How much notice did workers actually get?

Updated 2026-09-07 · recomputed on every daily refresh of the 45,775-notice archive.

The WARN Act is a law about a number of days. Every state publishes the notices; none of them publishes the number. So we compute it — for every notice in the archive, back to 1988, every day.

12,901notices gave under 60 daysof 26,003 with both dates on file
1,242,246workers in those noticeswhere a headcount was published
50%share under the 60-day mark1988 to today
1,213same-day noticeslayoff date = notice date
60days — median notice periodacross 39 states
Read this before you quote any number above. This page reports the number of calendar days between the notice date and the layoff date as published by the state. It is not a finding that any employer broke the law and it is not legal advice. The federal WARN Act applies only to covered employers and covered events, and it excuses short notice in defined circumstances — a faltering company seeking capital, business circumstances that were not reasonably foreseeable, and natural disasters. Several states also run their own notice laws with different periods. In several states the published notice date is the day the state received the filing, which can trail the day workers were told by days or weeks — so an interval a little under 60 is often state processing lag, not short notice. A short interval here is a question worth asking, not an answer.

The trend

42.6%20062025
Share of measurable WARN notices filed under 60 days ahead of the layoff date, by year of notice (2006–2025, complete years only — 2026 is still running and is excluded). Recomputed on every daily refresh. · Source: 39 state WARN portals, normalized by this pipeline · as of 2026-09-07

By state

TX (983 of 1,546)63.6%TN (401 of 649)61.8%CA (6,239 of 10,262)60.8%CT (416 of 730)57%AL (505 of 981)51.5%CO (243 of 475)51.2%FL (1,044 of 2,094)49.9%WA (612 of 1,315)46.5%KY (219 of 491)44.6%MD (398 of 1,026)38.8%WI (177 of 463)38.2%OR (438 of 1,172)37.4%
States ranked by the share of their measurable notices filed under 60 days ahead. Only states with at least 200 measurable notices are ranked; the bracket shows the underlying counts. Filing practice and state law differ, so this ranks records, not conduct. · Source: 26,003 notices with both dates, 1988–2026 · as of 2026-09-07

The largest short-notice events in the free window (2024–today)

Ten notices dated 2024 or later, ranked by workers affected, where fewer than 60 days separated the notice from the layoff. Every one of these rows, with its computed notice_days, is in the free CSV below.

CompanyStateWorkersDays notice Notice dateLayoff date
Ideal US Talent Systems Worker OpCo LLCRI9,891572026-05-052026-07-01
United States Cellular CorporationGA4,100562025-04-072025-06-02
Gebbers Farms, EtalWA3,465462025-10-102025-11-25
Meta Platforms, Inc.CA2,212572026-05-262026-07-22
Tyson Foods, Inc (Amarillo B-Shift Operations) UpdatedTX1,761352026-01-202026-02-24
Tyson Foods, Inc. (Amarillo B-Shift OperationsTX1,761552025-11-262026-01-20
Stemilt Ag Services LLCWA1,561462025-10-062025-11-21
Century Blvd., HillsboroOR1,54482025-07-072025-07-15
Tesla, Inc.CA1,452502024-04-252024-06-14
Agrimacs, Inc.WA1,368462025-09-292025-11-14

What we excluded, and why

Of 45,775 notices in the archive, 26,003 carry both a notice date and a layoff date on or after it. Those are the only ones counted above. The rest are excluded rather than guessed at:

That exclusion list is the product. A number you cannot audit is worth nothing, and the states' own files are messier than anyone selling a layoff feed will tell you.

Get the data

Free — 7,365 notices dated 2024 onward, with the computed notice_days column, as CSV. No account, no email, no rate limit.

Free — the whole-archive summary, 356 state-year rows back to 1988: how many measurable notices each state published each year and how many of them fell under 60 days (notice_period_by_state_year.csv). Aggregate counts only, so it is free at full archive depth. The share column is left blank for any state-year with fewer than 30 measurable notices — a percentage over a handful of records reads as a fact and is not one.

$199 one-time — the full archive: all 45,775 notices back to 1988 with notice_days computed on every row, plus the 9,984 flagged reversed-date records so you can audit them yourself. Same file the charts above are built from, refreshed daily for a year.

Get the full archive — $199 one-time Download the free CSV

Also published as a Hugging Face dataset with a full method card: us-warn-act-layoff-notice-period-days (same rows, rebuilt by the same daily run).

CC BY 4.0 — credit "WARN Feed". Refunds: full refund on request within 30 days, no questions, via the Gumroad receipt. Privacy & terms.

Method

notice_days = effective_date − notice_date in calendar days, computed per notice from the dates the state published, with no interpolation and no imputation. Notices are counted by their own notice date, so a year's figure never moves once that year is complete except when a state back-fills a record. Both dates ship in the free dataset; this page adds the arithmetic, the exclusions and the daily rebuild.

This page reports the number of calendar days between the notice date and the layoff date as published by the state. It is not a finding that any employer broke the law and it is not legal advice. The federal WARN Act applies only to covered employers and covered events, and it excuses short notice in defined circumstances — a faltering company seeking capital, business circumstances that were not reasonably foreseeable, and natural disasters. Several states also run their own notice laws with different periods. In several states the published notice date is the day the state received the filing, which can trail the day workers were told by days or weeks — so an interval a little under 60 is often state processing lag, not short notice. A short interval here is a question worth asking, not an answer.