The Brief Standalone · Last updated July 2026
STIR/SHAKEN and call branding in 2026: the answer-rate fix mid-market operators keep deferring.
STIR/SHAKEN says who signed the call. It does not say who is calling. In 2026 the FCC closed the gap in three moves — annual RMD recertification, a KYC FNPRM adopted April 30, and a Call Branding FNPRM adopted October 29, 2025. The mid-market playbook is now four moves, and the vendor decision is coverage math against your buyer base.
8 min · The Brief · Standalone
Questions this article answers
- What changed in FCC STIR/SHAKEN and RMD rules in 2026, and what deadlines have already passed?
- Why does an A-attested call still land as "Spam Likely"?
- Which branded caller ID vendors reach which carriers and handsets?
- What KYC obligations do downstream buyers inherit from their voice provider?
- What does the October 2025 Call Branding FNPRM propose, and how should buyers scope for it?
- What operational controls actually protect a phone number's reputation from being flagged?
STIR/SHAKEN did what it was designed to do — it authenticates who signed a call. It does not tell the recipient who is calling, which is why answer rates for legitimate mid-market outbound still collapse under "Spam Likely" labels. In 2026 the FCC closed that gap in three moves, and the buyer playbook is now four.
The three regulatory moves stack in a single trailing twelve-month window. Annual recertification of the Robocall Mitigation Database became mandatory with a first deadline of March 1, 2026, and higher base forfeitures took effect February 5, 2026. The Commission adopted a Know-Your-Customer FNPRM on April 30, 2026 that would codify A/B/C attestation criteria and require originating voice service providers to verify customer name, physical address, government-issued ID, and right-to-use before granting A-attestation. A companion Know-Your-Upstream-Provider FNPRM followed on May 20, 2026. Comment and reply windows on the KYC rulemaking closed June 25 and July 27, 2026. Meanwhile, the October 29, 2025 Call Branding FNPRM proposes that terminating providers display a verified caller name for A-attested calls. Together, those three moves shift branded caller ID from a marketing lever to a regulatory foundation.
What attestation means in practice for a mid-market outbound program
A, B, and C describe how much the originating provider knows. TransUnion's plain-language explainer is the reference: A means the provider knows the customer and the customer's right to use the number; B means the provider knows the customer but not the number-right; C means the provider is merely an entry point, most common on international gateways. A mid-market outbound program signing new voice contracts in the second half of 2026 should treat A-attestation as a written deliverable in the MSA, not an aspiration — and should demand per-call attestation reporting the carrier can actually produce.
Why an A-attested call still gets tagged "Spam Likely"
Attestation and labeling are separate systems. Attestation is a SIP header signed at origination. Spam labels are analytics-engine calls made downstream by Hiya, TNS, First Orion, and the terminating carrier's own scoring pipeline based on velocity, complaint rates, and list hygiene. A clean A-attested call can still be labeled if the number's reputation breaks a threshold. Branded caller ID does not override reputation. It renders business name, logo, and — where the carrier supports it — reason-for-call on the recipient's handset when the reputation and attestation both hold. The two systems compound rather than substitute.
Which vendor reaches which carrier
Vendor selection is coverage math against the buyer base, not a feature grid. TransUnion TruContact Branded Call Display launched in-network with AT&T and added reason-for-call for AT&T wireless subscribers in February 2025. Hiya Connect Branded Call distributes through Samsung, AT&T, and T-Mobile relationships; the vendor cites 500M+ mobile handsets in coverage. First Orion INFORM Branded Calling positions as a carrier-integrated, multi-carrier display service with vendor case-study lifts up to 34 percent (treat as directional). TNS Enterprise Branded Calling deploys through its carrier-network footprint. Numeracle orchestrates Google Verified Calls, Hiya Connect, and TNS from a single management surface, which is the shape most useful when the buyer base spans multiple carriers and the buyer wants one contract to enforce coverage. TransNexus, Bandwidth, and iconectiv operate the underlying voice-provider infrastructure — authentication, RMD filing, SPC token administration — rather than buyer-facing branding.
The KYC file the buyer inherits
The KYC FNPRM's practical effect is not a new form to file. It is that the evidence your voice provider holds on you now determines whether your calls are eligible for A-attestation. Per the Wiley alert, A-attestation under the proposed rule requires KYC plus verified right-to-use of the calling number; B keeps KYC and drops number verification. A buyer that cannot produce right-to-use documentation for every DID in the outbound program should assume the pool of A-eligible numbers will shrink when the rule finalizes. Enumerate the evidence — name, address, government ID, right-to-use per DID — and store it in the same folder as the RMD certification. That folder is what an FCC inquiry asks for.
The number-reputation program that keeps answer rates from collapsing
Reputation is managed the way marketers manage sending domains. Cap daily volume per DID in the first fourteen days after activation. Monitor spam-labeling against a reputation dashboard — First Orion Ferret, Numeracle Reputation Cloud, TNS Call Guardian — and rotate DIDs on a scheduled cadence rather than reactively after a flag. For any call using generative or cloned voice, publish the AI-voice disclosure at call open and hold Prior Express Written Consent per the FCC's 2024 Declaratory Ruling under TCPA 47 U.S.C. § 227(b). The 2024 Enforcement Bureau sweep removed 1,388 non-compliant providers from the RMD; that removal is silent and immediate downstream, and the outbound program discovers it via a sudden answer-rate cliff, not a notice.
The buyer artifact
Answer-rate preservation checklist
- RMD posture. Confirm every originating carrier in the voice stack is registered and recertified in the RMD at fcc.gov/robocall-mitigation-database. Set a March 1 recertification reminder for every carrier annually.
- Written attestation policy. Ask each carrier what triggers A, B, and C on your traffic, and whether per-call attestation reporting is available in the customer portal.
- KYC file. Enumerate name, address, government ID, and right-to-use documentation the carrier holds on you, per DID. This is what protects A-eligibility under the pending rule.
- Branded caller ID selection. Map your buyer base to carrier coverage. TransUnion for AT&T reach, Hiya for T-Mobile and Samsung, First Orion for multi-carrier lift, TNS for network-level footprint, Numeracle as an aggregator across all four.
- Number-warming and rotation. Cap daily volume per DID for the first fourteen days; rotate on a scheduled cadence; monitor labels on First Orion Ferret, Numeracle Reputation Cloud, or TNS Call Guardian.
- AI-voice consent. For any generative or cloned voice, publish disclosure at call open and hold Prior Express Written Consent per FCC 2024 Declaratory Ruling.
- Audit folder. RMD certification, branded caller ID contract carrier coverage list, reputation dashboard export, and AI-call consent evidence in one place.
Where these programs actually break
Failure mode one. Branded call display silently drops on carriers the vendor does not cover. A vendor may reach AT&T, Verizon, and T-Mobile at the parent-carrier level but not every MVNO subscriber; the branded frame disappears on unsupported handsets and the call renders as generic caller ID with no error surfaced. Contract for a per-carrier and per-MVNO coverage list, updated quarterly.
Failure mode two. A-attestation is not portable. If a call transits an intermediate carrier that re-signs at B, the terminating carrier's analytics engine may downgrade the trust score. Buyers who assume "originating A equals displayed A" get surprised. Ask the carrier for the intermediate-hop policy in writing, and audit a sample of live calls through a per-call attestation report.
Failure mode three. Reputation labels are analytics-engine calls, not FCC calls. Even a clean, A-attested, branded call can be tagged "Spam Likely" if velocity, complaint rate, or list hygiene breaks a threshold. Branded ID does not override reputation. Own the reputation dashboard the same way the marketing team owns sender score, and treat rotation as scheduled maintenance, not incident response.
Three rules for a mid-market operator
Rule one. Do not treat attestation and branding as substitutes. STIR/SHAKEN answers "who signed this call." Branded caller ID answers "who is calling." Both are required to hold answer rates in 2026.
Rule two. Buy branded caller ID against the carrier mix of your actual buyers, not the vendor's marketing claims. A vendor with strong AT&T coverage delivers nothing to a book that dials T-Mobile.
Rule three. Assume the Call Branding FNPRM becomes a rule. Scoping the KYC file, the RMD posture, and the branded caller ID contract now is cheaper than retrofitting after a final rule drops.
In short
- STIR/SHAKEN authenticates who signed the call. Branded caller ID renders who is calling. Mid-market operators need both to hold answer rates in 2026.
- First annual RMD recertification hit March 1, 2026; removal is silent and immediate downstream and every downstream carrier must block the traffic.
- The April 30, 2026 KYC FNPRM would tie A-attestation to verified customer identity and number right-to-use; scope the KYC file now.
- The October 29, 2025 Call Branding FNPRM would require terminating carriers to display verified caller name for A-attested calls.
- Vendor selection is coverage math: TransUnion for AT&T, Hiya for T-Mobile and Samsung, First Orion multi-carrier, TNS network-level, Numeracle as an aggregator.
- AI-voice outbound is a robocall under TCPA regardless of intent per FCC 2024 Declaratory Ruling; Prior Express Written Consent required.
Sources
- FCC, “Robocall Mitigation Database,” program page. fcc.gov
- FCC, “Fact Sheet: Improving Verification & Presentation of Caller ID Information (Call Branding FNPRM),” adopted Oct. 29, 2025. docs.fcc.gov
- FCC, Enhancing Know-Your-Customer Requirements NPRM (FCC-26-32), adopted April 30, 2026. docs.fcc.gov
- Federal Register, “Enhancing Know-Your-Customer Requirements,” May 26, 2026. federalregister.gov
- The CommLaw Group, “First Annual RMD Recertification Due March 1, 2026.” commlawgroup.com
- TransUnion, “What are the attestation levels for STIR/SHAKEN?” transunion.com
- Wiley, “FCC Proposes Stronger Know-Your-Customer Rules,” alert. wiley.law
- Davis Wright Tremaine, “FCC Proposes Expanded KYUP and STIR/SHAKEN Standards.” dwt.com
- Mac Murray & Shuster, “The FCC's Call Branding FNPRM at a Glance.” mslawgroup.com
- TransNexus, “FCC adopts proposed rules for presentation of caller name.” transnexus.com
All linked sources were live at time of publish (July 2026). Verify before quoting in a procurement document.
Want a calibration on your outbound compliance posture?
Run the Tier 1 benchmark.
Submit your current voice-carrier contract, RMD status, and DID inventory. We return a benchmark PDF in five business days showing where your A-attestation eligibility, branded caller ID coverage, and number-reputation program hold — and where they break under the pending KYC and Call Branding rules. Free. No follow-up sales drip.
Run the Tier 1 benchmark →Related reading
See Voice AI agents in 2026: where it works for the consent posture around generative-voice outbound. The anchor on Where AI actually fits in your tech stack frames the voice layer inside the broader AI decision. And the Cardinal Method explains how buyer-side compensation isolation works in practice.