SMS segment
An SMS segment is the billing unit of text messaging. A message is not billed per message; it is billed per segment, and how many segments a message occupies depends on its length and on which characters it contains. One curly apostrophe can double the cost of a district-wide send.
| Encoding | Single message | Per segment when concatenated | Triggered by |
|---|---|---|---|
| GSM-7 | 160 characters | 153 characters | Plain Latin letters, digits, common punctuation |
| GSM-7 with extension | 160, but some characters cost 2 | 153, same rule | [ ] { } \ | ^ ~ € and a newline in some handsets |
| UCS-2 | 70 characters | 67 characters | Any emoji, any accented character, curly quotes, em dashes |
Concatenated segments are shorter because a few bytes of each segment carry the header that lets the handset reassemble them in order.
The characters that silently downgrade a message
The downgrade from GSM-7 to UCS-2 is all-or-nothing. One character outside the GSM-7 set re-encodes the entire message, and the budget drops from 160 to 70 in an instant. The characters that do it are almost never typed deliberately:
- Curly quotes and apostrophes. Word, Google Docs and macOS all convert straight quotes to curly ones automatically. Paste from a document and the message is UCS-2 before anyone has read it.
- The em dash and en dash. Same autocorrect path.
- Accented characters. Every á, é, í, ó, ú and ñ.
- Emoji. A single one costs more than the sentence around it.
- Ellipsis as one character rather than three full stops.
The arithmetic is worth internalising. A 150-character message in plain ASCII is one segment. The same message with one curly apostrophe is UCS-2 at 70 characters per segment, so it becomes three. The message did not change; the bill tripled.
What that costs a real district
Take a 6,000-family district and price the same send at one, two and three segments. Use a round one cent per segment per recipient as a deliberate upper bound — real rates are lower and vary by carrier, and estimating high is the safe direction for a budget.
| Segments | One send | Weekly for a 36-week year | Plus 20 all-district alerts |
|---|---|---|---|
| 1 | $60 | $2,160 | $3,360 |
| 2 | $120 | $4,320 | $6,720 |
| 3 | $180 | $6,480 | $10,080 |
The difference between the first and third rows is roughly the cost of the communications platform itself, for the same messages, containing the same information. It is entirely determined by choices made in a composer by someone who has no way of seeing the consequence — unless the composer shows it.
This is why Kastr renders a live per-channel cost and reach estimate as the sender ticks channels and types, before the send. It turns an invisible line item into a design decision, and to our knowledge no other platform in this category does it.
The bilingual trap, and how to work around it
Now apply the encoding rule to a district that translates every message, which is what language access obligations increasingly require.
Spanish carries accented characters and the ñ. Vietnamese carries extensive diacritics. Arabic, Chinese, Korean, Russian and Somali are outside GSM-7 entirely. Every one of those translations is UCS-2 by definition, at 70 characters per segment. A one-segment English message becomes a three-segment Spanish one, and the district's bilingual families cost roughly three times as much to reach for the same content.
Do not solve this by translating less. The cost difference is real and the obligation is not optional. Solve it by writing shorter source messages, moving detail to a linked page rather than into the SMS body, and using SMS as the alerting channel with email carrying the long form. A 60-character SMS survives translation into one or two segments in most languages; a 155-character one never does.
Two practices that help immediately: draft in plain ASCII and let translation add the accents where they belong, and check the segment count of the longest translation rather than the English source. The English version is the one that looks cheapest and the one nobody is billed for.
Questions people actually ask
Why does one emoji triple the cost of a text message?
Because a single non-GSM-7 character re-encodes the whole message as UCS-2, dropping the per-segment budget from 153 characters to 67. A 150-character message goes from one segment to three. The emoji does not cost extra by itself; it changes the encoding for everything around it.
How many characters can a school text message be?
160 in plain GSM-7 for a single segment, or 153 per segment once a message runs long enough to be split. If the message contains any accent, emoji, curly quote or em dash, the budget is 70 and 67 respectively.
Do Spanish messages cost more to send?
Yes, materially. Accented characters force UCS-2 encoding, so a translated message occupies roughly twice as many segments as the same content in plain English. Districts that translate everything should budget for it and should write shorter source messages, not send fewer translations.
Does a split message arrive as one text or two?
On modern handsets it reassembles into one message, because each segment carries a header describing its position. Older devices and some carrier paths show them separately and occasionally out of order. Billing counts every segment regardless of how it displays.
One price. Every feature. Locked for three years.
$3.50 per student per year under 5,000 students. No tiers, no add-on modules, no per-message fees. Published on the site because you should not have to book a call to learn a price.