB industries to Preqin companies. For every company on Preqin Pro, we show which of the 26 factors are relevant based on the industries that the company operates in. To make it easier for users to pick the E, S, and G risks, we have mapped the five dimensions that SASB provides into three categories.
SASB dimension and Preqin category mapping
SASB dimension | Preqin category |
|---|
Environmental | Environmental |
Business model and innovation | Environmental |
Social capital | Social |
Human capital | Social |
Leadership and governance | Governance |
Real estate carbon and energy metrics
Preqin tracks real estate carbon emissions estimates, energy usage estimates, and green building certificates at the asset-level. Although there is increasing interest in disclosing real estate sustainability metrics, investment professionals often face significant challenges with access to standardized and high-quality, asset-level data. Leveraging Measurabl’s Whole Buildings Estimates data service, we adopt its best-in-class machine learning (ML) models to provide comprehensive and standardized carbon and energy metrics at the granular level. This first-of-its-kind solution increases transparency and accountability for sustainability in private real estate investment.
Introduction to Measurabl
Measurabl, the world’s most widely adopted ESG technology for real estate, is at the forefront of driving sustainable real estate. With a database consisting of 80,000+ buildings and over 18 billion square feet of real estate across 93 countries worldwide, Measurabl leverages ESG data to create investment-grade estimates on sustainability metrics for any building in the world with high accuracy. Through its data services, Measurabl empowers real estate organizations to optimize their sustainability performance, manage physical risk exposures, and accelerate sustainable finance initiatives such as decarbonization.
Introduction to Measurabl’s ML model
Measurabl utilizes its proprietary ML model to generate historical usage estimates of buildings with varying features. These datasets undergo a rigorous data cleaning process to ensure maximum data integrity and quality. The ML model is also subject to a ‘training’ process whereby patterns between usages and input features are learned, allowing the model to estimate usage for any combination of building features. This process enables the model to generate any historical usage estimates for new buildings based on its specified building features. The ML model is used in conjunction with inputs and Measurabl’s database to derive Whole Building Estimates.
How Measurabl uses Whole Building Estimates to calculate metrics
Energy and carbon estimates at the building level are calculated using a three-step process:
The ML model is used to compute energy usage estimates from a trained database. Measurabl’s database draws from 545,000+ monthly data points across 100+ primary use types. These estimates are computed in both log and linear space to minimize the impact of outlier data.
Drawing from similar buildings, the energy mix fraction is calculated for each building. We characterize similar buildings as those that share the same use type and geographical location.
Using both computed energy mix fractions and GHG methodology, the final step involves taking the energy usage estimates to calculate carbon emissions estimates. These estimates show the expected energy consumption and carbon emissions for each building over the duration of the specified historical timeframe.
The ML model consists of:
Building features: By utilizing building features that are universally available across all building use types, we can better understand the variance in usage patterns.
High-level use types: By incorporating standardized, high-level use types such as ‘retail’, ‘residential’, ‘office’, etc., the model can differentiate between subtypes that belong to the same category. In situations where only the highest use type is given, the average is calculated for all buildings that exhibit the least granular subtype.
Granular subtype: The granular subtype may be used to calculate an average for all buildings based on the specific subtype. Granular subtypes separate high-level use types into further categories. For example, if ‘residential’ is the high-level use type, the granular subtype may include ‘residential: multifamily housing’, ‘residential: senior care facility”, etc.
Time series data: Calendarized monthly usages taken from 2017 onward are directly collected from utility bills and aggregated at the building level. Datasets are updated monthly to ensure the data used is the most recent.
For further information on Measurabl’s methodology, please refer to its collateral here.
Next > Fund-level solutions
