How the Zip Score works

What it measures

The Zip Score rates a New Jersey municipality across nine measured pillars: household prosperity, school quality, safety, amenity access, downtown & street life, parks and open space, highway access, rail commute, and market momentum. Every input traces to a public source — Census ACS, NJ DOE, FBI crime data, GTFS transit schedules, county tax records, Economic Census, and OpenStreetMap/Overture places data. No third-party ratings, no editorial opinion, no hidden inputs.

The pillars are not weighted equally. Pillars that tend to move together statewide (like transit and highway access, both forms of "getting around") are grouped into a theme, and the theme counts once rather than double-crediting two versions of the same idea.

What actually drives the score

Here is what each theme is really worth in the final number today — not the internal planning target, the number the score actually produces once every pillar's weight is accounted for:

Market momentum (recent price/sales trend) is tracked and shown on every card, but carries no weight in the score.

Household weighting

A town's score is the average of its Census block groups' scores, weighted by how many occupied homes each block group actually has — a block group with 900 households counts far more than one with 20, so a handful of largely-empty commercial or industrial blocks can't quietly drag a town's real, lived-in neighborhoods around.

Reading the number

The score is shown on a familiar 0-100 scale, where 100 is New Jersey's top-ranked town and every other town is shown proportionally below it. Every ranking, percentile, and "value" read is computed on the underlying raw composite before that display stretch — the 0-100 number is a presentation choice, not a different calculation.

These tier badges are an experimental, secondary read layered on top of the score — they never change the number itself.

When a score is withheld

Some municipalities don't get a numeric Zip Score at all, and we say so plainly rather than guess:

• Too small to rank — under 500 residents. A handful of block groups at that size can't produce a statistically stable comparison against towns hundreds of times larger. We'll still share whatever pillar-level facts genuinely exist (schools, safety, transit, etc.) as context — just not a single composite number.
• Insufficient data — fewer than 5 of the 9 pillars have real data for that town (most often no local police department reporting crime data, or Census income data suppressed for privacy at a very small sample size). We name exactly which pillars are missing and why.
• No geographic match — a small number of tiny enclave municipalities have no Census block group that resolves to them at all, so there's no underlying data to score.

Scored vs. context metrics

Not every number on a town's card counts toward its score. Market momentum, downtown-depth detail, retail sales per resident, and several demographic/trajectory signals are shown as honest context — real, sourced facts that help explain a place — but they never move the Zip Score itself. Every context metric is labeled as such.

Borrowed access

Parks and amenity access (dining, culture, everyday retail) count wherever they physically sit, not just inside a town's own borders — a park or restaurant a five-minute drive away genuinely serves nearby residents regardless of which side of a town line it's on. We separately disclose what share of that credit is "borrowed" from a neighboring town versus inside the town's own boundary, so the number is useful without being misleading. Downtown & street life is the one exception: it measures a town's own identity, so only what's physically inside that town's own borders counts.

Limitations, honestly

• Data vintages vary by source — Census figures are a 2020-2024 blend, school data is 2024-25, crime data is the most recent full FBI reporting year (2023). Nothing here is real-time.
• Small towns are inherently noisier than big ones — a handful of incidents or a thin sales sample moves a small town's numbers more than the same events would move a city's.
• Safety is measured municipality-wide — NJ crime data has no neighborhood-level resolution, so one number describes the whole town even where real safety varies block to block.
• This is a quality-of-place score, not a life-outcome guarantee, an investment recommendation, or a measure of who lives somewhere. It says nothing about any individual property, and it is never used to rank or screen towns by the demographic makeup of who lives there.