Skip to main content
Jump toLoading 0 records…
Section one

Spaces

Zips ranked by 24-month closures against active density, then the recently-vacated storefronts inside them with no current tenant.

Opportunity zips

Higher score = more recent closures in dense areas.

ZipActiveClosed (24 mo)TurnoverScore

Fresh 2nd-gen spaces

Storefronts vacated in the last ~24 months with no active tenant. Ranked by Opportunity Score — freshness, prior-failure count, neighborhood churn vs. city median, and citywide durability of the last cuisine.

How to read this number: A ranking heuristic, not a probability. It orders spaces by four hand-weighted signals and has not been backtested against reopening outcomes. Unlike the Concept Fit Score, it is not a survival estimate.

AddressZipLast operatorPrev cuisineSuggested conceptVacant daysPrior attemptsScoreWhy

Watch one neighbourhood.

This page is the whole city. Pick the area you actually care about and we will email you when a new restaurant licence appears there, and stay quiet when none does. It covers openings only: a closure takes about three years to surface fully in the public record, and we will not send an alert we cannot stand behind.

Section two

Concept reference

The citywide inputs behind the Suggested-concept column above: which cuisines a zip is short of, and which concepts actually last.

Every survival figure in this section is unconditional Kaplan-Meier — P(reach that age at all), not P(reach it given it reached the previous horizon). The conditional form runs roughly 14pp higher at 4 years and further out beyond that, and bare year headers read as absolute.

Whitespace by zip

Cuisines with meaningful shortfall vs city share. Delta = expected active − observed. A space whose suggested concept carries ✦ sits in one of these gaps.

ZipActiveTop gaps (expected − actual)

Best-fit cuisines by 4-yr survival

Cuisine types with the highest 4-year survival citywide. Use as a prior when choosing what to open in a reused space.

CuisineSample2yr4yr6yrMedian days

Restaurant DNA — cuisine × service profiles

What "successful" looks like for each concept, citywide. Ranked by 4-yr survival among combos with n≥15 at that horizon — the same cuisine × service cohort signal validated in the concept-fit lift experiment (AUC 0.62 location-only → 0.73 with cuisine × service). This is the table the Suggested-concept column draws from.

CuisineServiceSample2yr4yr6yrMedian yrsAvg ratingTypical priceConfidence

Confidence = share of this cohort with a real Google price/category read backing its format label, vs. a signal-free cuisine/name-shape default. Fallback-only assignments run 12–48pp lower survival than signal-backed ones for the same label — rows marked Low skew toward "restaurants Google never indexed" rather than the format itself. Read those numbers as directional.