ENERGY STAR Portfolio Manager Data

Portfolio Manager is the EPA's interactive resource management tool for benchmarking the energy and water use, waste and materials, and greenhouse gas emissions of any building type - and it also serves as Canada's national benchmarking tool. Operators enter property details, meter data and operating characteristics; the tool computes normalized performance metrics, headlined by the 1-100 ENERGY STAR score, which compares a building to similar buildings nationwide adjusted for weather and operating characteristics. Fifty is the median; a score of 75 or higher can qualify a property to apply for ENERGY STAR certification.

Two facts make these records unusually valuable. First, scale: EPA states that nearly 25% of US commercial building space actively benchmarks in the tool. Second, compulsion: many state and municipal benchmarking ordinances require its use, so disclosure-grade data accumulates as a legal byproduct rather than a marketing exercise. What Datadory adds is shape - the same records reshaped as rows keyed on propertyId, ready to join, filter and model. Get a sample of this dataset cut to the property types you follow.

The block above reads top to bottom the way a delivery arrives: identity first, computed metrics second, modules third. A property identified as 1234567 carrying 120,000 square feet posts a weather-normalized score of 82 - comfortably past the 75-point threshold where certification becomes possible - with a site intensity of 45.2 against a source intensity of 98.6. The gap between those two numbers is the generation-and-transmission story: source intensity counts what producing the energy cost upstream, site intensity counts what burned on the premises.

Below the metrics sit the module records - electric-meter consumption, waste diversion, and the no-score alert list where one applies. Multiply the block across a portfolio and you get a panel: the same building measured month over month, comparable building to building because every figure carries the same normalization rules.

Eleven documented fields define the delivered table. Three anchor each observation - the property identifier, the gross floor area that turns absolute consumption into comparable intensities, and the score itself. Four carry the computed performance layer: site energy use intensity, source energy use intensity, total greenhouse gas emissions and tracked water use. The remaining fields hold the meter layer, typed across electric, gas, water, IT, flow and waste categories, plus the waste-module contents and the no-score alert list.

• Geography: the United States and Canada - the tool serves as Canada's national benchmarking service as well - with individual records covering any benchmarked property worldwide that its operator chooses to enter. North America is the core; multinational operators extend it.
• Temporal: rolling monthly and annual reporting periods as entered per property, with historical meter data retained per building. Because benchmarking ordinances keep obligating disclosure year after year, the longitudinal depth of a well-benchmarked asset routinely exceeds what any one-off survey captures.
• Granularity: individual properties and the meters inside them, within one operator's portfolio. There is no aggregate extract of every benchmarked building - the corpus is deliberately bottom-up, which is precisely why shaped, joined delivery beats a raw dump.

API, files, or your warehouse. Daily, weekly, or hourly.

Most teams backfill history once and let new reporting periods rotate into place on the cadence their models expect. Every delivery ships the complete field dictionary above, verified sample rows and the coverage profile mapped to your scope - no reverse-engineering on your side.

• ESG and sustainability teams anchor disclosed figures in the same system raters, regulators and tenants already recognize, instead of assembling a parallel methodology nobody else can audit.
• Acquisition analysts read a candidate asset's score and intensities during due diligence - an 82 against a neighborhood of 60s is a pricing input, not a trivia point.
• Compliance managers treat ordinance-driven benchmarking as both filing and baseline: the same records that satisfy a municipal disclosure requirement track whether retrofits worked.
• Energy services providers size their pipeline by ranking buildings on score and intensity, then prove post-project improvement in the same units they promised it in.
• Researchers and consultants use measured, normalized building performance to characterize commercial stock from the bottom up rather than extrapolating from surveys.

Market researchers and consultants get a measured denominator for building-stock work, normalized consistently across millions of square feet. Competitive intelligence and product teams get the performance context for anything sold into commercial buildings - equipment, software, services. Investors and quant researchers get property-level scores and intensities that drop into real estate factors and transition-risk screens beside financial series. Data scientists and ML engineers get tidy numeric rows keyed on propertyId that join to geography, floor area and time without prose-parsing. Across all four, the constant is normalization: one set of rules applied to every building, which is what makes cross-portfolio comparison defensible.

Within this slice the neighbours answer different questions. The DOE Building Performance Database holds anonymized characteristics for more than a million US buildings - peer-group context rather than identifiable properties. EIA's CBECS is the national sample survey, one record per responding building on a multi-year cycle, ideal for weighting a whole market. The EPA Greenhouse Gas Reporting Program covers facility-level emissions for large emitters including municipal and industrial landfills - industrial scale rather than office-tower scale.

Portfolio Manager owns the middle ground: named, operator-entered records for the ordinary commercial buildings ordinances care about, refreshed as often as their operators report. Run it beside CBECS for representativeness checks, or beside the GHGRP where campus-scale emitters overlap the commercial world.

Notes worth having in hand:

• The score is normalized, not raw - 1-100 compares similar buildings nationwide adjusted for weather and operating characteristics, so two identical-looking bills can earn different scores.
• Ordinances make it infrastructure - many state and municipal benchmarking requirements mandate the tool, and nearly 25% of US commercial building space actively benchmarks in it per EPA.
• Bottom-up by design - records resolve to individual properties and meters; there is no aggregate dump of every benchmarked building, so scope your questions per portfolio or segment.
• Certification threshold - 75 or higher can qualify a property to apply for ENERGY STAR certification, which is why the score column earns its keep in screening models.

Name the property types, geographies and metrics you care about and the sample arrives in exactly the field shape documented above, with the meter and waste modules included where they apply.