After a Chicago Restaurant Is Found Out of Business, a New Food Business Shows Up in Inspection Records at 56 Percent of Addresses Within Two Years

Chicago health inspectors visit the city’s licensed food businesses, and their findings are public. When an inspector arrives and finds nobody operating under the license they came to check, the inspection is recorded as “Out of Business.” That record is the closest thing the city publishes to a running list of restaurants found out of business, but it has to be read carefully: it is one inspector’s observation on one day, not a legal closure filing. We checked the record against itself before using it: in 7.8 percent of these findings, the same business later passed or failed an inspection at the same address — the “out of business” call was premature or the closure temporary. That is a floor on the finding’s fallibility, since it can only catch the errors a later inspection happens to reveal. Roughly one finding in thirteen, at minimum, travels with every number below. Wherever this page says an address “emptied” or a restaurant was “found gone,” it means exactly this: an inspector recorded that finding, with the error just described.

What counts as a refill — the definition is the analysis

The whole question turns on one definition, so here it is in the open. An address “refills” when a license newly observed at it — the registration Chicago issues each food business at an address — shows up in inspections at that address. Two ways of counting that fail: counting only licenses first seen after the finding date gives 27 percent, because inspectors only discover a closure on their next visit and a fast-moving successor is often licensed and inspected before the “Out of Business” finding is ever written down — those real replacements get missed. Counting any other license inspected after the finding gives 50 percent, and misses the same early arrivals. The rule this page uses starts the clock earlier, at the departed license’s last recorded operating inspection — the last day the old business was observed operating — and counts a refill when a license never observed at the address in records from 2016 onward appears after that day and passes or fails an inspection by two years after the finding. A successor that opened before the closure was even recorded counts, because the deadline runs from the finding and the start runs from that last observed operating day. That gives 56 percent. One more guard: at addresses where a different, pre-existing license kept recording operating results after the departed one’s last observed operating day — a second tenant in the same building, still open — the address never emptied of food businesses, so those 306 addresses are excluded from the count entirely.

The citywide count: 56 percent of emptied addresses have a new food business by two years after the finding

From 2019 through July 2024 — the cutoff that gives every address a full two-year follow-up — inspectors recorded a first Out of Business finding at 3,016 qualifying restaurant addresses. By two years after the finding, 1,685 of those addresses (55.9 percent) had a licensed food business newly observed at the address, and 1,331 (44.1 percent) did not. Every address gets the same deadline, so a space that emptied in 2019 is judged by the same clock as one that emptied in 2024. Stretch the follow-up longer and the share keeps climbing: 65.0 percent of these addresses had seen a successor license by the end of the data on July 28, 2026. How long does the gap between food businesses last? The file cannot say exactly, because it records a closure only when an inspector happens to find it. What it can bound: the median gap from a departed license’s last recorded operating inspection to its successor’s first appearance is 15 months — and the true vacancy is shorter than that gap, since the closing date falls somewhere inside it.

Four of the five areas with the most emptied addresses sit below the citywide rate

Chicago’s statistics divide the city into 77 official community areas, and 33 of them reach this page’s minimum of 25 emptied restaurant addresses for naming an area. In most, the count is too small to separate the area from the citywide rate, so we name only the 9 that clear the screening rule described below. The four that screen below the citywide rate are the Near North Side (39 percent of 232 emptied addresses refilled — the community area containing River North, the Gold Coast and Old Town), the Loop (43 percent of 154), Lake View (47 percent of 207 — home to Wrigleyville and the Southport corridor) and the Near West Side (47 percent of 172, the West Loop’s restaurant row among them). Those four rank first, second, fourth and fifth in the city for emptied restaurant addresses; third-ranked West Town sits too close to the citywide rate to call either way. The most emptied addresses and below-citywide refill go together in these four — a pattern in this data, not a law. The 5 areas that screen above the citywide rate — West Ridge on the far North Side, Portage Park and Belmont Cragin on the Northwest Side, and Hyde Park and Chicago Lawn on the South and Southwest Sides — run from 69 to 79 percent. The inspection file records what happened at each address, not why; rents, landlords, demolitions and redevelopment are all invisible to it, so this page makes no claim about causes.

How we counted

• Source: the city’s food inspections file (4ijn-s7e5), 193,768 inspections of licensed food businesses from January 1, 2016 onward. Retrieved July 28, 2026; the pull was verified complete against the dataset’s own row count. Findings are counted from 2019, so every counted finding has at least three years of address history behind the “first seen at this address” checks.
• A restaurant address: any street address that ever hosted a business the file types as a restaurant — 9,421 of them. Food halls and other addresses with six or more distinct licenses were dropped (444 addresses), because one closure inside a food hall does not darken the storefront. Matches are by street address, not by unit, so at a multi-unit address the successor may occupy a different space in the same building. Refills count any licensed food business, not just restaurants — a bakery in a former taqueria counts as a refill.
• “Out of Business”: an inspector’s recorded observation that nobody was operating at the address under that license. It is not a closure filing. Computed on this data: in 7.8 percent of findings, the same license later passed or failed an inspection there — the finding was premature or the closure temporary. That is the reversal rate later inspections happened to reveal, a floor on the finding’s fallibility, not its total error.
• The refill rule: a license counts as a refill only if its first observed appearance at the address, in records from 2016 onward, comes after the departed license’s last inspection with an operating result (Pass, Pass with Conditions, or Fail), and it records an operating result within two years of the finding. Alternatives move the citywide figure between 27 and 50 percent, chiefly by missing successors that arrived before the inspector recorded the closure; the rule here is the one that counts them.
• Never-emptied addresses: 306 addresses were excluded because a different, pre-existing license recorded an operating result after the departed license’s last observed operating day — another food business was recorded open there, so the address never emptied and cannot answer the refill question.
• The two-year rule: only first findings on or before July 28, 2024 count, and a refill counts only within exactly two years of the finding. This keeps the comparison fair: an open-ended window lets older vacancies accumulate more chances and inflates the rate.
• Mapping addresses: each address is placed in its community area by checking its inspection coordinates against the city’s official community-area boundary file; the handful of addresses without coordinates are left out of the area figures only.
• Naming areas: a community area is named only if it had at least 25 qualifying addresses AND a 95 percent confidence interval on its own rate (normal approximation to the binomial) sits entirely above or below the citywide figure. This is a descriptive screen against the observed citywide rate — which is itself an estimate that includes each area’s own addresses — not a formal test that an area’s underlying rate differs: with 33 areas checked at once, an area that clears the bar narrowly can land on the list by chance, and 4 of the 9 named areas clear it by less than three percentage points. The Near North Side clears it by ten.
• What this is not: a closure rate for restaurants as businesses — a restaurant that moves shows up here the same as one that folds. Nor is it a vacancy measure: the file sees only food businesses, so a former café that becomes a clothing store shows up here as no refill, not as occupied. It is a measure of food-business licensing at the address.

Computed by KCM Desk from Chicago food-inspection records, January 2016 through July 28, 2026; published July 28, 2026. If you spot an error, corrections come first.

Related Keep Chicagoland Moving coverage

• Chicago Licenses 1,000 Fewer New Businesses a Year Than Before the Pandemic. The Gap Is Downtown.
• Data Centers Face Pauses and Bans Elsewhere. Chicago’s Permit File Shows $492 Million in Reported Construction Since 2015.
• Chicago Issued 698 New-Construction Permits in the First Half of 2026. Here’s Where They Went.