Picture this: every morning, before you pour your first cup of coffee, America has already lost enough treated drinking water to fill 9,000 Olympic swimming pools. By the time you go to bed, it will have lost 9,000 more. That water was pumped from aquifers and reservoirs, chemically treated to meet Safe Drinking Water Act standards, pressured through miles of distribution mains — and then it disappeared into the ground through cracked pipes, failed joints, and corroded mains that nobody can see.
According to the American Society of Civil Engineers' Infrastructure Report Card, U.S. drinking water systems lose at least 6 billion gallons of treated water per day — roughly 2.1 trillion gallons per year. There is a water main break every two minutes in this country. And the financial toll is staggering: Bluefield Research estimates that non-revenue water costs American utilities $6.4 billion annually in lost revenue, wasted energy, and treatment chemicals poured into water that never reaches a tap.
This is not a drought problem. This is not a climate problem. This is an infrastructure problem — and it is solvable.
The Pipes Beneath Us
America's drinking water infrastructure was largely built between the 1920s and 1970s. Many of the cast-iron and ductile-iron mains running under our cities and towns have exceeded their expected service lives. The EPA's 2023 national needs assessment determined that the nation requires $625 billion over the next 20 years to bring drinking water infrastructure to a state of good repair — an increase of more than $150 billion over the previous 2018 estimate. That number keeps climbing because we are falling further behind, not catching up.
The American Water Works Association (AWWA) and the Value of Water Campaign have pegged the broader need — including stormwater and wastewater — at $3.4 trillion over 20 years. The ASCE's 2024 "Bridging the Gap" study found a $309 billion gap between current drinking water investment levels and what is actually needed, a gap projected to widen to $620 billion by 2043 if we stay on the current trajectory.
These are not abstract numbers. They show up as boil-water advisories, brown water flowing from kitchen faucets, sinkholes opening over failed mains, and contamination events that erode public trust in the water system.
The Small-System Problem
Here is a fact that should trouble anyone who cares about equitable infrastructure: approximately 92 percent of America's community water systems serve fewer than 10,000 people. These small systems collectively serve roughly one-third of the U.S. population, often in rural communities and small towns where the tax base cannot support the engineering staff, sensor networks, and capital budgets that larger cities take for granted.
A 2025 analysis by Bluefield Research found that small and very small utilities report average water losses above 20 percent of total supply, compared with 16.7 percent for larger systems. The reasons are structural: limited financial resources, shortage of trained operators, aging infrastructure with no condition data, and difficulty accessing federal funding programs that require matching dollars and technical applications these communities struggle to produce.
In other words, the communities least equipped to find and fix leaks are the ones losing the most water. That is not just an engineering failure — it is an equity failure.
The Smart Water Revolution
The good news is that we are in the early stages of a technological transformation that could fundamentally change how water utilities detect, locate, and prevent losses. Three categories of innovation are converging:
Advanced Metering Infrastructure (AMI). Traditional water meters are read once a month — or once a quarter — by a technician walking or driving a route. AMI systems replace these with smart meters that transmit consumption data every hour or even every 15 minutes. This granular data does more than improve billing accuracy. It reveals continuous-flow patterns at individual service connections that indicate leaks, detects abnormal nighttime usage, and identifies step changes that suggest a new service-line break. Vendors like Sensus (now part of Xylem), Itron, Badger Meter, and Neptune Technology Group have deployed AMI across hundreds of American utilities, large and small. When a utility can see every gallon flowing through every meter in near-real-time, apparent losses from meter under-registration and billing errors drop dramatically.
Acoustic and Pressure Sensors. Fixed-network acoustic loggers, deployed on hydrants or pipe fittings throughout a distribution system, continuously listen for the sound signatures of leaking water. Machine-learning algorithms filter out background noise — traffic, pumps, normal flow — and pinpoint leak locations with remarkable precision. Mueller's Echologics platform and Xylem's Visenti system are among the leaders, and utilities like Atlanta's Department of Watershed Management have used acoustic networks to identify thousands of previously undetected leaks, recovering millions of gallons per day. Complementing acoustics, networked pressure sensors detect transients and anomalies across distribution zones, enabling operators to manage pressure to reduce both break frequency and leak flow rates.
Satellite and AI-Driven Analytics. Perhaps the most striking development is the use of synthetic aperture radar (SAR) from orbiting satellites to detect subsurface moisture anomalies indicative of pipe leaks. Companies like ASTERRA (formerly Utilis) process satellite imagery to identify likely leak locations across entire service territories — without a single sensor in the ground. Utilities can then dispatch crews to confirm and repair, dramatically reducing the time between leak onset and repair. Meanwhile, AI platforms are integrating data from AMI, SCADA systems, asset databases, and weather feeds to predict which pipes are most likely to fail next, enabling utilities to replace infrastructure proactively rather than reactively. DC Water in Washington, D.C. has become a national leader in asset-failure prediction, using data analytics to prioritize its capital investments where they will prevent the most damage.
The Federal Funding Window
The timing for this transformation is both fortunate and urgent. The Infrastructure Investment and Jobs Act (IIJA), signed in 2021, dedicated $50 billion over five years for water infrastructure, with approximately $8 billion per year flowing through FY2026 via the Clean Water and Drinking Water State Revolving Funds. The EPA's WIFIA program has made $6.5 billion in low-interest financing available, plus $550 million through the new State WIFIA (SWIFIA) program. These are not grants for studies — they are capital for construction, including AMI deployments, SCADA modernization, pressure and flow sensor networks, and the data platforms that tie them together.
But there is a hard deadline. IIJA's supplemental water infrastructure funding expires on September 30, 2026. The National League of Cities and water-sector advocates are urging Congress to maintain IIJA-level funding, but the President's FY2026 budget request proposed nearly 90 percent cuts to State Revolving Funds and the WIFIA program relative to IIJA levels. If those cuts hold — and if no successor legislation passes — utilities that have not yet applied for smart-water modernization funding will face a dramatically narrower path to financing.
The result is a race against the clock. Utilities that move now can lock in historically favorable financing to deploy technology that will reduce losses for decades. Those that wait may find the window closed.
What "Build It Right" Looks Like
At the EPR Foundation, we believe infrastructure investment should be guided by a simple principle: build it right the first time, and build it smart enough to tell you when something goes wrong.
That means:
Prioritize data before concrete. Before a utility spends $50 million replacing mains, it should spend $2 million deploying sensors and AMI to understand where its losses are actually occurring. Too many pipe-replacement programs are driven by age-based assumptions rather than condition data. Smart water technology lets you target the 20 percent of pipes causing 80 percent of your losses.
Design for small systems. The 92 percent of water systems serving small communities need solutions scaled to their budgets and staffing realities. That means shared-services models where a regional operator manages smart water platforms for multiple small systems, grant programs with simplified applications, and technology vendors willing to offer cloud-based analytics without requiring on-site IT staff. The Bluefield data is clear: small systems lose proportionally more water and have the fewest tools to address it. Federal and state programs should weight funding toward these communities.
Mandate water audits. You cannot manage what you do not measure. AWWA's free water audit methodology gives every utility a standardized framework for quantifying real losses (leaks and breaks) and apparent losses (metering and billing errors). Yet many utilities — particularly small ones — have never completed a validated water audit. Several states, including Georgia, Texas, and California, have begun requiring annual water-loss reporting. This should be a national standard, tied to SRF eligibility. If a utility wants federal money, it should demonstrate that it knows how much water it is losing and has a plan to reduce it.
Think in systems, not silos. The most effective smart water deployments integrate AMI data, SCADA, hydraulic modeling, asset management, and GIS into a unified platform. Isolated sensor deployments produce isolated insights. Connected platforms produce operational intelligence — the kind that lets a utility dispatcher see a pressure drop at 2 a.m., correlate it with acoustic data showing a new leak signature, and have a repair crew on site before dawn.
The Invisible Made Visible
There is something almost poetic about the challenge of water loss. Unlike a pothole or a crumbling bridge, a leaking water main is invisible. The water seeps into soil, runs into storm drains, or simply saturates the ground beneath our feet. We do not see it, so we do not fix it. We lose 2.1 trillion gallons a year not because the problem is unsolvable but because it is underground and out of mind.
Smart water technology changes that equation. It makes the invisible visible. It gives operators eyes and ears in places where, for a century, they have had nothing but guesswork and hope. And it does so at a moment when federal financing makes deployment more affordable than it has been in a generation — or may be again.
The water main breaking beneath your street right now does not care about funding cliffs or budget proposals. It is losing water at the same rate whether Congress acts or not. The question is whether we will use this narrow window to build the intelligent water systems that can find, fix, and prevent these losses — or whether we will let another trillion gallons drain away while we debate the cost of doing something about it.
At the EPR Foundation, we think the math is straightforward: $6.4 billion in annual losses is the cost of doing nothing. Smart water infrastructure is the cost of doing something. We know which number we would rather see on the ledger.
Sources: ASCE 2021 and 2024 Infrastructure Report Cards; EPA 2023 Drinking Water Infrastructure Needs Survey and Assessment; Bluefield Research 2025 U.S. Water Loss Analysis; AWWA/Value of Water Campaign infrastructure needs assessment; EPA WIFIA program announcements; National League of Cities water infrastructure advocacy (2025–2026); Environmental and Energy Study Institute water infrastructure briefing (2026).