You drive past a restaurant you've eaten at a dozen times and there's a printed notice taped to the glass. The word on it is usually something like CLOSED, SUSPENDED, or PERMIT SUSPENDED — DO NOT ENTER. It looks like the end of the story. Most of the time it isn't. A large share of the restaurants that get closed by a health inspector are back open within a day or two, and a meaningful number of them are cleaner afterward than the place next door that never got closed at all.

That doesn't mean a closure is nothing. It means the notice by itself is a low-information signal, and what you actually want is the context around it: what triggered it, how the place responded, and whether it has happened before. Here's how to read one.

What triggers a closure

Health departments don't close restaurants for a long list of small mistakes. They close them for what most codes call an imminent health hazard — a condition where the department can't reasonably guarantee food coming out of that kitchen is safe until the condition is corrected. The categories are remarkably consistent from jurisdiction to jurisdiction:

Notice what's mostly not on that list: a bad score. In most jurisdictions a low score alone doesn't close a restaurant. It triggers a re-inspection, sometimes a fee, sometimes a conference with the operator. Closure is a separate lever, pulled for a specific set of conditions.

Notice something else: several of those triggers are infrastructure failures, not hygiene failures. A water heater that died Tuesday morning at a spotless kitchen produces the same red notice as a roach problem someone let go for six months. The sign doesn't distinguish. The record does.

Closure vs. suspension vs. voluntary

The vocabulary shifts by jurisdiction, and the differences matter when you're reading a record.

Permit suspension is the formal action. The department suspends the operating permit, which makes continuing to serve food a violation in itself. This is the version that generates a posted notice and, in most places, a public record entry that survives long after the doors reopen.

Closure order or closure is often used interchangeably with suspension in public-facing language, though some departments reserve it for the physical act of shutting the doors during an inspection.

Voluntary closure is when the operator closes before the department orders it — sometimes on the inspector's suggestion, sometimes on their own after a walk-in dies overnight. Some departments log this distinctly and some fold it into the same bucket. Where it's logged separately, it's worth reading as a mildly positive signal: the operator moved first.

Emergency or administrative closure shows up in some systems for hazards outside the operator's control, like a boil-water advisory covering an entire district. In a citywide water event you'll sometimes see dozens of closures on the same date. If you see a closure and the date lines up with a known local utility incident, that's usually what happened.

The practical takeaway: don't read the word. Read the reason code and the narrative attached to it, which nearly every department publishes alongside the action.

How fast a place can reopen

Faster than most people assume. The typical model is fix and call back: the operator corrects the condition, calls the department, and requests a re-inspection. If the department has staff available and the fix is verifiable, that re-inspection can happen the same day or the next business day.

A dead water heater replaced Tuesday afternoon can pass a Wednesday-morning re-inspection. A sewage backup cleaned and sanitized by a remediation crew can clear in 24 to 48 hours. Pest infestations take longer, because the department usually wants evidence of professional treatment plus a follow-up visit to confirm the activity actually stopped — that's a multi-day to multi-week process depending on severity.

Some jurisdictions charge a re-inspection fee, and some require the operator to appear at an administrative hearing before the permit is restored, which adds days. Weekends and holidays add days too, since re-inspections generally happen on business days.

So a two-day closure tells you the fix was mechanical and verifiable. A three-week closure tells you something structural was wrong, or the operator was slow to act, or both. Duration is a better signal than the closure itself.

What the record shows afterward

This is where people get tripped up. In most jurisdictions the closure doesn't disappear when the restaurant reopens. It stays in the inspection history as an enforcement action, typically with the date, the reason, and the date the permit was reinstated. How long it stays visible varies — some departments publish a rolling two or three years of history, others keep everything indefinitely.

What generally does not stay posted is the physical notice on the door. Once the permit is reinstated, the placard comes down and, in grading jurisdictions, a new grade card goes up reflecting the re-inspection result. That's why you can walk into a restaurant showing a clean current grade that was closed nine weeks ago. Both facts are true. Only one of them is on the window.

If you're checking a place after hearing it got closed, the current placard is the least useful thing to look at. The re-inspection report and the enforcement history are the useful things.

When a closure is and isn't a red flag

Here's the honest framework, and it comes down to pattern versus event.

Usually not a red flag: a single closure, infrastructure-related (water, hot water, power, refrigeration failure), corrected within a day or two, at a location whose surrounding inspection history is unremarkable. Equipment breaks. In an older building with an older kitchen it breaks more. An operator who shuts down, fixes it, and gets re-inspected in 24 hours is behaving exactly the way you'd want.

Worth a closer look: a vermin-related closure, even a single one — because pest problems rarely appear overnight, and the closure usually means it reached a level an inspector could see during a scheduled visit.

A real red flag: repeat closures. Two or more within a couple of years, especially for related causes, is the pattern that matters. So is a closure preceded by a run of inspections flagging the same violation repeatedly — that's the signature of a known problem nobody fixed until the department forced it. Same for a closure that follows a suspected illness cluster, which is a different category of seriousness than a broken appliance.

And the counterintuitive one: a restaurant with a long, clean, boring history and one two-day closure is often a safer bet than a restaurant with no closures and a steady trickle of repeat critical violations that never quite escalated. Enforcement action is visible. Chronic mediocrity frequently isn't.

How to check before you go

Pull the history, not the placard. Nearly every county and city health department publishes inspection records online, and the good ones let you see the full timeline for a single location rather than just the most recent result. What you want on screen:

If your local department's site is hard to navigate — and many are, since they were built for compliance rather than for diners — pulling the same records into one readable timeline is exactly the problem FoodCodeScout was built to solve. Either way, the method is the same: read the history, not the sign on the door.

A closure notice is a snapshot of one bad day. The inspection record is the movie. Watch the movie.

Check before you eat

Look up any restaurant's real inspection history — free

FoodCodeScout covers 57,000+ restaurants across NYC, Seattle, and LA with real health-department records. Takes 30 seconds.

Search your city →

About the author — Rick Jenkins is the founder of AI Revenue Forge. ARF builds vertical-specific AI virtual receptionists for service businesses in HVAC, dental, medspa, real estate, home health, credit repair, and pawn shops. Headquartered in Charlotte, NC. Part of Jenkins Worldwide Enterprises.