Methodology

PearlAudit analyzes New York City tax lots. Its zoning and floor-area-ratio (FAR) analysis is grounded in the NYC Zoning Resolution itself, and every rule it reports carries a citation to the governing section of the Resolution. When a dossier says a lot has unused development rights, it names the section the ceiling comes from — so the claim is checkable at its source, not taken on faith.

Flood determinations are made at the level of the individual parcel, using the FEMA National Flood Hazard Layer: a lot is reported as inside or outside a Special Flood Hazard Area based on its own boundary, never a ZIP-code approximation. That same per-lot precision runs through the rest of the analysis — every geographic result is tied to the specific lot in the City’s official parcel records, not a neighborhood average.

The parcel records and most municipal datasets refresh on a monthly cycle, matching the City’s own publication schedule. Recorded documents such as deeds and mortgages update more frequently, since the City publishes them a few weeks after recording. Every dossier and every data surface cites the as-of date of the records behind it, so the freshness of any figure is visible right where it appears. See our data sources page for the full source list.

Most of what a dossier shows is the record itself, passed through with its citation: an owner of record, a recorded deed, a violation, a vacate order, a court finding. We do not editorialize these — statuses render as the agency publishes them, including the agency’s own typos.

A smaller set of figures is computed, and each names its inputs: buildable floor area is the recorded lot area times the cited zoning ceiling; flood coverage is the lot boundary intersected with the federal flood map; carbon exposure applies the law’s own published limits to the building’s own filed energy data. Computed figures are arithmetic on cited records — never a model, never a score.

Where records co-occur — several distress signals on one lot, or common ownership across lots — we report the co-occurrence as a count of individually cited records. The join is ours; every component remains checkable at its source.

Every figure PearlAudit shows traces to a public record. We do not estimate, interpolate, or model a value where the record is silent — absence renders as “no record on file,” never as a guess. If two authoritative sources disagree, we say which one governs. If a rule cannot be resolved for a specific lot, the dossier says so plainly instead of picking an answer.

The same doctrine applies to what a record’s absence means: no open violations on file is a statement about the file, not a certification that a building complies. Some datasets only ever contain buildings that have the condition — for those, a building’s absence is ambiguous, and we treat it as “not evaluable” rather than claiming a clean history we cannot see.

No valuations. No price predictions. No risk scores. No claims about anyone’s intent — a set of recorded facts is context for your own diligence, never a determination that an owner is “distressed” or “assembling a site.” A filed case is an allegation unless a judgement or finding is recorded, and we word it that way. A billed city repair charge is not a statement that a lien was filed, and we word that too.

These are not stylistic choices. The product is built to be handed to an attorney or a lender, and a characterization we cannot cite is a liability we will not print.

New York City only — five boroughs, tax lots. Recorded documents publish weeks after the underlying event, so the newest deed or mortgage may not be visible yet; every surface shows its as-of date so you can judge. Condominium units and co-op buildings have thinner public records at the unit level, and the dossier says so on the affected sections rather than papering over it. Agency records themselves contain errors — a misdated order, a garbled name — and because we render the record as published, you will occasionally see the agency’s error verbatim, flagged with the instruction to verify with the issuing agency.

PearlAudit is a research tool, not legal advice. Zoning analysis, flood determinations, and public-record summaries are produced from official sources with care, but records contain errors, publication lags, and edge cases — verify independently before transacting, and engage licensed professionals (an attorney, architect, or engineer) for any decision that depends on these figures. This is the same limitations notice printed on every dossier, because the document travels without the website.

Where sources disagree on a lot’s area, we use the New York City Department of Finance record — the record the City assesses and bills against.