E.164: The International Phone Number Format

Phone numbers look like data but behave like text. The same subscriber line can be written a dozen ways, all of them correct to a human reader and all of them different to a database. E.164 exists to collapse that ambiguity into a single canonical form.

How an E.164 number is built

Three parts, in order:

• The plus sign. Notation, not a digit. It marks the string as an international number so no dialling prefix has to be guessed.
• The country code. One to three digits assigned by the ITU. 1 covers the United States, Canada and a number of Caribbean territories; 44 is the United Kingdom; 91 is India.
• The national significant number. Everything else — in the US, a three-digit area code, a three-digit exchange code and a four-digit line number.

Total length after the plus is capped at fifteen digits. There is no minimum defined by the standard; national numbering plans set their own lengths, which is why a valid E.164 number from one country can be materially shorter than one from another.

Why it matters for a calling list

The practical value is deduplication and lookup. A purchased lead list assembled from several sources will contain the same subscriber written several ways, and a naive duplicate check treats them as different people. You pay to process the same number more than once, and your agents call it more than once — which is both wasted spend and, if the contact has asked you to stop, a compliance problem.

Normalising to E.164 first makes the duplicates collapse. It also makes every downstream check work: carrier and line type lookups, litigator database matching, and suppression lists all key on a canonical number. A list that has not been normalised will return misses that look like clean results.

Where normalisation goes wrong

• Assumed country codes. A ten-digit string with no country code is only unambiguous if you already know the country. Tools that default to +1 will happily produce a valid-looking US number from a foreign one.
• Leading trunk codes. Many countries write domestic numbers with a leading 0 that is dropped in international format. Keeping it produces a number that fails to route.
• Spreadsheet coercion. Opening a CSV in a spreadsheet application frequently strips leading zeros and reformats long digit strings into scientific notation. This corrupts numbers before any tool sees them, and it is the single most common cause of "the file was fine when I sent it".
• Extensions. E.164 has no room for an extension. Extensions belong in a separate field, not appended to the number.

Format valid is not the same as real

A number can be perfectly formed E.164 and still be worthless to call. The standard describes structure, not status: it says nothing about whether the line is in service, whether it is a mobile or a landline, or who it belongs to now. +15005550000 is valid E.164 and reaches nobody.

That is the distinction between syntactic validation, which a regular expression can do for free, and carrier-level validation, which requires querying live routing data. Only the second tells you whether a number will connect. See line type for what that query returns.

NumberBroom normalises every uploaded row to E.164 before any check runs, so the same number written five ways is billed and dialled once. Try a single number free.

Frequently asked questions