Methodology

How we know โ€” and how we don't bluff.

PriceWeather uses sourced, dated, and reproducible data. Every number in every report traces back to a specific public source, with a specific refresh cadence. No anonymous submissions. No hidden algorithms.

โœ“ Last reviewed: September 5, 2026

Our data principles

Every metric on every PriceWeather report is subject to four rules:

Where our data comes from

We use public and authoritative sources only. PriceWeather is not affiliated with, endorsed by, or partnered with any of the organizations listed below โ€” we cite their data with appropriate attribution.

Housing
Zillow Research + Redfin Data Center
Median home prices, median rent, inventory, price-per-square-foot at metro and ZIP level.
Monthly refresh
Insurance
State Department of Insurance filings
Actuarial rate filings from state insurance regulators. Reflects filed rates, not customer quotes.
Quarterly refresh
Crime (official)
FBI Uniform Crime Reporting (UCR)
Reported incidents by category. Sourced from crime-data.fbi.gov and local police disclosures.
Annual refresh
Climate & risk
NOAA + FEMA National Risk Index
Weather patterns, natural hazard exposure, 30-year climate baselines updated annually.
Monthly refresh
Daily costs
BLS Consumer Price Index + retailer scrapes
Groceries, gas, utilities, transportation. Tiered by budget / mid / premium baskets with example items.
Weekly refresh
Schools
NCES + state Departments of Education
Test scores, per-pupil funding, student-teacher ratios, graduation rates at district and ZIP level.
Annual refresh
Life expectancy & health
CDC + state vital statistics
Life expectancy at birth, mortality rates, hospital access, air quality tied to health outcomes.
Annual refresh
Jobs & salaries
BLS OEWS + Lightcast (planned)
Employment counts, median wages by occupation, job posting volumes by metro and industry.
Quarterly refresh
Air quality
EPA Air Quality Index
PM2.5, ozone, other pollutants. Tied to allergy severity indexing.
Daily where available
Taxes
IRS SOI + county assessor records
State income tax rates, local property tax rates, sales tax by jurisdiction.
Annual refresh

How we're different from user-submitted comparison tools

Tools like Numbeo, Expatistan, and Areavibes rely heavily on anonymous user submissions. That approach creates specific problems we structurally avoid:

Reliability

Their approach: anonymous user submissions can be manipulated by anyone with an agenda โ€” a landlord bumping their city's rent numbers, a chamber of commerce boosting a positive metric.

Our approach: sourced data only. Every number traces back to a named public dataset. Community signals (when we introduce them) are visibly labeled separately and never mixed with authoritative data.

Precision

Their approach: vague metric definitions like "meal at a good restaurant" mean different things to different users.

Our approach: precise scope on every metric. Groceries shown in three tiers โ€” budget basket, mid-tier, premium โ€” with example items so users know exactly what they're comparing.

Freshness

Their approach: stale prices from years ago that no one has refreshed, with no visible timestamp on the metric.

Our approach: every metric ships with a last-updated timestamp on the card. Outdated metrics fade visually and trigger a refresh in our data pipeline.

Perception vs. reality

Their approach: the "crime index" is a survey of resident perception, not an actual crime count.

Our approach: official FBI UCR crime statistics and resident-perception scores shown separately. Users can compare what the numbers say to what residents feel โ€” never blended into a single misleading figure.

Local granularity

Their approach: metro-level averages hide material ZIP-level variance.

Our approach: ZIP code is our atomic unit of data where the source supports it. Metro averages exist as a headline rollup only. Property taxes in Mt. Lebanon vs. East Pittsburgh are not the same number, and we don't pretend they are.

Verification and traceability

Every metric on a PriceWeather report can be traced back to its origin. If you ever see a number you'd like to verify:

Ask us to trace any number

Email hello@priceweather.com with the city, metric, and date you're checking. We'll reply within 5 business days with the specific source, the refresh cadence, and the calculation methodology behind that figure.

This isn't a marketing promise โ€” it's a commitment to reproducibility. If we can't trace it, we shouldn't have published it.

Important disclaimer: Forecasts and rankings on PriceWeather are informational estimates based on publicly available data. They are not financial, legal, tax, or investment advice. Individual circumstances vary, and users should consult qualified professionals before making major relocation, housing, or financial decisions. PriceWeather does not guarantee the accuracy or completeness of any metric and is not liable for decisions made based on our content. Full liability terms are in our Terms of Use.

Changes to this methodology

When we materially change how we source or calculate metrics โ€” for example, adding a new data provider, changing tier definitions, or updating refresh cadences โ€” we update this page with a new "last reviewed" date and log the change in our internal audit trail. Subscribers to the weekly Price Weather Report receive an email notification for material changes.

For any questions about our methodology, our data sources, or our approach to a specific metric, contact us at hello@priceweather.com or through the contact form.