CE Clean Eats NYCInspections + ratings, in one place
A bustling NYC restaurant kitchen during dinner service, illustrating the moment a restaurant fails a health inspection

Reader manual · Updated May 6, 2026

When a Restaurant Fails a Health Inspection: A Diner's Manual

You see it on Eater, you see it in the window, you see it in a friend's text at eleven on a Wednesday: the place you go for breakfast every Sunday just got a B. Or worse. Pending. Closed. The instinct is to overreact — to swear off the room you've been eating in for five years and never come back. The other instinct is to ignore it. Both are wrong. When a restaurant fails a health inspection in this city, what you should do is read the receipts. Then decide.

First, an honest framing

About one in five inspections in this city ends in a B or a C on first visit. Tens of thousands of New York establishments end up in the public record where a restaurant fails a health inspection every year, and the lights stay on at almost all of them. A failed inspection is not the same thing as a public health crisis. It is one inspector's report of one Tuesday afternoon in one kitchen, scored against a fifty-three-item code that includes everything from active rodent infestation to a missing thermometer in a beverage cooler. Some failures are catastrophic. Most are not. The work is in telling the difference.

Anyone who has stood on a line will tell you the same thing: kitchens have bad days. The line cook called in sick, the sous-chef hired her replacement an hour before service, the cooler that was cycling fine yesterday started losing temperature at four. A clean kitchen on most days can have one inspector arrive on its worst day. When the failure happens, the first question to ask is which day it was — the bad one or the average one. The full inspection record for any NYC restaurant is searchable at NYC Open Data, going back to 2011.

When a restaurant fails a health inspection: the hierarchy of violations

Not all violations are created equal. Glance at the cited list before you draw any conclusions. The order, roughly, from most to least concerning:

If the failure is mostly facility and documentation issues, the score might still hit B but the actual food-safety question is muted. If pest and temperature issues are on the list, treat the result with the gravity the data deserves.

When a restaurant fails a health inspection, pattern beats incident

One bad inspection is noise. Two in a row is signal. Three is a problem. Pull the full history before you decide.

The directory shows full inspection history per restaurant so you can scan the pattern in one look. We also flag improvement: when a follow-up inspection passes cleanly, we add a "Violations Closed" badge. That second visit is, in many cases, the more meaningful data point.

Recency is everything

A failed inspection from two weeks ago is a different animal than one from two years ago. Restaurants change ownership. Chefs leave. The cooler that was cycling badly in 2024 may have been replaced. The pest problem from 2023 may have driven new contracts and new vigilance. A failed inspection in the past tense is archaeology. A failed inspection in the present tense is news.

One nuance: the city's data has a one- to two-week publishing lag. So if you hear about a failure today, it may already be ten days old, and the follow-up re-inspection may not be in the data yet. Check back in three weeks. The most useful question after a fresh failure is what the next inspection said, and whether the next inspection is in the record yet at all.

What to do at a place you actually love

Most of the time, at a place you've been going to for years, the right move is to keep an eye on it without writing it off. Specifically:

When to walk away

Some failures cross a line where the risk-reward doesn't work, even at a place you'd defend in a fight. At this severity, give it months, not days, before going back. Specifically:

These are not moral judgments. They are statistical signals that the kitchen has structural problems. Restaurants do recover from these patterns; it usually takes new ownership, new chefs, or a renovation. Until you can see the change, don't volunteer to be the test case.

Subscribe to inspections, not just outrage

Most diners only learn about a failed inspection when it makes the news, which means the failure was unusually bad. To catch the smaller signals before they become the news, subscribe to alerts on your favorites and your neighborhood. We email you when a saved restaurant drops below A, when a restaurant in your saved area gets cited for criticals, or when the DOHMH closes one on the block. Free. The point of public data is to put the warning system in your pocket, not to wait for a tabloid to translate it for you.

The right level of vigilance

This city's inspection regime is among the most rigorous in the country. The fact that, every week, somebody fails an inspection is, paradoxically, the system working. Failure is how kitchens get better. Don't punish the place that took its B and earned its A back. Punish the one that has been quietly losing points for three years and has never once felt the need to change. Calibrated skepticism is the move. The data already did the hard work. Your job is to read it.

Look up your favorite restaurant's record →