Last updated 30 August 2026. Compiled while evaluating quotes from three aggregators for a US launch. Structure and public schedules only — no vendor's confidential pricing appears here.
Ask an aggregator what RCS costs and you will get a price per message and a price per session. Those are the numbers you asked for, and for a business starting out they are rarely the numbers that decide anything.
The fees that decide it are recurring, per brand, and charged whether you send anything or not. If you run messaging for several brands — an agency, a platform, a company with multiple products — they scale with your customer count rather than your volume, which is the opposite of how you probably modelled it.
Here is what to ask about before you sign, roughly in order of how much it will hurt.
Every brand needs its own agent. That is not a platform limitation; it is the point. The agent is what carries the verified name, logo and colour, and a shared agent would mean every brand messaging under someone else's identity — which discards the one thing RCS sells.
So a per-agent monthly fee is a per-brand monthly fee. At ten brands it is ten times. At fifty, fifty. Compare it against your expected traffic per brand: below roughly a couple of hundred thousand messages a month, a fee of a few hundred dollars per agent can exceed everything that brand spends on actual messages.
Two questions worth asking directly:
At least one major US carrier charges a one-time activation fee for RCS campaign service. Published schedules state the amount and frequently do not state the unit.
That single missing word is the difference between a few hundred dollars and a five-figure line item at fifty brands. Ask explicitly, and get the answer in writing rather than in a thread.
Verified sender status is vetted, and the vetting is a fee — usually annual, usually modest, and sometimes charged per image you submit. A logo and a banner is two.
Modest individually, and it multiplies by brands and then by years. Budget it as a recurring per-brand cost rather than a setup cost, because that is what it is.
This one is not a fee until it is. Brand vetting can carry a validity window — commonly around 45 days — within which a two-factor verification must be completed by a named contact. If that window closes, the submission is void and the fees apply again. At least one schedule states plainly that the decision is final and cannot be appealed.
So the cost of a missed email is the whole vetting fee, plus the weeks. Put a calendar reminder against every brand submission, and make sure the contact address is a mailbox somebody reads.
Which raises a failure worth its own paragraph: the authorisation request goes to a point-of-contact address, and that address is often on the brand's own domain. If the domain uses registrar forwarding rather than a real mailbox, it will forward web traffic and quietly receive no mail at all. The request goes nowhere, nobody sees an error, and the window closes. Check that the address actually receives mail before you submit — not that it looks right.
Carriers fine for prohibited content. Published tiers run from several hundred dollars for a general violation to several thousand for phishing or social engineering, and aggregators reserve the right to pass those charges on.
If you host brands whose messages you do not write, you are carrying that risk on their behalf. Two things follow, and both are cheaper before you sign than after:
Carrier pass-through fees differ substantially between US networks — published schedules show per-session fees varying by more than a factor of six between the cheapest and most expensive, and the expensive ones are not the small networks.
You cannot route around this. Your recipient's carrier is a property of your recipient. Two identical conversations can differ several-fold in cost, so your blended rate depends on the carrier mix of your audience — which means a pilot with one audience can seriously mislead you about the economics of another.
Ask for the pass-through schedule and compute a blended figure against your own list rather than accepting a headline rate.
Each brand needs a legal entity, a reachable website with a privacy policy and messaging terms, a described use case, a named contact with a title, an opt-in description, and often a video or written instructions for a human reviewer. Google's brand review is quoted at around 15 days when nothing needs correcting. Carrier launch is separate and slower.
And a constraint that surprises people: domains younger than about 90 days are frequently auto-rejected. If a brand registered its domain last month, no amount of paperwork shortens the wait.
A rejection costs weeks, so the cheapest thing you can build is whatever catches a bad submission before it is submitted.
For a single high-volume brand, the rate card is most of the bill and the fees above are noise.
For a platform serving many smaller brands, it inverts: recurring per-brand fees can be several times the traffic cost, and the per-agent line alone can decide whether the business works. If you are in the second category, negotiate that line and treat the rate card as secondary — which is the reverse of how these conversations usually go, and the reason they usually go badly.
One API and an MCP server for RCS with automatic SMS fallback. Brand onboarding, carrier readiness and the checks in this article are built in, so the failures above are reported rather than discovered.
Get a free sandbox key — no card, nothing to verify first — or read the agent-facing reference.
Its companion, on the rate card itself: why your bill will not match your quote. And before any of it matters, why an approved agent still reaches nobody.