If your customers send texts to their own contacts, give each one a sub-account with its own numbers and registration. Use a single shared sender only when the platform itself is the sender, such as login codes or product notices that come from your brand. The two models differ most in registration, opt-outs, billing and what one bad actor can break.
Plenty of platforms start shared because it's quicker, then regret it once customers want their own name on the text. Here's the trade, area by area, so you can decide before you build.
The two models
Sub-account per customer. Each customer is a sub-account with its own numbers and carrier registration. Your platform creates them through the API (POST /v1/sub-accounts) and lists them with GET /v1/sub-accounts. See the sub-accounts docs.
Shared sender. All customers send from one number or pool of numbers, registered once under the platform's brand. Customers never see a number of their own.
Registration
Carriers register a brand, the business behind the texts, and a campaign describing what it sends. With a shared sender, the brand is your platform, and every customer's messages ride on your registration. That works if your platform is genuinely the author of the texts. It strains when a plumber's customers get a message that's really from the plumber, but registered to you.
With sub-accounts, each customer registers their own brand. Ready registers businesses with an EIN, so each customer needs one. Registration is $25/mo per customer, including the brand and its first campaign. Extra campaigns are $20/mo each, and each number is $2/mo. Review is the slow part, though most approvals take 2 to 7 days. For the structure, see 10DLC for SaaS platforms.
Opt-outs
A person who texts STOP to a number expects to stop hearing from that sender. With sub-accounts, the answer follows the structure: a STOP always blocks that number in the sub-account that received it. Whether it also blocks the parent and the other sub-accounts depends on a setting, so don't assume. Check GET /v1/opt-outs/:phone before a send when you're unsure.
With a shared sender, a STOP is a STOP for everyone on that number. One customer's heavy-handed campaign can opt people out of every other customer's messages. That's rarely what the other customers want, and they won't know why their list shrank. Our SMS opt-out API guide covers the mechanics.
Billing
| Feature | Sub-account per customer | Shared sender |
|---|---|---|
| Registration cost | Per customer, which you can pass on | One, absorbed by you |
| Number cost | Per customer number | One pool |
| Usage attribution | Natural, one sub-account per bill | You must tag and meter every send yourself |
| Pricing plans | Easy to pass through or mark up | Harder to explain per customer |
With sub-accounts, usage lines up with customers without extra work. With a shared sender you build the metering. Our posts on metering SMS usage per customer and fixed-cost 10DLC pricing for SaaS go deeper.
Deliverability
Carriers judge a sender by its history. On a shared number, one customer sending a poor list or spammy copy damages the reputation every other customer relies on. A single complaint pattern can get the shared registration scrutinized, and everyone's texts feel it.
Sub-accounts contain the damage. A problem customer gets filtered or paused, and the rest keep sending. That isolation is the main reason to choose them for a multi-tenant product. Each customer also builds its own reputation, which is a benefit once they send well.
When a shared sender is the right call
Sub-accounts aren't always better, and a shared sender has real advantages: nothing to register per customer, nothing to wait on and one number to manage. Shared is fine when:
- The platform authors every message, such as codes, alerts and receipts, and the customer's contacts expect them from you.
- Volume per customer is too low to justify a registration.
- You're prototyping before you commit to a structure.
Even then, keep marketing and transactional traffic apart. Our post on separating MFA from marketing numbers explains why.
Offboarding and support
The model also changes what happens when a customer leaves or complains. With a sub-account, you can pause or close one customer and their numbers without touching anyone else, and support can answer "why didn't my text arrive" by looking at one account's history. On a shared sender, every support question means digging through traffic from all customers, and a departing customer leaves no clean boundary to cut along.
A quick decision guide
Work through these in order. The first honest yes usually settles it.
- Does the customer write or choose the message content? Use a sub-account.
- Does the customer have an EIN and a website? Sub-accounts work.
- Do you need per-customer billing and reports? Sub-accounts.
- Are all the texts authored by your platform, with no customer-specific content? A shared sender can work.
- Will customers expect their own business name on the text? Sub-accounts.
You can also mix them: a shared sender for platform notices and a sub-account for each customer who texts their own contacts.
The takeaway
For a platform that texts on behalf of its customers, sub-accounts cost more up front and save you in opt-outs, billing and blame. Read our multi-tenant SMS architecture guide for the build, then try the model with the API quickstart.
Accounts opened before November 12, 2026 keep their signup prices, shown under Billing.