Why Are My Business Texts Not Delivering? SMS Deliverability Guide

"Customers aren’t getting our texts" is a symptom with exactly five root-cause families: registration blocks, carrier filtering, throughput queuing, bad destinations, and handset-side issues. Diagnosing by guesswork takes days; diagnosing by error code takes minutes. Here is the decision tree.

Step 1: Read the actual message status

In the Twilio Console (Monitor → Logs → Messaging), every message has a status and, if it failed, an error code. The status tells you which family you are in:

• undelivered / failed with an error code → look the code up (families below).
• queued / accepted for a long time → throughput or provider-side backlog.
• sent but never delivered → downstream carrier or handset; no explicit rejection.
• delivered but customer says nothing arrived → device-level filtering or wrong number.

Step 2: Triage by error family

Registration blocks

30034 (unregistered 10DLC number) is the big one: the number isn’t attached to an approved campaign. Fix via the 30034 guide. Toll-free cousins exist for unverified toll-free senders. These errors are absolute: 100% failure until registration is fixed.

Carrier filtering

30007 (message filtered) is registered traffic that a spam filter scored badly: content, links, list quality, or volume behavior. Partial and probabilistic (some messages pass, some don’t). Systematic fix in the 30007 guide.

Bad destinations

The 30003/30005/30006 family covers unreachable, unknown, or landline destinations. These are list-quality problems: validate numbers (a lookup API catches landlines and disconnects) and prune hard bounces automatically.

Throughput queuing

Messages that sit queued and deliver late are draining through your carrier rate limits: AT&T per-minute classes and T-Mobile daily caps. Not an error, but a launch-day disaster if unplanned. Fix by vetting up your brand tier, spreading sends, or shortening messages (segment math).

Step 3: The silent failure modes

When statuses look fine but customers still report nothing:

• Device-level filtering: modern iOS/Android route unknown senders into "junk"-style folders. Branded, expected, consistent content minimizes this; asking customers to save your number eliminates it.
• Wrong or recycled numbers: the customer’s number changed hands; "delivered" went to a stranger. Periodic re-confirmation for critical notifications helps.
• Opt-out suppression: the customer texted STOP months ago (perhaps to a different keyword) and your platform is correctly suppressing. Check the suppression list before debugging carriers.

Build the early-warning system

1. Track delivery rate per send and per carrier. A drop isolated to one carrier is a filtering event; across all carriers, a registration or content event.
2. Alert on error-code anomalies. Any 30034 in production means a number escaped the sender pool; a 30007 rate above ~1–2% means content drift.
3. Watch opt-out and reply rates, the leading indicators of list fatigue that precede filtering.
4. Test through real handsets on the three major carriers before each big send. It is a $30/month test-phone habit that catches template problems before 50,000 recipients do.

Deliverability is downstream of the same fundamentals as approval: registered correctly, consented lists, honest content, sane volume. Programs with those four never spend long in this guide.

Frequently asked questions

What is a good SMS delivery rate?

Registered 10DLC traffic to a clean, consented list typically delivers in the high 90s (%). Sustained rates below ~95% signal list-quality or filtering issues worth investigating; below 90% something is systematically wrong.

Why do my texts deliver to some carriers but not others?

Each carrier applies its own filters and limits. One-carrier failure patterns usually mean that carrier’s content filtering (see error 30007) or, on T-Mobile, an exhausted daily cap. Per-carrier delivery breakdowns in your logs isolate this quickly.

Can recipients’ phones block my messages without any error?

Yes. Device-level spam filtering happens after carrier delivery, so your logs show "delivered" while the message sits in a junk folder. Consistent branding, expected timing, and having customers save your contact reduce it.