What is the difference between A2P and P2P messaging?

A2P is application-to-person: software originates the message and a person receives it. P2P is person-to-person, two humans messaging at conversational volume. Carriers treat them as separate products because A2P traffic is high volume, one-directional and sent in bursts, which is the same shape as spam. A2P therefore carries registration requirements, throughput limits, content filtering and its own pricing, none of which apply to ordinary personal texting.

Why are my A2P messages being filtered or blocked by carriers?

Several causes produce the same symptom. The sending number may not be registered to a campaign, or the traffic may not match the use case it was registered under. Content filters block prohibited categories and often flag links from public URL shorteners. The number may be a landline or a VOIP line that discards SMS. Because SMS rarely returns diagnostic errors, the practical approach is to eliminate the knowable causes first, starting with validating line type.

Do I need consent to send A2P messages to my customers?

Yes. Under the TCPA, autodialled messages to a mobile number generally require prior express consent, rising to prior express written consent where the content is marketing. This is entirely separate from carrier registration: registering a 10DLC campaign authorises carriers to carry your traffic but grants no permission from the recipient. A fully registered campaign sending without consent carries the same statutory exposure as any other unconsented send.

Can A2P messages be delivered to landline numbers?

No. A landline has no mechanism for receiving SMS, so the message fails regardless of how your campaign is registered or how clean the content is. Some carriers convert texts to a landline into an automated voice call that reads the message aloud, which is rarely a useful customer experience. Because you are billed per send attempt, landlines in a list cost full price and deliver nothing, and the resulting failures feed into the reputation carriers score your traffic on.

How do I know whether an A2P message was actually delivered?

Less precisely than you would like. Delivery receipts indicate that a carrier accepted the message, which is not the same as a handset displaying it, and the granularity varies considerably by carrier and route. Filtered messages frequently return either nothing or a generic status indistinguishable from an unreachable handset. The practical response is to remove the knowable failure categories in advance and then monitor delivery rates by segment rather than trusting individual receipts.

Should transactional and marketing A2P traffic use the same campaign?

No, and separating them is one of the higher-leverage decisions available. Carriers apply more scrutiny and tighter filtering to promotional content than to transactional messages such as passcodes and delivery alerts. Running both through one campaign subjects your time-critical transactional traffic to the constraints earned by your marketing traffic. Registering them separately costs an additional recurring campaign fee and is almost always worth it.

A2P messaging

Application-to-person SMS — messages sent by software to a person, rather than between two people.

Upload a CSV and validate your phone list

No account required · Results in under 60 seconds

A2P messaging, short for application-to-person, describes any SMS originated by software rather than typed by a human on a handset. One-time passcodes, delivery notifications, appointment reminders, payment alerts and marketing campaigns are all A2P traffic, and together they account for the overwhelming majority of business text messaging.

The distinction matters because carriers treat A2P and person-to-person traffic as fundamentally different products. A2P is subject to registration requirements, throughput ceilings, content filtering and separate pricing. Sending application traffic over channels intended for conversational messaging is one of the most reliable ways to have messages silently filtered, and it is a mistake businesses make routinely when they scale outreach faster than they scale their messaging setup.

What is A2P messaging?

The defining characteristic of A2P messaging is that the sender is a system, not a person. A platform triggers the message in response to an event: a purchase completes, a login attempt needs verification, an appointment approaches, a campaign schedule fires. The recipient is a person, which is what distinguishes A2P from machine-to-machine traffic.

Person-to-person messaging sits at the other end of the spectrum. Two humans exchange messages at conversational volume and pace, with roughly symmetrical traffic in both directions. Carriers have long tolerated this traffic with minimal scrutiny because its shape makes abuse self-limiting.

A2P traffic looks nothing like that. It is high volume, heavily one-directional, often sent in bursts to thousands of recipients simultaneously, and originated from a small number of sending numbers. Those same characteristics describe spam, which is precisely why carriers built separate infrastructure and rules to handle legitimate A2P rather than trying to distinguish it case by case.

✓Registered 10DLC — send from a standard local number after registering a brand and campaign.
✓Short codes — five or six digit numbers with individual carrier approval and very high throughput.
✓Verified toll-free — lighter verification than a short code, with solid throughput for national traffic.

In the United States, A2P is delivered through three sanctioned channels. Registered 10DLC lets businesses send from standard local numbers after registering a brand and campaign. Short codes are five or six digit numbers leased for high-volume campaigns with individual carrier approval. Verified toll-free numbers occupy a middle ground, requiring a lighter verification than short codes but offering solid throughput.

The category also spans wildly different risk profiles. A two-factor authentication code is high-value, time-critical, and explicitly requested by the recipient moments earlier. A promotional blast is none of those things. Carriers price, throttle and scrutinise the two differently, which is why they are registered as separate use cases rather than lumped together as business messaging.

Internationally the picture fragments further. Many countries require sender ID registration, some mandate local number origination, and a few prohibit certain content categories outright. A business sending across borders is managing several regulatory regimes simultaneously, each with its own registration and filtering behaviour.

How A2P messaging works

An A2P message begins as an API call from an application to a messaging provider. The provider validates the request, applies its own content and rate checks, and hands the message to a carrier or an aggregator that maintains carrier connections. The carrier then attempts delivery to the recipient's handset through its network.

Each hop can reject or drop the message. Providers block content that violates their acceptable use policies. Aggregators enforce throughput limits. Carriers apply spam filtering, check registration status, and verify that the traffic matches the declared use case. A message that clears every layer is delivered; one that fails a check often disappears without a clear error reaching the sender.

✓Your provider applies acceptable-use and content checks before the message leaves.
✓An aggregator enforces the throughput ceiling for your campaign.
✓The carrier checks registration, filters content, and attempts delivery to the handset.
✓Any stage can drop the message, often without returning a usable error to the sender.

That silence is the defining operational difficulty of A2P. Unlike email, where bounces return structured diagnostics, SMS filtering frequently produces either no feedback or a generic status that does not distinguish a blocked message from an unreachable handset. Senders are left inferring causes from aggregate delivery rates rather than reading errors.

Throughput governs how fast a queue drains. A campaign granted a modest messages-per-second allowance does not fail when sent to a large list; it simply takes far longer than expected. For time-sensitive content this is functionally equivalent to failure, since a reminder arriving hours late has lost its purpose.

Delivery receipts add another layer of ambiguity. A receipt indicating the carrier accepted a message is not confirmation a handset displayed it, and the granularity varies by carrier and route. Treating receipts as proof of delivery consistently overstates real performance.

Line type determines whether delivery is possible at all before any of this machinery engages. A landline has no mechanism to receive SMS, so the message fails regardless of registration, throughput or content. VOIP numbers vary entirely by provider, with some supporting SMS and others silently discarding it.

Opt-out handling is mandatory and automatic on sanctioned channels. Replies such as STOP are intercepted and enforced at the carrier or provider level, and continuing to send after one is both a delivery problem and a compliance exposure.

Watch out

A2P failures are usually silent. A message blocked by a content filter, one sent to a landline, and one sent to a disconnected number can all return the same unhelpful status. Validating line type before the send is the only way to remove an entire category of failure from that ambiguity.

Why A2P messaging matters for SMS and outreach teams

A2P is where most businesses actually reach customers by text, so its constraints shape what outreach is possible. Understanding throughput, filtering and line type is the difference between a channel that performs predictably and one that produces unexplained results.

Cost is the most immediate consequence. A2P is billed per send attempt, not per delivery, and carrier surcharges apply on top of provider rates. Every message aimed at a landline, a disconnected number, or a VOIP line that discards SMS is billed in full and delivers nothing. Across a large list the waste is not marginal.

Sender reputation is the compounding consequence. Carriers assess traffic partly on delivery outcomes, and sustained failures are the signature of a poorly sourced list. That assessment degrades delivery for the portion of the list that would otherwise have succeeded, so undeliverable numbers actively damage deliverable ones rather than merely costing their own send fee.

The distinction between use cases matters commercially too. Transactional A2P such as passcodes and delivery alerts is tolerated far better by both carriers and recipients than promotional traffic. Mixing them through a single campaign drags the high-value transactional messages into the scrutiny applied to marketing, which is an avoidable self-inflicted constraint.

There is a customer experience dimension as well. A person who receives a text they did not expect from a number they do not recognise is likely to report it as spam, and enough such reports affect carrier treatment of your traffic directly. Clear opt-in and recognisable sender identity are deliverability mechanisms, not just courtesies.

Finally, A2P sits inside a legal framework. Carrier permission to carry traffic is entirely separate from the recipient's permission to receive it, and the TCPA governs the latter regardless of how thoroughly the former is satisfied.

Getting started with A2P messaging

Start by cataloguing what you actually send. List every automated message your systems originate, group them by purpose, and identify which are transactional and which are promotional. Most organisations find messages they had forgotten about, frequently sent by a tool nobody has reviewed in a year.

1
Inventory your traffic — List every automated message your systems send and group them by purpose, separating transactional from promotional.
2
Pick the right channel — Registered 10DLC suits most senders, verified toll-free suits national support traffic, and short codes earn their cost only at high volume.
3
Register use cases separately — Keeping passcodes on their own campaign protects the messages where delay costs the most.
4
Validate before the first send — Append line type and validity, then remove landlines and dead numbers before they generate failures.
5
Instrument by segment — Track delivery by carrier, line type and acquisition source, because one aggregate number hides the cause.

Then revalidate on a schedule. Numbers are disconnected and reassigned continuously, so a list validated once at import decays steadily. Checking before each significant campaign catches the drift that a one-time check never will.

Key takeaways: A2P messaging

✓A2P messaging is SMS originated by software rather than a person, covering passcodes, alerts, reminders and marketing.
✓Carriers treat A2P and person-to-person traffic as different products with separate registration, throughput, filtering and pricing.
✓In the US, sanctioned A2P channels are registered 10DLC, short codes, and verified toll-free numbers.
✓Failures are frequently silent, so a blocked message and an unreachable landline can look identical to the sender.
✓Billing is per send attempt, meaning undeliverable numbers cost full price and return nothing.
✓Carrier permission to carry traffic is separate from the recipient's consent to receive it, which the TCPA governs.

Frequently asked questions about A2P messaging

What is the difference between A2P and P2P messaging?
Why are my A2P messages being filtered or blocked by carriers?
Do I need consent to send A2P messages to my customers?
Can A2P messages be delivered to landline numbers?
How do I know whether an A2P message was actually delivered?
Should transactional and marketing A2P traffic use the same campaign?

10DLC — The US carrier framework for sending registered application-to-person SMS from a 10-digit local number.
SMS deliverability — The proportion of sent SMS messages that are actually delivered to a recipient handset.
Short code — A five- or six-digit number leased for high-volume, carrier-approved SMS campaigns.
Opt-in — A recipient explicitly granting permission to be contacted by call or text.
Line type — The classification of a phone number as mobile, landline, VOIP, or invalid.

Check your list before your next send

Upload a CSV and PhoneCheckr flags every mobile, landline, VOIP, and invalid number so you only pay to reach contacts who can actually receive your message.

Validate your phone list

Back to the full glossary