ASCE Infrastructure Report Card - US Infrastructure Grades Data

The 2025 card in four blocks - two national categories, one state rollup, one edition summary - exactly as they land in a delivery:

category: Bridges        grade: C     total_bridges: 623218
condition_good_pct: 44.1    condition_fair_pct: 49.1    condition_poor_pct: 6.8
avg_age_years: ~47         load_posted_structures: 63085
funding_gap_10yr: $373B

category: Roads           grade: D+
category: Ports           grade: B

geography: California     overall_gpa: C-     categories_assessed: 17
bridges_total: 25392      bridges_poor: 1,500+ (6%)
roads_in_poor_condition: ~1/3    roads_shortfall_10yr: $70B    wastewater_needs: $65B

edition: 2025             overall_grade: C     categories_assessed: 18
cumulative_investment_gap_10yr: $3.7 trillion

Three things to notice. First, the unit of analysis is one grade per category per geography - compact by design, and precisely what makes the card joinable onto spending ledgers, permit pipelines and account lists. Second, the condition shares always resolve across good, fair and poor, so a category's story reads in three numbers instead of a paragraph. Third, state rollups carry their own totals - California prints 25,392 bridges with more than 1,500 rated poor and one-third of its road miles in poor condition - so a statewide pitch deck starts from the recipient's own report card, not the national average.

Get a sample of this dataset and we route rows scoped to your categories and jurisdictions the same day.

Six documented fields define the card, each verified against the publisher's own category and state pages. The table below is the complete core; everything else folds out on request.

Geography - the United States nationally, then individually for all 50 states, DC and Puerto Rico: 52 jurisdiction report sets alongside the national card.

Temporal - quadrennial editions since 1998, through 2001, 2005, 2009, 2013, 2017 and 2021 to the current edition released March 2025, whose condition figures mostly reference 2023-2024 inventories. Trend statements inside the card reach further back still - the bridges narrative quotes a 2014 baseline - and interim reports update specific categories between editions.

Granularity - one grade per category per geography, supported by several dozen statistics per category page and per state fact sheet. This is an assessment layer, not a per-asset ledger: it tells you the grade and the gap, and pairs naturally with the inventories underneath.

More on the industry this serves lives on our construction engineering data hub.

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

Name the categories, the jurisdictions and the cadence when you request the sample; the sample ships first either way, with the full field dictionary attached so your engineers can validate shape before anything recurring starts. Because the card moves by edition, most teams take a standing pull timed to releases plus interim updates - others snapshot it annually into a warehouse table and join it to faster-moving feeds.

• Opening market deliverables with an authority quote - consultants and market researchers lead construction-sector decks with the category-by-state grades, because no client argues with a C-minus on their state's roads. The workflow is written up on our market researchers use cases page.
• Citable condition claims - journalists and academics cite the 2025 grades while crediting the underlying federal inventories directly; see our journalists and academics use cases.
• Territory and account prioritization - sales teams rank DOTs, municipalities and engineering buyers by their own state's grades and funding gaps, working the worst-graded markets first; more on our sales growth teams use cases page.
• Infrastructure-spending theses - investors and quants frame the decade ahead with condition and funding-gap statistics across 18 categories spanning seven editions since 1998; see our investors use cases.
• Labeled outcomes for models - data scientists harvest letter grades and gap figures as targets for infrastructure-investment models; see our data scientists use cases.
• Anticipating where competitors chase public money - competitive intelligence teams read the worst-graded asset classes as a map of upcoming funded workstreams; see our competitive intel product teams use cases.

Market researchers and consultants and journalists, academics and students rate this dataset their top relevance: one opens deliverables with it, one cites it. Sales and growth teams and investors and quant researchers rate it close behind - the first works worst-graded territories in order, the second anchors spending theses on quadrennial statistics that predate most alternatives by decades. Data scientists and ML engineers, competitive intelligence and product teams, and developers building data products treat the card as the labeled, quotable overlay on top of faster-moving ledgers.

The card is the opinion layer; nearly everything else in the catalog is a ledger. Value of Construction Put in Place prices what actually got built, month by month, back decades - pair it with the card's funding gaps to separate need from spend. New Residential Construction counts housing units permitted, started and completed, and Chicago Building Permits runs about 845,000 permit rows since 2006 for city-level pipeline work. On the asset side, OSHA Enforcement & Establishment Data and CPWR Construction Safety & Health Data score how the workforce fares while building it. The practical stack: the card tells you where the need is graded worst, the ledgers tell you who is already working there. Head-to-head methodology notes live on the best construction engineering datasets ranking.

Four things worth knowing upfront.

First, a grade is an assessment, not a measurement. Each letter compresses eight criteria into one ordinal value, so regression on the raw letters treats a D-plus-to-C move as meaningful without proving the spacing is uniform - model the condition shares and dollar gaps alongside it.

Second, granularity stops at category times geography. If your analysis needs bridge-level or dam-level condition records, the card is the index that tells you which asset class and state to pull, not the asset table itself.

Third, the publisher's own surfaces occasionally disagree: California's bridge count prints 25,392 on its state page and 25,848 on its fact sheet, and some state grant modules carried placeholder zeros at review. We confirm which figure governs before anything ships rather than silently picking one.

Fourth, the cadence is editorial, not continuous: a new edition lands every four years under the pattern running since 1998 - the next would fall around 2029, though no date is announced - with interim topical reports filling specific categories in between. Tell us your horizon and we scope the standing delivery to match.