The week number that doesn't depend on which app you asked.
Converts calendar dates to ISO 8601 week-numbering dates and back. Gets right the part every ad-hoc script gets wrong: week 1 is the week containing the first Thursday of the year, so late-December and early-January dates routinely belong to a different ISO year than the calendar year.
Built for agents too
The same engine behind this page is available as a JSON API and an MCP server.
curl 'https://iso-week.gumballtools.com/api/v1/run?input=2027-01-01'Worked examples
Questions people actually ask
Why does my spreadsheet say a different week number than this?
Most spreadsheet WEEKNUM functions default to a US-style convention: week 1 starts on the week containing January 1st, and weeks start on Sunday. ISO 8601 uses a different rule entirely: weeks start Monday, and week 1 is the week containing the year's first Thursday. The two conventions can disagree by a whole week number, and around New Year's they can even disagree about which YEAR a date belongs to.
Why is January 1st sometimes "week 52" or "week 53" of the previous year?
Because ISO week 1 is defined by the first Thursday, not by January 1st. If January 1st falls on a Friday, Saturday, or Sunday, that year's first Thursday is January 5th, 6th, or 7th — so the days before it (Jan 1-3, or Jan 1-2, or just Jan 1) belong to the last week of the PREVIOUS ISO year. For example, 2027-01-01 is a Friday, so it falls in ISO week 53 of 2026.
Why is December 31st sometimes "week 1" of next year?
Same rule, other end. If December 29th, 30th, or 31st falls on a Monday, Tuesday, or Wednesday, it starts the week containing the FOLLOWING year's January 4th, which is that year's week 1. For example, 2029-12-31 is a Monday, so it starts ISO week 1 of 2030, even though the calendar year is still 2029.
Does every year have 52 weeks?
No — a year has 53 ISO weeks whenever January 1st of that year is a Thursday, or when the year is a leap year and January 1st is a Wednesday. That happens roughly once every 5-6 years. 2020, 2026, and 2032 are all 53-week ISO years; most years, including 2025 and 2027, have 52.
What is ISO week date notation, like "2026-W53-5"?
It is the ISO 8601 standard's own way of writing a date by week instead of by month: YYYY-Www-D, where YYYY is the ISO week-numbering year (which can differ from the calendar year), Www is the week number 01-53, and D is the weekday 1-7 (Monday through Sunday). This tool accepts that notation directly as input.
Why do you reject dates like "03/04/2025"?
Because it is genuinely ambiguous — read as month/day it is March 4th, read as day/month it is April 3rd, and those land in different ISO weeks. Rather than guess and give a confident but possibly wrong answer, this tool refuses slash-separated dates and asks for the unambiguous YYYY-MM-DD order instead.
Does this handle timezones?
There is nothing to handle — it operates purely on calendar dates (year, month, day), never on a timestamp or a time of day. A date is the same date everywhere; there is no timezone conversion involved in assigning it a week number.
Is this the same as the ISO week numbering used in business reporting or retail calendars?
No. Retail and fiscal calendars often use a "4-4-5" or similar week-grouping scheme with its own start-of-year rule, which is a different system that also gets called "week numbers." This tool implements only the ISO 8601 standard.
Not a general calendar or scheduling tool, and it does not know about business fiscal-year or retail 4-4-5 week numbering, which are different systems that also call themselves "week numbers." It has no concept of time of day or timezone — it operates on calendar dates only. Not for US-style "week of the month" or non-ISO week-start conventions (e.g. Sunday-start week numbering used by some US calendars).