Phone number extractor: pull every number out of any text
A phone number extractor scans text for phone numbers and returns each one once in a standard format. Paste text, HTML or a CSV above and pick the default country for numbers written without a + code; you get every number in E.164 with its country. Dates, prices and order numbers are ignored. It runs in your browser.
Numbers written without a + code are read as numbers in this country.
Sample result
Found 3 unique numbers in 4 mentions, across 2 countries.
- Unique numbers
- 3
- Mentions found
- 4
- Countries
- 2
What this means for you
Pull numbers out of emails, pages or notes without copying them by hand or keeping repeats.
- Finds every number in pasted text once
- Skips dates, prices and order numbers
- Each result in E.164 and international form
- Counts how often each number appears
- Extract
- E.164
- Dedupe
By SimpleVerifier Team · Last updated · Runs in your browser; nothing you paste is sent to us
How to extract phone numbers from text
- Pick the default country: the country most of the local numbers in your text belong to.
- Paste the text or upload a CSV, TSV or TXT file. Web page source, email signatures, notes and spreadsheet exports all work.
- Press Extract numbers. Each number is converted to E.164 and counted once, however many ways it was written.
- Download the CSV or copy the E.164 list into your CRM, dialer or the phone number formatter.
Example
- You enter
- A block of text that mentions
(415) 555-0132twice, plus415.555.0133and+44 20 7946 0958, with United States as the default country. - You get
- 3 unique numbers from 4 mentions across 2 countries:
+14155550132(counted twice),+14155550133and+442079460958, each in E.164 and international format. - Limits
- Uploaded files can be up to 25 MB; there is no cap on the number of rows, and it all runs in your browser.
How do I extract phone numbers from text online?
Paste the text above, choose the default country and press Extract numbers: every phone number in the text is listed once, in E.164, with its country, the way it was first written and how many times it appeared. Notes, email signatures, a page's HTML source, chat exports and CSV or TXT files all work, because the tool reads the text line by line, not a particular column.
Text is full of digit runs that aren't phone numbers: dates, prices, order IDs, postcodes. A simple pattern catches all of them. This extractor uses the matcher from Google's open-source libphonenumber, the same library Android uses, which only accepts a run of digits if it is a valid number for its country's numbering plan, at a valid length, in an allocated range.
That makes it strict in a useful way. 2024-10-15 and $1,234.56 are skipped, and so is 555-1234, because seven digits isn't a complete US number. The cost of strictness is that a number with a typo in its area code is skipped too. If you expected a number that isn't in the result, check it in the phone number validator to see why.
Why does the default country matter?
The default country decides how numbers written without a + code are read. A number written with a + code carries its own country and is always read correctly. A local number such as 020 7946 0958 only makes sense with a country: it is a London number if the default is the United Kingdom and nothing at all if the default is the United States. Pick the country most of your local numbers come from.
For text that mixes local numbers from several countries, run it once per country and combine the downloads; duplicates collapse to the same E.164 value. The format rules for each country are on phone number formats by country.
What can't phone number extraction tell you?
Extraction can't tell you whether a number is in service, who uses it now, or whether it is a mobile. Every number in the result fits an allocated range, which shows it could exist and nothing more. In the US and Canada mobile and landline numbers share area codes, so the digits genuinely don't say which a number is; that takes a carrier lookup. This matters if you plan to text the list: most landlines can't receive SMS, and US rules on calling and texting mobiles are stricter, as covered in TCPA rules for mobile vs landline.
How do I extract phone numbers from text in Excel?
A regular expression works for one tidy shape of number. Google Sheets has REGEXEXTRACT, and recent Excel for Microsoft 365 builds include it too. Compared with the extractor:
| This tool | Excel or Sheets formula | |
|---|---|---|
| Numbers per cell | All of them | First match only |
| Dates, prices, order IDs | Skipped: must fit a numbering plan | Matched if the digit count fits the pattern |
| Different countries' formats | Read per country | One pattern for every country |
| Output | E.164 plus international format | The text as written |
| Duplicates | Counted once, however they're written | Need a separate step |
A rough Sheets formula for the first number-like run in a cell: =REGEXEXTRACT(A2,"\+?\d[\d\s().-]{7,}\d"). Check what it returns, since it can't tell a phone number from an order ID.
Who uses this tool, and for what?
- Sales ops
- Reps leave numbers in CRM notes and activity logs instead of the phone field. Export the notes, extract the numbers, and match them back to contacts by E.164.
- Recruiters
- Pull the numbers out of pasted CVs, email signatures and job-board messages in one pass, each in the same format, without retyping.
- Agencies
- Turn a client's directory page, PDF copied to text or old contact sheet into a clean list, then format and deduplicate it before import.
- SMS teams
- Extract numbers from an export, then run them through the landline or mobile checker so you don't pay to send to landline ranges.
What can’t this tool tell you?
- It's strict on purpose: a number with a typo, a missing digit or a missing area code is skipped, not guessed.
- One default country per run. Local numbers from other countries are skipped or misread until you run the text again with their country.
- Text is read line by line, so a number broken across two lines, or written in words, is missed.
- It reads text, HTML, CSV, TSV and TXT. Scanned PDFs and images need to be turned into text first.
- Every result fits an allocated range. That doesn't mean the line is in service or belongs to the person named next to it.
Common questions
- How do I extract phone numbers from text?
- Paste the text into the extractor above, choose the default country for numbers written without a + code, and press Extract numbers. Every valid phone number is listed once in E.164 format with its country, and you can download the list as a CSV.
- How do I extract phone numbers from text in Excel?
- Copy the column, paste it here and download the CSV of numbers. Inside Excel or Google Sheets, REGEXEXTRACT with a digit pattern returns the first number-like run in each cell, but it also matches order IDs and dates and needs a separate step to deduplicate.
- Will it pick up dates, prices or order numbers?
- No. It only accepts digits that form a valid phone number for the country, using Google's libphonenumber numbering-plan data. Dates like 2024-10-15, prices and most order IDs don't match any country's numbering plan, so they're skipped.
- Why are some numbers missing?
- Usually because they are local numbers from a different country than the default you chose, or because the number is incomplete or has a typo, such as a seven-digit US number without its area code. Change the default country, or check the number in the phone validator to see what's wrong.
- Can it extract phone numbers from a CSV or website?
- Yes. Upload a CSV or TXT file, or paste a page's text or HTML source. Numbers are found wherever they are, including mixed into notes or address columns, and duplicates are counted once.
- Can I extract only the mobile numbers?
- Not in this tool, which extracts every valid number. Paste its E.164 list into the landline or mobile checker to split mobiles from landlines where the numbering plan allows it; in the US and Canada that needs a carrier lookup.
- Does it check whether the numbers are in service?
- No. It confirms each number fits its country's numbering plan, which shows it could exist, not that it's connected or who owns it. Line type and carrier come from a carrier lookup, which is part of the paid Email + Phone plan.
Sources
Well formatted is not the same as reachable.
This tool fixes how numbers are written. It cannot tell you whether a line is mobile, landline or VoIP, or which carrier holds it. That takes a live carrier lookup, which is what the Email + Phone plan adds.
- Upload a CSV up to 100,000 rows with an email or phone column
- 5,000 live carrier lookups a month: carrier name and real line type
- Emails in the same file get a real SMTP mailbox check
One plan, cancel anytime, no contract. Sign in with Google and your first 100 email checks are free, no card. The free tools stay free either way.