Phone Number Extractor
This phone number extractor finds telephone numbers written the way people actually write them — with or without a country code, and broken up by spaces, dots, dashes or parentheses. The difficult part is not finding candidates but rejecting the wrong ones, since any long run of digits looks like a phone number to a naive pattern. Candidates with fewer than seven digits or more than fifteen are discarded, which is what stops years, prices, invoice references and order numbers being reported: fifteen is the maximum length the international numbering plan permits, and seven is the shortest a local number gets. Normalising strips the formatting so you get comparable digits.
How it works
Finds phone numbers written the way people actually write them: with or without a country code, and broken up by spaces, dots, dashes or parentheses.
Candidates with fewer than 7 or more than 15 digits are discarded, which is what stops years, prices and reference numbers being reported as phone numbers. 15 is the maximum length the international numbering plan allows.
Everything runs in your browser — nothing you paste is sent to a server.
0 found · 0 distinct.
Rejecting is harder than finding
Finding runs of digits is trivial. The work is in deciding which ones are phone numbers, because a document is full of numbers that are not:
- Years —
2026 - Prices and totals —
1250.00 - Invoice and order references —
INV-1001 - Long identifiers — a 20-digit transaction ID
The digit-count rule handles all of these. The international numbering plan caps a full number at 15 digits including the country code, and 7 is about the shortest a local number gets, so anything outside that range is not a phone number regardless of how it is punctuated.
Normalising for comparison
The same number can be written a dozen ways. Normalising reduces each to digits with an optional leading plus, so
(555) 123-4567 and 555.123.4567 both become 5551234567 and can be compared
or de-duplicated properly.
What it will miss
Numbers spelled out in words, extensions written as x12, and vanity numbers using letters. There is
no universal grammar for phone numbers, so treat the result as a strong first pass rather than a guarantee.
Frequently asked questions
How do I extract phone numbers?
Paste the text into the left panel. Every number found is listed on the right, ready to copy or download.
How does it avoid matching years and prices?
By digit count. A candidate must have between 7 and 15 digits — 15 is the maximum the international numbering plan allows and 7 is the shortest local number — which rules out years, prices and most reference numbers.
What formats does it recognise?
Numbers with or without a country code, and with spaces, dots, dashes or parentheses as separators. All of these are found: +44 20 7946 0958, (555) 123-4567, 555.123.4567.
What does normalising do?
It strips everything except the digits, keeping a leading plus if there was one. That makes numbers written in different styles directly comparable.
Will it find every number?
No tool can. Phone numbers have no single grammar, and text like "call me on five five five" or an extension written as "x12" will be missed. Check the result against the source when completeness matters.
Is my text uploaded?
No. The search runs entirely in your browser.
Related tools
Email Address Extractor
This email extractor finds every address in a block of text and lists them.
Text Cleaner
This text cleaner runs several tidy-up passes in the right order, in one go.
Regex Tester
A regex tester for matching regular expressions against sample text, with live highlighting and capture group details.