Most integration guides can start with the software. This one cannot, because the hard part of a HubSpot SMS integration is not software at all. It is a phone number that a US carrier has agreed to let you send from, a consent record that can survive a regulator asking about it, and a throughput ceiling set by a vetting score you do not control.
HubSpot does have native SMS, and for the job it was built for it works. You buy an add-on, register a number, and your workflows can text people the way they already email them, with the messages landing on the contact timeline next to everything else. For a US company running campaign sends to US contacts, that is the whole answer and there is nothing to build.
Then someone asks to text a customer in Manchester, or a second brand wants its own sending number, or support wants inbound replies routed to a rep rather than into a marketing subscription. That is where this guide starts. What the native add-ons actually cover, why consent is the real integration surface, how the carriers cap your speed, and how to tell the difference between a problem you can configure and one you have to build. It sits alongside the broader guide to HubSpot integrations.
In this article
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Does HubSpot Have Native SMS?
Yes, and it is worth being precise about what you have to buy, because "HubSpot has SMS" and "SMS is included" are very different statements.
Marketing SMS requires either the Marketing SMS Add-On or the Marketing SMS Short Code Add-On. The Marketing SMS Add-On itself requires Marketing Hub Professional or Enterprise. The short code route requires a Marketing Hub Enterprise account plus the SMS Limit Increase. Transactional SMS is a third add-on, the Transactional SMS Add-On, and it grants its own separate long code number.
Underneath all of it, HubSpot's SMS provider is Twilio. That matters more than a vendor footnote usually does, because it means the rules that decide what you may send, to whom, and how quickly are Twilio's and the US carriers', not HubSpot's. HubSpot's own documentation points you at Twilio's forbidden message categories to check whether your business or message content is allowed at all.
That single sentence decides more architecture than anything else on this page. A holding company with four brands in one portal, a franchise network, or a sales team that wants local area codes all run into it immediately, and no amount of configuration produces a second marketing number.
What the Native HubSpot SMS Integration Actually Covers
There are two separate products here, they are bought separately, and they get separate phone numbers. Confusing them is the most common setup mistake.
| Marketing SMS | Transactional SMS | |
|---|---|---|
| Add-on required | Marketing SMS Add-On, or Marketing SMS Short Code Add-On | Transactional SMS Add-On |
| Subscription gate | Marketing Hub Professional or Enterprise. Short codes need Enterprise plus the SMS Limit Increase | The add-on itself |
| Number type | Long code (10DLC) or short code | Long code (10DLC) |
| Numbers per account | One | One |
| Who you can reach | US and Canada on a long code. US only on a short code | US and Canada |
| Typical use | Campaigns, promotions, nurture | Order confirmations, renewals, time-sensitive notices |
| Approval time | Six to eight business days for a long code, eight to twelve weeks for a short code | Six to eight business days |
Both exclude the US Virgin Islands, both require the recipient to have a US-format number beginning with +1, and neither is available to sole proprietors, which HubSpot states outright.
How to Set Up SMS in HubSpot
The setup is mostly a registration exercise, and the parts that take time are the parts you cannot rush by being organised.
1. Work out which add-on you actually need, and whether you need two
Marketing and transactional are different purchases with different numbers. If you send both campaign messages and order confirmations, you need both, and you should decide that before registration rather than discovering it after a number is already tied to a use case.
2. Choose long code or short code with the lead time in mind
A long code (10DLC) takes six to eight business days and reaches the US and Canada. A short code takes eight to twelve weeks, reaches the US only, and needs Enterprise plus the SMS Limit Increase. If a launch date depends on this, the short code decision has to be made a quarter ahead.
3. Gather the business registration details first
You will need a registered business address and an EIN or equivalent business number. This is A2P 10DLC brand registration, and it is verifying that your company is who it says it is. Sole proprietors are not supported.
4. Document every opt-in method you use, with evidence
You have to describe your SMS use case, provide sample messages, and list every way you collect consent. This is where submissions get rejected. If you collect consent on a web form, the consent checkbox has to be genuinely unchecked by default, and it has to look unchecked in whatever you submit.
5. Have live privacy policy and terms of service URLs ready
They have to resolve, and they have to describe the messaging programme you are registering. A placeholder page or a link behind a login will hold up the whole registration.
6. Set up the subscription type and test consent both ways
Before you send anything real, confirm that texting START subscribes a contact and STOP unsubscribes them, and that both show up on the record. Consent is the part of this integration that has legal consequences, so it is the part worth testing deliberately rather than assuming.
Consent Is the Real Integration
Every other connector guide treats consent as a footnote. On SMS it is the product, so it is worth understanding exactly how HubSpot models it.
Consent is held as an SMS subscription against the phone number, not as a general property of the contact. Contacts opt in by texting START, UNSTOP or SUBSCRIBE to your registered number, and their subscription status updates automatically. They opt out with STOP, STOPALL, UNSUBSCRIBE, CANCEL, END or QUIT. You can subscribe someone manually, but only where you hold explicit and verifiable consent obtained another way and can document the legal basis for it.
Two things follow from that design and both catch teams out.
The first is that consent is attached to a number, so a contact who changes phone numbers has not carried their consent with them, and a number that moves to a new person carries the old person's consent until something corrects it. The second is that the carrier keywords are handled at the messaging layer, which means an opt-out is authoritative in the SMS channel whether or not any other system in your stack ever hears about it.
Where Native HubSpot SMS Stops
Every item here is documented behaviour, which is why configuration does not get around any of it.
The limits that decide whether native SMS is enough
There is no international sending. You can only send from US-based numbers, long codes reach the US and Canada, short codes reach the US only, and recipients need a +1 number. For a company with buyers outside North America this is not a limitation to work around, it is the end of the native path.
One marketing number, one transactional number. Multiple brands, regions, business units or local area codes are simply not expressible. Everything you send comes from the same two numbers, so recipients cannot tell your brands apart and you cannot segment deliverability reputation between them.
Sole proprietors are excluded entirely. HubSpot states SMS is not currently supported for them, which is a carrier registration constraint rather than a HubSpot choice, and it rules out a whole class of small business outright.
Throughput is assigned to you, not chosen by you. HubSpot describes message volume as variable and dependent on a trust score determined by a third party. You cannot buy your way past it inside HubSpot, and it is shared across everything sending under your brand.
Consent lives in one channel's subscription record. That is correct for HubSpot's purposes and insufficient the moment another system also needs to know. A support desk, a billing platform or a mobile app that texts people has no native way to learn that someone replied STOP to a marketing message.
Sending is triggered by CRM events. Workflows fire on HubSpot data changing. If the thing that should trigger a message is a shipment leaving a warehouse, a payment failing in a billing system or a job finishing in an operational tool, that event has to reach HubSpot first before SMS can react to it.
Email forgives a bad address with a bounce. SMS does not forgive a bad consent record, because the consequence is not a deliverability metric, it is a regulator. That is why the interesting engineering in SMS is never the sending.
Native SMS, a Telephony App, or Twilio Directly
Because HubSpot's native SMS runs on Twilio, teams often ask whether to connect a messaging provider themselves instead. There are three sensible shapes and they suit different problems.
Native HubSpot SMS is the right default for US and Canadian campaign and transactional sending. It is the only option where messages, consent and reporting live natively on the record with no glue in between.
A telephony or conversations app from the marketplace, of the kind Aircall, RingCentral, Dialpad and similar vendors publish, suits teams whose real need is conversational: reps texting from a shared inbox, inbound replies routed to an owner, calls and texts logged against the same record. These are built around a phone system rather than around a marketing send, so the strengths and the gaps sit in different places from the native add-on.
Twilio, or another provider, connected directly is what you reach for when the constraint is structural: international recipients, several sending numbers, or messaging triggered by events that never touch HubSpot. You are taking on the A2P registration and the consent model yourself, which is real work, and in exchange the ceiling moves.
For what the middle category costs and how it behaves generally, the guide to iPaaS covers the pattern.
HubSpot SMS Integration: Native vs iPaaS vs Custom
Buy the add-on, register the number, build nothing. Consent, quiet hours, timeline logging and workflow sending all come as standard, and reproducing that correctly yourself is more work than it looks. Most teams should start and finish here.
Wire a messaging provider to HubSpot events without writing code, which buys you international sending and a provider of your choice quickly. You are accepting a per-operation subscription that scales with message volume, and you now own consent state yourself, because the native subscription record is no longer in the path.
Built against the provider's API and the HubSpot API directly, so numbers can be provisioned per brand or region, consent has one owner and propagates everywhere that needs it, inbound replies route to the right record and the right person, and a failed send surfaces to a human rather than disappearing into a log. Fixed cost, no per-message meter on top of the carrier's.
When to Build a Custom HubSpot SMS Integration
Nobody should start here. The native add-on is inexpensive next to the compliance machinery it hands you for free. Build when you can name the structural boundary you are standing against.
Signs you have outgrown native SMS
Your customers are not all in North America. The clearest signal there is, because it is not a preference. Native SMS cannot reach them, so either the messages do not happen or they happen somewhere your CRM cannot see.
You need more than two sending numbers. Multiple brands, regions or teams. One marketing number for an entire portal means every brand borrows the same reputation, and one brand's complaint rate becomes everyone's deliverability problem.
Consent has to be true in more than one place. If a STOP in HubSpot does not stop your billing platform, your support desk or your app from texting someone, you do not have a consent record, you have a consent copy. This is the same field ownership problem described in the data mapping guide, with a regulator attached.
The trigger lives outside HubSpot. A shipment, a failed payment, an appointment change, a provisioning job. Getting that event into HubSpot purely so a workflow can send a text is a long way round, and it makes the message dependent on a sync that has nothing to do with messaging.
Replies need to go somewhere. Marketing SMS is built around sending. If an inbound reply has to reach a deal owner, open a ticket, or update a record rather than just sit in a channel, that routing is the integration.
Nobody can answer who was texted and why. The audit question. When someone asks for the consent basis and the message history for one phone number across every system that texts it, the answer should take seconds. If it takes a person a morning, that is the build.
One thing to know before you scope a build
Throughput on US SMS is not bought, it is scored, and this is the part that most surprises people coming from email.
Sending speed is measured in message segments per second, where a segment is up to 160 GSM-7 characters. Longer messages and non-GSM characters split into more segments, so a single emoji in a template can quietly double what every send costs you against your ceiling.
The design consequences are specific. Any send has to be queued and rate-shaped against a ceiling that can change when your vetting is re-run, so a naive loop over a contact list is not a workable architecture. Transactional messages need a reserved share of the daily cap or a marketing campaign will starve them on exactly the day it succeeds. And because the cap counts segments rather than messages, message length belongs in the design review, not in the copywriting pass.
Ask whoever is quoting for this work how they will keep a campaign send from consuming the daily brand cap that order confirmations depend on. It is a quick way to find out who has run SMS at volume.
What a Custom HubSpot SMS Integration Costs
The honest comparison is not build versus free. The carrier costs and the registration exist on every path, and the native add-on is genuinely cheap for what it removes from your plate. It is build versus the running cost of the alternative: an iPaaS metered per message on top of the carrier, a second messaging tool bought purely to reach customers outside North America, and the risk carried by consent state that only one system can see.
A custom integration is a one-time build plus ongoing maintenance, priced by scope. Sending templated messages from HubSpot events through one provider is a much smaller build than multi-number, multi-region messaging with a single authoritative consent record and inbound routing. The cost teams underestimate is never the build. It is maintenance, because carrier rules change, vetting gets re-run, and a messaging pipe that nobody owns is one regulation update away from being switched off.
With SMS you are not buying the ability to send a message. You are buying a permission structure, and the integration work is making sure your permission structure and your customer records never disagree.
That is why we productized it. StackTie builds the connection against your messaging provider's API and the HubSpot API directly for a fixed fee, then maintains it on a flat monthly retainer, so consent, numbers and message history stay consistent across every system that texts your customers.
Need to text a customer who isn't in North America?
When your buyers are outside the US and Canada, when several brands need their own sending numbers, when a STOP has to be respected by every system you run, or when the event that should trigger a message never touches HubSpot, that is a build. StackTie builds custom HubSpot SMS integrations for a fixed fee and maintains them on a flat monthly retainer, alongside the native add-on where it still fits. Live in 14 days or your money back. Book a free audit and we'll map exactly where your native option stops.
The Bottom Line
HubSpot's native SMS is a real product and a reasonable buy. It gives you consent handling, quiet hours, timeline logging and workflow sending without you having to think about any of it, and for a US or Canadian business running campaigns and confirmations, that is the correct answer and there is nothing to build.
What it is not is a messaging platform. You send from US numbers only, you get one marketing number and one transactional number for the whole account, sole proprietors are excluded, and how fast you can send is a score a third party assigns to your brand and shares across every campaign under it. Those are carrier facts wearing a HubSpot interface, and no configuration screen moves any of them.
So the question to ask is not whether HubSpot can send a text. It can. The question is whether the boundary you are pressing against is one you can configure or one you have to build past. Geography, sending identity and consent that has to be true in more than one system are all the second kind. Everything else is an add-on purchase and an afternoon.
