Florida Restaurant Inspections, Closures & New Licenses

Florida's Division of Hotels and Restaurants licenses and inspects every public food service business in the state, from restaurants and caterers to food trucks and vending machines, and publishes the records as download files. This actor turns them into filtered rows.

The Division republishes those files about once a week. This actor reads a weekly copy of them rather than the Division's site directly, so rows are as current as the Division's latest publication, and every row carries the date the copy was taken. Pick one of four datasets:

• Inspections: every inspection this fiscal year, with the result and the violations found by category. Updated weekly.
• Emergency closures: restaurants the Division closed, with the reason (such as roach activity) and when they were allowed to reopen. Updated weekly.
• License changes: new food service licenses and approved changes of owner for a fiscal year, with seats, address and phone.
• Active licenses: every active food service license with seats, risk level, last inspection date and phone, and an option to deliver only the licenses that are new since your last run. Updated weekly.

Built for pest control and food safety companies looking for restaurants with recent violations or closures, restaurant suppliers, POS and payment providers, insurers, linen and beverage distributors looking for new and newly sold restaurants, and local news.

What you get

Every row has dataset, licenseNumber, businessName, address, city, zip, sourceUrl, sourceLastModified and sourceFetchedAt, plus the fields of its dataset. Blank values are null; phone numbers are shown as published. sourceLastModified is when the Division published the file; sourceFetchedAt is when our copy of it was taken, so you can see the age of what you are reading.

Dataset	Main fields
inspections	county, inspectionDate, inspectionType, disposition, totalViolations, highPriorityViolations, intermediateViolations, basicViolations, violations (for example "41: † Chemicals/toxic substances (count: 1)")
emergencyClosures	closureDate, closureTime, conditionsForClosure (for example Roach activity), reopeningDate, reopeningTime, district
licenseChanges	changeType (new or changeOwner), approvedDate, licenseeName, licenseType (for example Permanent Food Service, Mobile MFDV, Catering), seats, phone, mailing address
activeLicenses	county, licenseeName, seats, phone, mailing address, status and risk level codes, last inspection date, and changeType: "new" in the new-since-last-run mode

How to use it

Restaurants with serious violations this month:

{
  "dataset": "inspections",
  "counties": ["Orange", "Seminole", "Osceola"],
  "dispositions": ["warning", "emergency order", "administrative complaint"],
  "dateFrom": "2026-09-01",
  "minHighPriorityViolations": 1,
  "maxResults": 1000
}

dispositions matches any part of the result text, without case. The results the Division uses include Inspection Completed - No Further Action, Warning Issued, Call Back - Complied, Call Back - Extension given, pending, Administrative complaint recommended, Emergency order recommended and Emergency Order Callback Complied.

This week's emergency closures statewide: {"dataset": "emergencyClosures", "weeks": 1}. Use up to 12 weeks. Closure files have no county or ZIP column, so filter by cities.

New restaurants and new owners:

{ "dataset": "licenseChanges", "cities": ["TAMPA"], "dateFrom": "2026-01-01" }

changeTypes picks new, changeOwner or both; licenseTypes narrows to, for example, ["Permanent Food Service"]. These files cover a July-to-June fiscal year and have no county column, so filter by cities or zipCodes.

A weekly feed of newly licensed restaurants: use activeLicenses with newSinceLastRun: true, create a key-value store in Apify Console and select it as stateStoreId, then schedule the task weekly. The first run saves the current list and delivers nothing. Each later run delivers only the license numbers that were not on the list before. Add minSeats to skip vending machines and carts.

Counties and cities are not case sensitive, and Miami-Dade and Dade both work. zipCodes takes ZIP codes or their first 3 or 4 digits. Dates are inclusive.

Pricing

Pay per event. There are no platform usage charges on top.

Event	Price	When
actor-start	$0.01	Once per run, after the source answers
record	$0.003	Per row delivered, any dataset

Examples: the 66 Orange County inspections with a warning in the first half of September 2026 cost $0.21. 1,000 rows cost $3.01. A run that matches nothing costs $0.01. A run that cannot reach the data at all, or that refuses it for being out of date, costs nothing.

Set a maximum charge on the run for a hard ceiling; the actor stops cleanly at the cap and never delivers a row it has not charged for.

Good to know

• The source is the Division of Hotels and Restaurants' public records downloads on myfloridalicense.com. Rows are what the Division published, abbreviations included.
• The data is refreshed weekly from Florida DBPR, not read live. Runs read a copy that is re-downloaded on a schedule, which is why results are identical between two runs on the same day. sourceFetchedAt on every row tells you when that copy was taken. If the copy ever falls more than ten days behind, the run fails and tells you the date instead of delivering old rows as though they were current.
• An inspection is a snapshot of the day it happened. The violation list shows counts by the Division's violation categories, not the inspector's written notes.
• The inspection files cover the current fiscal year, from July 1 to the Division's latest weekly update. The license change files cover a whole fiscal year; the 2025-26 files were published on July 9, 2026 and end in June 2026.
• Inspections and active licenses read all seven district files, so a county filter does not make those runs faster.
• The actor is not affiliated with the Florida Department of Business and Professional Regulation.

Questions or a filter that does not behave? Open an issue on the actor page and include the input you used.