SMS API providers for Australia

Australian SMS is not a commodity market pretending to be one — it is a genuinely fragmented one pretending to be simple. Three things make it different from sending to the US or the EU: alphanumeric sender IDs are permitted and widely used, per-message wholesale rates are three to eight times higher, and roughly half the credible vendors will not let you send a message until someone from their revenue team has spoken to you. That last one is the field worth optimising for first, because it determines whether you ship this week or next quarter.

If you want to be sending inside an hour, the shortlist is short. ClickSend, SMS Broadcast, SMSGlobal, Cellcast and Kudosity all take a card, hand you an API key, and publish a rate table you can read before you sign up. Twilio does too, and its documentation is still the best in the category by a wide margin — but Twilio's Australian list price is roughly double the local resellers', it does not do MMS to Australian handsets, and its volume discounts are only reachable through sales, which quietly reintroduces the problem you avoided at signup.

At the other end, MessageMedia, Esendex and Bird are enterprise-motion businesses. They are not worse products; MessageMedia in particular has genuinely strong Australian carrier relationships and a long compliance track record. But none of them publish per-message pricing, all of them route new accounts through a human, and if your integration is one webhook and a POST, you will spend more time in procurement than in code. Choose them when you are buying a relationship — deliverability escalation paths, a named account manager, a contract with an SLA in it — not when you are buying an endpoint.

Telstra sits in a category of one. It is the only provider on this roster that owns the network your message lands on, which removes an entire class of "the aggregator changed routes and delivery fell over" failure. The trade is that everything else about it — commercial terms, pricing transparency, the speed at which you can get to production — behaves like a carrier rather than like a developer platform. The dev portal will let you send test messages; getting to production is a Telstra account conversation.

Where this is heading: RCS is the live question. Australian carriers support it, Google's RBM ecosystem has matured, and every vendor on this list now has an RCS line on its marketing site. Very little of it is generally available, per-message pricing for RCS is almost universally quote-only, and the fallback-to-SMS billing behaviour is where the surprises will be. Do not pick a provider on its RCS roadmap in 2026. Pick on price, self-serve access and carrier route, then re-evaluate RCS when someone publishes a rate card for it.

The headline per-message rate is the smallest of the three costs in an Australian SMS budget. The first hidden cost is segmentation: as covered in the methodology, a single non-GSM character more than doubles the effective rate on anything longer than a tweet. The second is the dedicated virtual number. If two-way messaging is required, an alphanumeric sender will not do it, and that means a monthly number rental — typically five to fifteen dollars per number per month, sometimes with a setup fee, and almost never included in the per-message quote.

The third is the shape of the commitment. Prepaid credit models — ClickSend, SMS Broadcast, SMSGlobal, Cellcast, Kudosity — are pleasant right up until the expiry terms on unused credit surface, or until the good rate turns out to require buying a hundred thousand messages upfront. Postpaid and contract models push the other way: better unit economics, but a minimum spend committed before real send volume is known. For a product that has not launched, prepay at a worse rate and eat the difference; the option value of walking away is worth more than two cents a message.

One trap is specific to this market. Several providers quote "from" pricing achievable only on domestic traffic, sent from a domestic account, with a registered sender. Test with the actual message content, the actual sender ID and the actual recipient mix before committing, because international and roaming numbers sitting in a contact list bill at a completely different rate and nothing in the dashboard warns about it in advance.

Work in this order. First, decide whether two-way is needed. If it is, a dedicated number is required and the per-message maths changes; if it is not, an alphanumeric sender is cheaper, faster to set up, and immediately recognisable to recipients. Second, decide whether Australian data residency is a hard requirement — for health, government and most financial services work it is, and that eliminates Bird and Vonage outright while putting Sinch and Esendex into "ask them" territory.

Third, get to a sent message. This is where the killer field earns its keep. Sign up to two providers, not one, and send a real message to a real handset on each. Twenty minutes there teaches more than a week of comparing rate tables: whether delivery receipts actually arrive at the webhook, how long they take, whether the receipt distinguishes "accepted by carrier" from "delivered to handset", and whether sandbox behaviour matches production. ClickSend and SMS Broadcast are the fastest two through this loop; Twilio is the most pleasant.

Fourth, and only then, negotiate. Once there is volume data from a live integration, the enterprise providers become worth talking to, and the negotiation happens from a position where a working alternative already exists. Doing it in the other order — procurement first, integration second — is how organisations end up three years into a contract for an API nobody on the team has ever called.

SMS is the least locked-in category in this comparison set, and that should be exploited. The core operation is a POST with a destination, a body and a sender, plus a webhook that receives a status callback. Wrap that in a two-method interface on day one and switching providers becomes a config change rather than a project.

The genuine switching costs are elsewhere. A dedicated virtual number is portable in principle and painful in practice — porting between aggregators takes weeks and both parties have to cooperate. The opt-out list lives in the provider's suppression store unless it has been mirrored locally, and re-messaging someone who opted out at the previous provider is a compliance incident, not an inconvenience. Export it on a schedule. Delivery-receipt history is almost never exportable in bulk, so anyone reporting on deliverability should land those events in their own store as they arrive.

The one thing that cannot easily be replicated is reputation on a sender ID. A brand's alphanumeric sender that has been pushing clean traffic for two years is treated accordingly by carrier-side filtering, and a new provider starts that clock again. Plan an overlap period with both providers live and traffic shifted gradually, rather than a single cutover release that discovers the problem in production.

Prices are per outbound SMS segment to an Australian mobile number, in AUD, at three monthly volumes: 1,000, 10,000 and 100,000 messages. Where a provider publishes a public tier table, the figure is the list rate from that table. Where a provider quotes per account — MessageMedia, Telstra, Sinch and Esendex all do — the cell is null rather than a guess, because a made-up mid-market rate is worse than an honest gap. Twilio and AWS publish a flat per-destination rate with no public volume tiering, so their three tiers are near-identical; discounts exist but require sales. A segment is 160 GSM-7 characters or 70 UCS-2 characters. If your copy contains a curly apostrophe, an em dash or an emoji, you are paying UCS-2 rates for the whole message, which roughly doubles your bill on anything over 70 characters. Very few vendors say this on the pricing page.

The killer field, "self-serve API key", answers one question: from a cold start with a credit card and no prior relationship, can you obtain working credentials and deliver a message to a real Australian handset without speaking to a human? "Yes" means signup to sent message with no manual approval step. "Partial" means credentials are self-serve but something between you and production is not — a sender ID registration ticket, a manual account activation, a trial that cannot send to arbitrary numbers. "No" means a salesperson is in the critical path. A serious caveat on ACMA: Australia's SMS Sender ID Register regime has changed recently and the obligations on senders and on carriage service providers are in flux. Every cell in the ACMA Sender ID field is therefore deliberately left unknown rather than guessed. Vendor pages on this topic lag the regulator by months and several are simply out of date. This field is pending verification against ACMA directly and must not be relied on for any compliance decision — treat it as a prompt to go and check, not as an answer.