How we test and source this site
We test and source this site by tiering every source: each iPhone fact cites a support.apple.com page we opened and dated. Older Apple text and community reports are labelled as such, never stated as current fact. SMS counts follow 3GPP TS 23.038, run through our own tested script. Anything we haven't tested on a real iPhone is marked untested.
The short answer
Our methodology is simple: every claim points to a source we opened, and the source's strength is labelled. Every iPhone fact cites a support.apple.com page, with the date we last opened it. Apple's older, archived wording is labelled "Apple's older guidance" and never presented as current. Community reports from forums are labelled as reports and never decide a fact on their own. SMS character and segment counts follow the 3GPP standard TS 23.038 and run through our own script, which checks itself before every run. Anything we haven't tested on a real iPhone is marked "untested". When we get something wrong, it goes on a public corrections log. The three facts every page repeats are summarized on the Batchrunner home page, next to the free message writer.
Who writes this site
The site is written by the people who make Batchrunner, an iPhone app for sending one text to many people. The developer named on the App Store listing is Nattapon Kittisuphat. We use AI tools for research and drafting. Every factual claim is tied to a source listed on the page. There are no named experts on this site and no "reviewed by" badges, because we don't have outside reviewers. Because we make an app in this space, every page that mentions apps says "Batchrunner is our app."
Source tiers
We sort every source into one of these tiers. The tier decides how a page may word the claim.
| Tier | Examples | How a page may word it |
|---|---|---|
| 1. Apple current | support.apple.com articles and iPhone User Guide pages | Stated as fact, with the link and check date |
| 2. Apple archived | Old Apple support text read through the Wayback Machine (e.g. HT202724, 2014) | "Apple's older guidance (2014) said…" plus "may vary by iOS version and carrier" |
| 3. Carrier official | Verizon support FAQs | "Verizon says, for its network…" — never generalized to all carriers |
| 4. Regulator / industry body | FCC orders and guides, CTIA Messaging Principles, The Campaign Registry | Quoted with date; business pages add "not legal advice" |
| 5. Standards and technical docs | 3GPP TS 23.038 (via ETSI), Twilio's SMS glossary | Used for SMS encoding and segment math |
| 6. Community report | Apple Community threads, Reddit, forums | "Users report…" — never stated as fact |
| 7. Our own reasoning | Rankings on our picker, practical list-size estimates | Marked "our estimate" or "our assessment" |
If Apple doesn't document something, we say so. For example: Apple publishes no maximum group size, so we don't print the 20, 25 or 32-person figures that circulate online. We say "Apple doesn't publish a limit" instead.
If a source won't load, we don't fill the gap. On 4 Nov 2026, WhatsApp's broadcast-list help article would not load for us, so our method comparison gives no WhatsApp list-size number.
What we test, and what we can't
We test:
- Our SMS segment script, with 25 built-in checks that run before it produces any numbers.
- Every link and quote on a page, opened on the date shown in its Sources list.
- Our own app's behavior, described from how it is built: it opens Messages batch by batch, you tap Send, and a ledger marks sent, remaining and skipped.
We can't test:
- Every carrier. Group-message handling and long-text delivery depend on the carrier. Verizon, for example, says texts over 160 characters go as EMS on its network. We don't have lines on every US carrier, so we quote carriers' own pages and say which network each applies to.
- iOS betas. We describe released iOS versions only. Apple's unversioned User Guide pages now show iOS 27 text; where wording differs, we cite the version-specific page.
- Android phones. We make no claims about Android behavior that an official source doesn't state.
- Other companies' apps. We didn't install competitor apps. Our app comparison lists name, developer and link only, and gives a checklist you can apply yourself.
On-device tests still pending (iOS 26):
- Whether an
sms:link with one recipient and a message body opens a filled-in compose screen (used by our send list). - The group-text table rows marked "untested on device" — rows 4, 6 and 12.
- Whether iPhone's Smart Punctuation turns straight quotes into curly ones in a text, which would switch it to UCS-2.
Until these are done, the pages say "untested."
How we count SMS segments
Our SMS character limit counter and our template data use the same rules:
- GSM-7: one SMS holds up to 160 characters (3GPP TS 23.038: "(140x8)/7=160 characters"). Longer texts split into parts of 153 characters each (Twilio).
- Extension characters
^ { } \ [ ~ ] | €count as 2 each, because each needs an escape code (TS 23.038 §6.2.1.1). - UCS-2: one character outside GSM-7 — an emoji, a curly quote, an em dash — switches the whole text to UCS-2. Then one SMS holds 70 characters, and parts hold 67 each (TS 23.038; Twilio). An emoji counts as 2.
- An escape pair or an emoji is never split across two parts.
{firstName}is counted as an 8-letter name ("Jonathan"). Real counts change with each person's name.- Toll-free US numbers use 152/66 per part (Twilio). Our counts assume an ordinary phone number.
| Example text (with an 8-letter name) | Encoding | Characters | SMS parts |
|---|---|---|---|
Hi {firstName}, practice moved to 6pm tonight. Bring water. | GSM-7 | 56 | 1 |
Same idea with a curly apostrophe: Hi {firstName}, it’s on for tonight at 6pm. Bring water. | UCS-2 | 53 | 1 |
| 161 plain characters | GSM-7 | 161 | 2 (153 + 8) |
| 71 characters including one curly quote | UCS-2 | 71 | 2 |
iMessage and RCS don't use 160-character parts; these counts matter for green-bubble SMS. Apple doesn't document whether an iPhone sends a long text as several SMS parts or as MMS.
How the group-text table is built
Our table of what happens when you text a group from an iPhone crosses three inputs: who is on the list (all iMessage, mixed, unknown), the sender's MMS Messaging switch (on, off), and the send method (one message to all, one per person). That makes 12 rows. For each row we record the thread type, who sees replies, and whether others can see who got it. Each row carries the tier of its weakest cell. A row with no current Apple source is marked "untested on device" until we test it. The table has a dated changelog, and the CSV carries the same tiers.
How templates are counted
Our 42 text message templates are generated by the same script as the counter:
- 42 templates in 11 categories (party invite, wedding RSVP, RSVP reminder, team practice change, rainout, church event, parent notice, tenant notice, appointment reminder, holiday greeting, thank-you).
- Every template includes
{firstName}, so each copy greets by name. In Batchrunner 2.2, include {firstName} to open one conversation per person; without a merge token, a batch can open as a group. - With an 8-letter name, lengths run from 92 to 147 characters (median 127).
- 40 templates are GSM-7 and fit in one SMS. 2 are UCS-2 and take two parts: one with an emoji, one with curly quotes.
- 5 business templates include a "Reply STOP to opt out" line; one more notes where to add it for recurring sends.
How we compare apps
Apple's terms keep us from copying App Store prices, ratings or descriptions. So iPhone mass-text apps: what to check before you pay lists apps by name, developer and link, plus a 10-point checklist. Three of the ten points are ones our own app fails.
Business and legal topics
Pages about texting customers quote the FCC and CTIA with dates and say "not legal advice". FCC 26-67, adopted 30 Sep 2026, takes effect 30 days after Federal Register publication; the FCC has said it will announce the exact date. We re-check it before publishing texting customers: consent and opt-out basics.
Why replies sometimes go to everyone
This is the claim we're asked about most, so here is our sourcing for it. Apple 118236 says that in a group iMessage or group MMS, everyone can see all responses. In a group SMS, Apple says responses arrive as individual texts the others can't see. Messages picks the type itself based on settings, network and carrier plan. That's why every page on this site recommends one text per person.
Corrections
When a fact changes or we get one wrong, we fix the page, update its date, and add a line to corrections with the date, the page and what changed. To report an error, use the contact link in the footer. Our app's own facts — price, limits, how it sends — are kept the same on every page and on Batchrunner for iPhone.
Sources (9)
- ETSI TS 123 038 V18.0.0 (3GPP TS 23.038). www.etsi.org/deliver/etsi_ts/123000_123099/123038/18.00.00_60/ts_123038v180000p.pdf — checked 3 Nov 2026.
- Twilio — What is the SMS character limit? www.twilio.com/docs/glossary/what-sms-character-limit — checked 3 Nov 2026.
- Twilio — What is GSM-7 character encoding? www.twilio.com/docs/glossary/what-is-gsm-7-character-encoding — checked 3 Nov 2026.
- Apple Support — Send a group text message on iPhone (118236). support.apple.com/en-us/118236 — checked 4 Nov 2026.
- Apple Support — If you can't send or receive messages on iPhone (118433). support.apple.com/en-us/118433 — checked 4 Nov 2026.
- Apple Support HT202724 (archived 2014). web.archive.org/web/20141123171913/http://support.apple.com/en-us/HT202724 — checked 3 Nov 2026.
- Verizon — Text messaging FAQs. www.verizon.com/support/text-messaging-faqs/ — checked 3 Nov 2026.
- CTIA Messaging Principles and Best Practices (May 2023). api.ctia.org/wp-content/uploads/2023/05/230523-CTIA-Messaging-Principles-and-Best-Practices-FINAL.pdf — checked 3 Nov 2026.
- FCC 26-67. docs.fcc.gov/public/attachments/FCC-26-67A1.pdf — checked 3 Nov 2026.