Methodology

1. How we map an address to a utility

When you enter an address, we run a point-in-polygon lookup against a database of service-area boundaries to find the public water system (PWS) that legally supplies that parcel. The resolution process produces one of three confidence tiers.

Tier 1 — Explicit Boundary (confidence 1.00): the address falls inside a GIS-verified service-area polygon sourced from TEMM boundary data. This is the highest-confidence resolution and covers the majority of urban and suburban addresses.

Tier 2 — Matched Proxy (confidence 0.85): the address lies in a county or municipality for which a unique water system serves the overwhelming majority of parcels, but a precise boundary polygon is unavailable. We match to that system as the most likely supplier.

Tier 3 — Modeled Radius (confidence 0.50): no boundary or county proxy is available. We fall back to a modeled service radius around the nearest known water-system intake point. This is an estimate. Results pages and utility profiles for Tier 3 matches display an explicit disclosure; we always recommend verifying against your water bill and ordering a lab test before purchasing hardware.

2. Where the contaminant data comes from

All contaminant measurements are sourced from EPA's Safe Drinking Water Information System (SDWIS). Utilities regulated under the Safe Drinking Water Act are required to submit monitoring data to SDWIS on a quarterly or annual basis depending on contaminant type. We also ingest supplementary data from state primacy agencies where they maintain more granular or more current records than the federal system.

The data as of date shown on every results page and utility profile reflects the most recent sample date in the record — not the date we last synced with SDWIS. This distinction matters: a result can show a sample dated six months ago if the utility's monitoring schedule is semi-annual.

We report utility-level sampling, not point-of-use measurements. The samples are taken at distribution-system monitoring points selected by the utility, not at individual taps. Contamination introduced by your home's internal plumbing — most commonly from legacy lead service lines or lead solder in older buildings — will not appear in utility data. See the plumbing-limitation disclaimer on /legal for the full statement.

3. What MCL and MCLG mean

These two acronyms appear on every contaminant card throughout the product. They represent fundamentally different thresholds.

A measurement below the MCL is legally compliant, but may still exceed the MCLG. We always show both values so you can evaluate a result against both the legal standard and the health goal. Contaminants with an MCLG of 0 are flagged explicitly throughout the product.

4. How we verify filters

For every filter SKU in the registry, we read the official NSF/ANSI certification listings published by NSF International directly — never manufacturer marketing claims or self-reported performance data. A product's certification page lists the exact contaminant reductions it has been tested for, the standard under which each reduction was verified, and the model numbers in scope.

The NSF/ANSI standards we track, and what each actually certifies:

• NSF/42Aesthetic effects — Chlorine taste/odor, particulates, zinc. No health-contaminant reduction.
• NSF/53Health effects (carbon) — Lead, VOCs, cysts, some metals. The primary standard for pour-through and faucet filters.
• NSF/58Reverse osmosis systems — All NSF 53 contaminants plus nitrate, arsenic, barium, chromium, PFAS, radionuclides, and more.
• NSF/372Lead-free materials — Certifies that materials used do not leach lead — not that the filter removes lead from water.
• NSF/401Emerging contaminants — Pharmaceuticals, herbicides, PFAS (some). Applied alongside 53 or 58.

Materials-only products hold only NSF 42 and/or 372 — standards that verify taste, odor, or materials safety, but make no health-reduction claims. We flag these explicitly in the registry so you can distinguish them from health-certified alternatives. A product that holds only NSF 42 and 372 will never appear in a filter recommendation for an exceeding contaminant.

5. How we rank recommendations

Filter recommendations are produced by a three-stage filter and sort:

1. 1.Certification match — A product must hold verified health certifications (NSF 53, 58, or 401) for every contaminant that exceeds its MCL in your address's utility. Any product missing coverage for even one exceeding contaminant is excluded entirely.
2. 2.Dwelling and form-factor fit — Renters are restricted to pitcher and countertop RO units — form factors that require no plumbing modification. Homeowners see the full eligible set including under-sink RO and whole-house systems.
3. 3.Price — Within the eligible, form-factor-matched set, products are ordered by ascending price so the lowest-cost certified option appears first.

Affiliate relationships never reorder this ranking. A product that is affiliated but ranks below a non-affiliated product on price will appear below that product. A product that loses its NSF certification — or that a new certification filing reveals no longer covers a listed contaminant — is removed from recommendations immediately upon our next data refresh, regardless of any commercial relationship.

6. What we deliberately don't do

• No proprietary risk scores — We do not distill contaminant measurements into a single "water quality score" or star rating. Aggregation of heterogeneous health effects into a single number is inherently misleading and obscures the information a reader needs to make a decision.
• No medical claims — We report engineering measurements and certification data. We do not predict individual health outcomes, diagnose exposure, or suggest that drinking your tap water will cause or not cause any specific condition.
• No scare thresholds — We use MCL and MCLG as our reference thresholds — the values set by EPA's scientific review process. We do not apply third-party advocacy thresholds designed to make more contaminants appear alarming.
• No vertically-integrated hardware sales — WaterQualityLens does not manufacture, warehouse, or directly sell any filtration hardware. We have no inventory position and no incentive to recommend a more expensive product.

7. Frequently asked questions

How accurate is the address match?

Accuracy depends on confidence tier. Tier 1 (Explicit Boundary) uses GIS-verified service-area polygons and is highly accurate. Tier 2 (Matched Proxy) uses a county/municipal record match and is accurate for the vast majority of U.S. addresses. Tier 3 (Modeled Radius) is an estimate — you should confirm the utility name printed on your water bill and consider ordering a Tap Score lab test for certainty.

What if my home uses a private well?

Private wells are not yet in scope. WaterQualityLens relies on EPA SDWIS data, which covers only regulated public water systems (serving 25 or more people or having 15 or more service connections). No centralized municipal data exists for private wells. If you use a private well, we recommend ordering a certified lab test directly.

Do you sell filters?

No. WaterQualityLens is vendor-agnostic. The filter recommendations on /results and /registry are surfaced because those products hold independently verified NSF/ANSI health certifications for the detected contaminants — not because of any commercial relationship. Affiliate links are disclosed per FTC guidelines and never reorder results.

How often is the underlying data refreshed?

EPA SDWIS is updated on a rolling basis as utilities submit quarterly and annual reports. We target a 30-day data freshness window. The 'data as of' date shown on each results page and utility profile reflects the most recent sample date in the record, not the date we last fetched from SDWIS.