Your texting bill has four layers. The quote shows one.

The line item is small enough that nobody audits it. A dental group texts appointment reminders. A home-services franchise texts arrival windows. A restaurant group texts waitlist positions. Somewhere a provider quoted a fraction of a cent per message, someone put that fraction next to another fraction, and the contract was signed on the smaller number. The structural fact is that the quoted rate is one of four charges paid to three different parties, and the three that go unquoted behave nothing like the one that does. Volume discounts move the first layer and do not touch the rest.

A US messaging bill is four charges with three payees.

Layer one is your provider's per-segment rate, the number on the quote. Layer two is a per-segment carrier pass-through charged by AT&T, T-Mobile, Verizon and the regional carriers for delivery onto their networks, which Twilio documents as applied on top of standard US messaging pricing. Layer three is one-time registration with The Campaign Registry, covering brand creation and whatever vetting your brand type requires. Layer four is a recurring monthly subscription for every campaign use case you register, payable whether you send one message or a million. A quote showing only layer one is not dishonest so much as incomplete, and the incompleteness always runs one direction.

You cannot buy registration at cost, because brands are not permitted to register directly.

The Campaign Registry says so on its own homepage: direct registration is not available for brands, and a business wanting to use SMS must work through a registered Campaign Service Provider who handles registration on its behalf. That constraint sets the economics of layers three and four. The registry publishes a fee schedule; a mandatory intermediary sets its own rate card on top of it; the buyer sees only the second number, with no native way to separate registry cost from provider margin. What makes the spread visible is that some providers publish. Bandwidth and Vonage both post full 10DLC fee tables in public support documentation, and read together they show the identical registry standard-campaign tier carried at monthly rates ranging from roughly $10 to $20, and the Agents and Franchises tier from roughly $30 to $60. A rate card is a negotiable surface once you know one exists.

Use-case classification moves your recurring cost more than volume does.

The campaign use case chosen at registration sets the monthly subscription, and the spread across tiers is wider than most volume discounts a mid-market buyer will ever win. Published provider schedules put a low-volume mixed campaign near $1.50 to $3 per month, standard use cases near $10 to $20, and the Agents and Franchises use case near $30 to $60. For a single site that is a rounding error. For a forty-location franchise system registering a campaign per location, the same decision separates roughly $1,400 a year from roughly $29,000 on identical traffic. It is usually made in about six seconds by whoever performs the registration, which in mid-market deployments is the provider or the scheduling, CRM or field-service application doing the sending. A field carrying that much operating leverage should not be filled in by someone with no stake in the outcome.

Two adjacent mechanics are documented and almost never read. Vonage's fee article states that the $15 campaign vetting fee is charged per vetting event, and that a campaign returned for updates after initial review is charged another $15 for each subsequent event, turning a sloppy first submission into a cash cost rather than a delay. Bandwidth's fee article states that all campaigns other than the political use case carry an initial three-month commitment and are billed for the full three months even if cancelled earlier. Campaign count is a committed cost with a floor, not a dial you turn down mid-quarter.

Unregistered traffic is not cheaper. It runs roughly triple.

The intuition that skipping registration saves money is backwards. Vonage's published pass-through table lists compliant 10DLC SMS at $0.003 per segment on T-Mobile and AT&T and $0.004 on Verizon, against unregistered gray-route traffic at $0.008, $0.010 and $0.010, roughly two-and-a-half to three times the registered rate. Twilio's documentation makes the point from the other side: senders using a 10DLC number without registering receive additional carrier fees, plus heavier filtering and lower throughput. The realistic failure mode is not a decision to skip registration. It is a campaign that lapsed, a brand left unverified after a rejected submission, or a new location sending on a number never attached to an approved campaign. The pass-through penalty surfaces on an invoice ninety days later. The deliverability loss surfaces as reminders that never arrived and a no-show rate nobody traces back to a registration record.

Seven lines to normalize before you compare two messaging quotes.

Paste this into the renewal file or the RFP and require both bidders to answer in the same units. It is deliberately vendor-agnostic and works equally against a CPaaS provider, a UCaaS platform re-billing messaging, or a vertical SaaS application burying texting in a per-location fee.

Red flags in the answers: a single blended per-message price with no pass-through line; an inability to name your registered use case; campaign fees described as "carrier fees"; no stated throughput ceiling; a per-location fee for messaging with no campaign count behind it.

Why this is structural rather than a knock on any provider.

Every US sender meets the same architecture because the carriers built it. Long codes were designed for person-to-person traffic, were abused at scale, and the carrier response was a registry that verifies who is sending and what before traffic flows. Verification costs money, the registry charges for it, and the registry does not sell to brands, which makes the intermediary layer mandatory rather than optional. No provider can remove layers two through four and none of them invented the shape. Buyer leverage lies in seeing all four, knowing which are fixed and which are a rate card, and getting classification right before the first invoice rather than after the twelfth. Same discipline that surfaces unused licenses in a telecom expense audit and bundled features in a UCaaS AI review.

How Cardinal prices a messaging line.

When a client renews a CPaaS agreement or evaluates a platform that bundles messaging, we rebuild all four layers at the real segment mix and location count, benchmark the negotiable surface against the market rather than the first quote, and check registered brand type, use case and campaign count against how the organization actually messages. The reclassification finding is common and usually pays for the exercise on its own. The output is a like-for-like annual number.

See the Cardinal Method →  ·  UCaaS vendor selection →  ·  See the supplier pool →

Three rules, in order. Never compare messaging quotes on the per-message rate alone, because it is the one layer designed to be compared. Read your own registration before you read anyone's rate card, because the use case field carries more annual dollars than the negotiation will. And treat campaign count as a committed cost with a minimum term, not a variable expense, because the published fee schedules say exactly that and the invoice will enforce it.