FCC Robocall Scorecard Follows Years of Provider Purges

The FCC on September 2, 2026, sought comment on a robocall scorecard for phone companies and cut 14 providers off U.S. networks the same day. The Consumer and Governmental Affairs Bureau published the plan as DA-26-932 in CG Docket 26-239, with comments due September 22, 2026, and replies due October 2, 2026.

Chairman Brendan Carr framed the grade as the next consumer tool in a yearslong fight. The bureau was careful about what the tool is not. It is not a new rule, and it is not a finding that any carrier broke the law.

Fourteen Providers Were Cut Off the Same Afternoon

While the consumer bureau asked how to rate retail phone companies, the Enforcement Bureau finished a much older file. Acting Chief Hunter Deeley signed an order that same day removing 14 certifications from the Robocall Mitigation Database, the public roster a provider must stay on to hand traffic to the rest of the U.S. network.

The FCC’s own release said fourteen companies will be blocked from connecting to U.S. phone networks. Other carriers have two business days to stop taking their traffic, with 911 and public-safety calls carved out. The companies may not refile unless both the Enforcement Bureau and the Wireline Competition Bureau consent.

The 14 are not household names. They include Skycom Healthcare, Dixie Net Communications, Conference America, Opex Communications, Apps Communications, and Reachme.com Inc, along with CFX Business Solutions, Convergence Technology Solutions, CSB Technologies, Digital Division, HIGHCOMM, Inatech Solutions, makrodepot, and a filer listed as SECURE. They were among 35 firms told on March 24, 2026, to fix deficient filings. None of these 14 answered.

The paper trail is longer than one missed deadline. Wireline staff flagged the same certifications on March 29, 2024, after expanded database rules took effect on February 26, 2024. A December 10, 2024, show-cause order named 2,411 filers. An August 25, 2025, order then stripped 1,203 certifications, including these companies. After that purge, the bureaus provisionally put some of them back to limit service disruptions, then warned them again, then dumped them on September 2, 2026. Five notices, still no cure.

Combatting the scourge of illegal robocalls remains the FCC’s top consumer protection priority. And since I became Chairman, we have been tackling the problem at every point of the call path.

Brendan Carr, Chairman, FCC news release, September 2, 2026

Carr followed the cutoff on September 3, 2026, writing that the agency had blocked an additional 14 providers from U.S. networks for failing robocall mitigation rules. That post, not the scorecard notice, is the piece of the day’s work that drew the heavier public argument, because a delisting actually stops traffic.

The Notice Creates No New Carrier Duties

The scorecard sits in a different lane. The bureau’s public notice seeking comment on a scorecard says, in plain terms, that this is not a rulemaking that will result in new rules or requirements for voice service providers. The FCC would publish a digestible comparison on its website and update it from time to time.

Carr’s public case is that providers should have to compete on blocking, the way they already compete on price and coverage. He put that wager on the record the morning the notice went out.

https://x.com/BrendanCarrFCC/status/2095151342167716296

The bureau wants a composite that looks past an administrative checklist, whether the company filed paperwork and offered a blocking toggle, and toward practices plus results, including how often legitimate calls are blocked. It also draws a hard line the other way. The score is not a verdict on compliance with robocall rules, which the notice calls a case-specific legal inquiry. A carrier could post a weak grade and still be inside the rules, or post a strong grade and still face a separate enforcement case.

Commissioner Anna Gomez said the scorecard could increase consumer awareness of how providers fight illegal robocalls if it includes the right information. Teresa Murray, consumer watchdog director at PIRG, said the agency’s efforts have been disappointing for years, going back to even before Congress passed the robocall crackdown law in 2019, and that companies should have to put the grades on websites, bills, or marketing materials.

A grade on an FCC page only moves a customer who can see it and can leave. Wireless contracts do not work like a grocery aisle. If the score lives in a docket folder, the person who already cannot stand the 8 a.m. loan pitch never meets it.

How the FCC Might Score a Carrier

The notice splits the yardstick in two. Conduct metrics ask whether a provider took named steps, even if those steps did not cut the calls. Outcome metrics ask whether customers actually got fewer illegal robocalls, and whether the filter ate real calls on the way.

CONDUCT METRICS VERSUS OUTCOME METRICS

The bureau asks whether to mix numbers with qualitative judgments, how many metrics are too many, and whether traceback help for someone else’s customers should count. It also asks whether 100 complaints should weigh more against a small company than against a national wireless brand.

Display is still a blank page. Commenters are asked whether to publish each metric, one composite, risk tiers (low, medium, high), a score such as 96 out of 100, a scale such as 4 out of 5 or 7 out of 10, letter grades from A through F, or medal bands such as platinum, gold, or silver. A second fork is absolute versus relative: a fixed bar every carrier can clear, or a curve that ranks them against each other.

That last choice decides whether the scorecard is a safety inspection or a league table. A curve can make a decent carrier look poor in a strong year. A fixed bar can give a pack of large carriers the same badge while customers still hear the same pitch.

STIR/SHAKEN Did Not Empty the Voice Mail

The grade is arriving after the technical mandates that were supposed to make it unnecessary. Congress signed the TRACED Act on December 30, 2019. The Robocall Mitigation Database opened on April 20, 2021. STIR/SHAKEN, the caller ID signing system, is now required on IP voice networks, with listed exemptions. Database rules that took effect on February 26, 2024, demanded a mitigation plan, a 24-hour traceback pledge, and a statement of how far STIR/SHAKEN actually runs on the network.

THE PATH TO A PUBLIC GRADE

1. December 30, 2019: TRACED Act is signed, expanding FCC tools against illegal robocalls and spoofed caller ID.
2. April 20, 2021: Robocall Mitigation Database opens; staying listed becomes the price of handing traffic to other U.S. providers.
3. February 26, 2024: Expanded certification rules take effect, including mitigation plans and a 24-hour traceback commitment.
4. December 10, 2024: Enforcement Bureau orders 2,411 filers to cure deficient certifications or explain why they should stay listed.
5. August 25, 2025: Bureau removes 1,203 certifications after those filers fail to cure, then provisionally restores some of them.
6. September 2, 2026: Bureau seeks comment on a public scorecard and, the same day, removes 14 reinstated filers again.

Installation still is not universal. PIRG Education Fund reviewed database listings as of September 28, 2025, and found that of 9,242 phone companies on file, 4,084 have completely installed the required robocall-fighting software, 44% of the file, down from 4,365 the year before. Another 1,696 had it on part of the network.

Call volume moved more than the filings did, and not always in a straight line. YouMail, whose Robocall Index estimates U.S. volume from traffic to its own users, counted 3.9 billion robocalls in August 2026, 125.4 million a day, down 10% from July. That is below August 2025’s 4.15 billion, and well below the 58.5 billion YouMail logged for all of 2019. It is still a pile of ringing phones.

U.S. ROBOCALL VOLUME BY YEAR

YouMail’s history series shows 32.7 billion robocalls so far in 2026, about 99.6 per person affected. A public carrier grade is being designed in a country where the annual total fell from the 2019 peak and then stuck near 52 billion, with August still near 4 billion. The database can exile a small filer. It has not given a household a way to tell Verizon from a regional VoIP shop before the next loan pitch.

Only Retail Providers Would Get a Grade

The bureau proposes to rate domestic voice providers that have retail customers, across wireless, wireline, and VoIP, including hybrid networks. Wholesale shops and intermediate carriers that never bill a household would sit out, including gateway providers that only sit in the middle of the call path.

It does not propose to rate every retail carrier. The aim is the set that together serves the vast majority of the retail market. Staff asks whether to pick the largest by subscribers, whether to use the agency’s Form 477 collection, and whether small or regional providers should be left off or scored on a different scale because of burden, traffic mix, and missing data.

Leave the small carriers off, and the scorecard becomes a poster for the national wireless brands that already sell spam-filter apps. Put them on, and a rural co-op with thin analytics can look worse than a nationwide network that sees a different load of junk. The notice flags that problem and does not solve it.

The inputs are a second fight, because whoever owns the data owns the grade.

THE DATA THE BUREAU WANTS TO MINE

• Database filings: Robocall Mitigation Database entries, which today vary in format and would have to be squeezed into comparable fields.
• Carrier tools: Blocking and labeling products the provider sells or gives to its own retail customers.
• Network filters: The provider’s own blocking practices, tracked over time, including mistakes.
• FCC complaints: Consumer Complaints Center records, which measure who bothered to file, not every call that rang.
• Enforcement files: Commission actions against a provider, which can lag the live traffic by months.
• Industry feeds: Third-party analytics firms and other private scores, if staff decides those scores are fair.
• Tracebacks: Industry Traceback Group work, whose underlying data is not public.
• FTC records: Aggregate Federal Trade Commission complaint data from the Consumer Sentinel Network.

Staff asks whether carriers are the only source for some of this, whether analytics vendors are unbiased, and whether any private score should be imported wholesale. A vendor that already sells labeling to one wireless group has an obvious interest in how that group lands. Traceback data can show who handled a suspected illegal call and still say nothing about a carrier’s retail filter. Complaint totals can punish the company whose customers are more likely to file.

Mistaken Blocks Are Part of the Grade

The notice keeps returning to one measurement the older mandates never put in public: the false-positive rate, the share of legitimate calls a filter kills. Schools, pharmacies, local governments, and political campaigns have spent years on the other side of “Scam Likely.” If outcome scores reward a low count of calls that reach the handset, the cheap way to win is to over-block.

The bureau says blocking data tracked over time could show which methods stop illegal traffic without eating the real thing, and could feed industry best practices. It also admits the scorecard will not be comprehensive, because a retail provider does not control every hop in a call’s life. A beautiful grade can sit on top of junk that entered the country through someone else.

Conduct scores have the mirror problem. A company can attest SIP calls, answer tracebacks in 24 hours, and ship a blocking app, then still pass through loan-scam campaigns that punch a hole in the feature list. Report-and-delete buttons on flagship wireless plans have not stopped those campaigns from reaching the handset. That is the gap an outcome column is meant to catch, if the data can be compared across carriers of different sizes.

Murray wants the grade where a customer already looks, on the bill and the ad. The bureau, for now, is talking about a page on fcc.gov. Those are not the same product.

CG Docket 26-239 Is Where the Grade Gets Built

Comments are due September 22, 2026. Replies are due October 2, 2026. The file is CG Docket 26-239. Nothing in the notice sets a launch date, a first roster of carriers, or a winning display format. Staff says the scorecard should be easy to read, drawn from sources whose limits are disclosed, and built so it does not inherently punish one kind of provider, then improved in later rounds.

The September 2, 2026, docket is the public half of a purge that has already taken 1,203 certifications off the database in a single August 2025 order and 14 more after those companies were let back in. The private half still runs through delistings and traffic blocks. The new piece is a grade that households can see, attached to no new legal duty, aimed at a market that still took 3.9 billion robocalls in August.

Carr said providers should be competing on this metric. The comment file is where carriers, analytics firms, and consumer groups will argue what the metric is. Until that argument ends, the mute button remains the only score that is already on the phone.