How to use the sms segment calculator
- Type or paste your message into the box. The count updates as you type.
- Read the encoding, the units used, the number of segments and the units left in the last segment.
- If the message uses a character outside the basic text message alphabet, check the note that lists it, since one such character changes the whole message to a shorter encoding.
- Use the split view to see where each part of a long message ends, then copy the segments or download them.
Formula
GSM-7: up to 160 units in one segment, or 153 per segment in a joined message
UCS-2: up to 70 units in one segment, or 67 per segment in a joined message
A few GSM-7 characters, such as € [ ] { } ^ ~ | and \, count as 2 units
In UCS-2 an emoji outside the basic range counts as 2 units
A message that uses only characters from the GSM 7-bit alphabet is sent in GSM-7. One character outside it makes the whole message UCS-2. Joined messages lose a few units in every segment to a header, and a two-unit character is never split across segments.
Where the limits fall
The sample message "Your table is ready. Reply 1 to confirm or 2 to move it to another time." is 72 characters in GSM-7, so it is one segment with 88 units to spare. A message of 160 plain letters is one segment, and 161 letters need two, since a joined message holds 153 units in each segment. A message of 306 letters is two segments and 307 is three.
Eighty euro signs use 160 units in one segment because each takes 2, and 81 need two segments. Seventy letters with an accent that is not in the GSM alphabet, such as ç, fit one UCS-2 segment, and 71 take two. Thirty-five emoji are 70 units, and 36 are two segments. Curly quotes also push a message to UCS-2: “quote” is 7 units.
Why one character can double the cost
A text message has a fixed size, so a long message is cut into segments, and some providers charge by segment. A message that fits in 160 units of the GSM alphabet is one segment, but a single emoji, a curly quote, a letter from a different alphabet or an accented letter outside the GSM set makes the phone use a different encoding with 70 units. Pasting from a word processor often adds curly quotes and long dashes, which are the usual surprise. If you replace them with straight quotes and plain punctuation, the message often drops back to GSM-7.
To count characters for social posts, use the Social Character Counter. For words, sentences and reading time, the Word Counter is the right tool.
What this calculator does and does not know
The counts follow the common rules for text messages, but carriers and apps can add their own header, trim whitespace or use messaging systems that are not SMS, so the number is an estimate. Check with your provider if you pay by the segment or send at scale. Your text stays in your browser and is never sent anywhere. For file sizes, try the Text Byte Size Calculator, and to check how easy your wording is to read, the Readability Score Calculator.
Assumptions and limits
- The message is plain text with no attachment or picture.
- GSM-7 here is the basic 7-bit alphabet plus its extension table, and everything else is treated as UCS-2.
- Characters are counted as whole characters, so an emoji is one character but two units.
- Nothing is uploaded, and the text is up to 20,000 characters.
Frequently asked questions
How many characters are in one text message?
A message in the basic GSM alphabet holds up to 160 characters in one segment. A message that needs UCS-2, for example because it has an emoji, holds up to 70.
Why does a long message hold fewer characters per segment?
When a message is joined from parts, each part carries a small header, so a segment holds 153 units in GSM-7 or 67 in UCS-2.
Why did one emoji make my message so much shorter?
An emoji is not in the GSM alphabet, so the whole message switches to UCS-2, which fits 70 units in a single segment instead of 160, and the emoji itself takes 2 of them.
How can I keep the message in one segment?
Replace curly quotes, long dashes and special letters with plain ones, and shorten the text. The note under the result lists the characters that force UCS-2.