Who Makes the Call? TCPA Liability for Texting Platforms and Users
The TCPA puts its robocall liability on whoever will “make” or “initiate” the call, and it never says who that is when a business texts through someone else’s software. The FCC’s answer since 2015 looks mostly at who controlled the message’s content, its timing, and its recipients. Courts have used that test to let some platforms out of lawsuits and keep others in. For a business that texts through a platform, those facts largely decide whether the platform shares the exposure or the business carries it alone.
What does it mean to “make” or “initiate” a call under the TCPA?
In the FCC’s reading, it means placing the call yourself, or being so involved in placing it that you’re treated as if you had. The statute makes it unlawful “to make any call” to a cell phone with an autodialer or a prerecorded voice without consent, and “to initiate any telephone call” to a residential line with an artificial or prerecorded voice, under 47 U.S.C. § 227(b)(1). A prerecorded call plays a recorded or artificial voice instead of a live person. An autodialer needs “the capacity to use a random or sequential number generator to either store or produce phone numbers to be called,” as the Supreme Court held in Facebook, Inc. v. Duguid.
Neither word is defined. In its 2013 Dish Network ruling, the FCC concluded that a party initiates a call “when it takes the steps necessary to physically place a telephone call,” and said the term generally doesn’t reach parties that “might merely have some role, however minor, in the causal chain that results in the making of a telephone call,” according to FCC 13-54. The Do Not Call rule uses the same verb: “No person or entity shall initiate any telephone solicitation” to a residential subscriber whose number is on the National Do Not Call Registry, the federal list of numbers whose owners opted out of telemarketing, under 47 C.F.R. § 64.1200(c).
How did the FCC apply that test to messaging apps?
It asked who decided whether, when, to whom, and what to send. In its July 2015 omnibus order, the FCC said it would weigh all the circumstances to decide “who took the steps necessary to physically place the call” and “whether another person or entity was so involved in placing the call as to be deemed to have initiated it,” according to FCC 15-72. It then applied that test to three app makers.
| Scenario in FCC 15-72 | Who made the key choices | FCC’s conclusion |
|---|---|---|
| YouMail texts a user’s auto-reply to someone who left a voicemail | The user picks whether to reply, who gets replies, and most of the wording | YouMail is not the maker |
| Glide automatically texts invitations to every contact in a user’s address book | Glide picks the recipients and the words, with little or no user input | Glide is the maker in that scenario |
| TextMe sends an invitation after the user taps through several screens and picks contacts | The user decides whether and when to send and to which contacts, but TextMe writes the message | The user is the maker |
The order states that “YouMail is not the maker or initiator of the text because it does not control the recipients, timing, or content,” according to FCC 15-72. In Glide’s automatic-send scenario, “the app user plays no discernible role in deciding whether to send the invitational text messages, to whom to send them, or what to say in them.” On TextMe: “These affirmative choices by the app user lead us to conclude that the app user and not TextMe is the maker of the invitational text message.” The FCC still called TextMe’s control of the wording “a reason for concern.” Our post on refer-a-friend texts applies those rulings to a business’s own invite program.
The same order flags other facts that matter to platforms. One is willfully enabling fraudulent spoofing of telephone numbers or helping telemarketers block caller ID. Another is “whether a person who offers a calling platform service for the use of others has knowingly allowed its client(s) to use that platform for unlawful purposes,” per FCC 15-72. If FCC staff warns a platform that clients are using it unlawfully and the platform lets that use continue, the agency will consider that a possible sign the platform is “actively participating in the making or initiating of the calls at issue.”
When have courts kept a texting or dialer platform in the case?
When the platform itself sent messages nobody else chose to send, ran the campaign, or is alleged to have done so. At the pleading stage, an allegation that an intermediary knew of and was involved in an unlawful campaign has also been enough, even without initiating the calls. When the customer made the key choices and the platform had no notice of illegal use, the platform has prevailed, as in Kauffman.
The clearest platform win is Kauffman v. CallFire, Inc. (S.D. Cal. Oct. 7, 2015), where the customers wrote the texts and supplied the numbers. The court granted the platform summary judgment and wrote that “Liability under the TCPA for an internet-based transmitter of text messages turns on the transmitter’s degree of involvement in selecting the content, timing, and recipients of a message as well as whether the transmitter had notice of the illegal nature of a message.” It also rejected the idea that lawsuits gave the platform notice: “A complaint is an allegation of an illegal act, not notice of an illegal act.”
Nunes v. Twitter, Inc. (N.D. Cal. July 1, 2016) came out the other way on different facts. Twitter texted tweets to phone numbers that had been reassigned to new owners. On cross-motions for partial summary judgment, the court held that Twitter, not the tweet’s author or the number’s former owner, made the calls, describing Twitter as “the actual sender of the text.” That ruling didn’t decide the claim’s other elements, such as whether Twitter used an autodialer.
In Cunningham v. Montes (W.D. Wis. May 2, 2019), the court denied summary judgment to the operator of an autodialing service, writing that “a provider of auto-dialing services cannot blithely sit back and blame his customers for any TCPA violations that result from their use of his service.” Setup help and training weren’t enough on their own. Evidence that the operator “actually set up and ran some of his clients’ campaigns from start to finish,” together with evidence that he knew clients were making unlawful calls, was enough to send the question to a jury, according to the court’s order. The court also noted his belief that a terms-of-use clause telling customers to follow the law “absolved him of any legal liability,” and treated it as evidence he knew the system could be misused.
In League of Women Voters of New Hampshire v. Kramer (D.N.H. Mar. 26, 2025), over AI-voice robocalls sent before the 2024 New Hampshire presidential primary, the court refused to dismiss claims against two companies in the call path. It assumed neither one initiated the calls and held the complaint plausibly alleged they might still be liable “depending upon their knowledge of, and involvement in, the scheme.” That was a pleading-stage ruling, not a finding of liability.
Most recently, in Connor v. ServiceQuik Inc. (D. Colo. Oct. 8, 2025), a putative Do Not Call class action, the court denied a texting platform’s motion to dismiss. The complaint alleged that the platform itself sent the texts and that its service went “[f]ar beyond merely providing the platform that enables them to send messages,” including “actually setting up and providing intimate support for their customers’ campaigns and strategies.” The court held those allegations pleaded a “direct connection” between the platform and the texts. The allegations haven’t been proven, no court has found the platform or its customer liable, and as of this post the case is pending.
How does the platform’s role shape who gets sued?
In my view, the business is the natural first defendant, and a platform gets added when the facts let a plaintiff argue it made the calls. A business that writes its own messages, uploads its own list, and schedules its own sends looks like the TextMe user, whom the FCC treated as “so involved in the making of the call as to be deemed the initiator of the call,” according to FCC 15-72. That’s direct liability, and it needs no agency theory. I’d read Kauffman from the customer’s side, too: the platform won because its customer, not the platform, made the sending choices, and because the platform had no notice of illegal use. A plaintiff would point to the customer’s choices in a suit against the customer.
Vicarious liability, where a seller answers for a vendor’s calls under agency law, is a separate question. Dish concluded that a seller that doesn’t initiate a call may still be vicariously liable for a telemarketer’s violations under federal common law agency principles, per FCC 13-54. In my view, agency theories fit most platforms poorly, because the platform usually isn’t the seller. Platform cases instead tend to turn on direct liability: whether the platform crossed from carrying messages to making them.
Do Not Call claims tilt further toward the business. A private suit under 47 U.S.C. § 227(c)(5) belongs to someone who received more than one violating call in 12 months “by or on behalf of the same entity.” Under the Dish ruling, that “on behalf of” liability runs through federal common law agency principles, per FCC 13-54, so a business is exposed for texts it sends itself and, through agency, for texts others send for it. Damages are actual monetary loss or up to $500 per violation, whichever is greater, and a court may increase them to up to $1,500 if it finds the violation willful or knowing. The same provision gives a defense to a company that “has established and implemented, with due care, reasonable practices and procedures to effectively prevent telephone solicitations in violation of the regulations,” under § 227(c)(5). Good platform records help prove that defense too.
My view: a platform that markets done-for-you campaigns, writes copy, picks send times, or supplies contact lists is volunteering for the defense table. And a business shouldn’t assume a hands-off platform spreads its risk. It usually confirms the messages were the business’s own. Indemnity clauses decide who reimburses whom later. They don’t decide whom the plaintiff can sue.
Voice service providers, the carriers that put calls onto the phone network, carry one more duty. Under 47 C.F.R. § 64.1200(n)(4), a voice service provider must “Take affirmative, effective measures to prevent new and renewing customers from using its network to originate illegal calls, including knowing its customers and exercising due diligence in ensuring that its services are not used to originate illegal traffic.”
Does McLaughlin change the platform analysis?
It changes how much weight the FCC’s factors carry. In June 2025, the Supreme Court held in McLaughlin Chiropractic Associates, Inc. v. McKesson Corp. that a district court should read the TCPA under “ordinary principles of statutory interpretation, affording appropriate respect to the agency’s interpretation.” Before that, some courts treated FCC orders as binding. The Cunningham court called the 2015 order “binding precedent here,” according to its order.
I expect the FCC’s control factors to stay at the center, because they track what “make” and “initiate” ordinarily mean, and Nunes leaned on the statute’s plain language as well as the order. The “knowingly allowed” factor is different. It’s an FCC gloss, and after McLaughlin I think a platform has a fair argument that knowing about a customer’s misuse doesn’t turn a carrier of messages into the party that makes them. It’s an open argument, so I wouldn’t build a compliance program around it.
How should a business document who controlled each text?
I’d build the record before anyone sues.
For content, keep each template and campaign message with the name of the employee who wrote or approved it, and note any wording the platform suggested or generated. For timing, keep the scheduling settings and any automated send-time or trigger rules, with who turned them on. For recipients, keep each list upload, note where it came from and when, and keep the consent records behind it.
Then put it in the contract. I’d require the platform to preserve per-message logs and export them on short notice: the user who created the campaign, the upload, the send time, delivery status, and replies. In a class action, those logs can be the only reliable count of who got what. And if your own app sends invite-a-friend texts, you’re the platform in the FCC’s analysis. The FCC found Glide was the maker in the scenario where Glide picked the recipients and the words, but found TextMe was not, because its users chose whom to invite and when.
What to keep, and what to do, once a texting claim arrives
- Put a litigation hold on the platform data: campaign settings, message logs, delivery reports, and the consent records for every number in the claim.
- Export the opt-out and internal do-not-call records from both your systems and the platform’s, with the date each request came in and was processed.
- Pull the National Do Not Call Registry scrub records, including which version of the Registry each list was checked against.
- Gather the platform agreement, vendor and lead-source contracts, message templates, and any recorded or AI-voice files.
- Talk to a lawyer before you answer a demand letter, contact the person who sued, tender the claim to the platform, or offer anything to settle.
The firm’s California TCPA defense lawyers page explains how it approaches claims involving texts sent through third-party messaging platforms.
Questions businesses ask about platform liability
Can our texting platform be sued along with us?
Yes, as Cunningham and Connor show, and the platform’s role decides whether the claim survives. A platform that only transmits what its customer wrote, to numbers the customer uploaded, on the customer’s schedule, has strong arguments that it didn’t make the texts. One that sets up or runs campaigns for its customers is in a weaker spot.
Does the platform’s terms of service protect us?
Not against the plaintiff. Terms of service and indemnity clauses divide costs between you and the platform, but they don’t bind the person who got the texts. In Cunningham, the court treated an operator’s belief that a clause telling customers to follow the law “absolved him of any legal liability” as evidence that he knew the system could be misused, according to the order. I’d also check what your agreement says about record access and cooperation. Our post on TCPA terms in vendor contracts covers the clauses that help when the bill arrives.
If the platform chose when to send, is it the maker instead of us?
Not necessarily. Timing is one factor among several. If you wrote the message and chose the recipients, a court could still treat you as the maker, or treat both of you as makers. I wouldn’t count on an automated send-time feature to shift liability unless the records show the platform, not the business, made that call.
If your company is sued along with its texting platform
If your company and its messaging platform are both named in a TCPA suit, the campaign logs and the platform contract usually decide who answers for each text. Rosenberger + Kawabata represents businesses defending TCPA lawsuits and class actions through its business litigation practice, including claims involving texts sent through third-party messaging platforms. Contact Rosenberger + Kawabata online for a free and confidential consultation, or call (310) 894-6921.
Sources
The statute is the Telephone Consumer Protection Act at 47 U.S.C. § 227. The FCC’s rules are at 47 C.F.R. § 64.1200, as published in the eCFR through November 15, 2025. The FCC orders are the Dish Network declaratory ruling, FCC 13-54, 28 FCC Rcd 6574 (2013), and the 2015 omnibus declaratory ruling and order, FCC 15-72, 30 FCC Rcd 7961 (2015), paragraphs 25 to 37.
The court decisions are Facebook, Inc. v. Duguid, 592 U.S. 395 (2021); McLaughlin Chiropractic Associates, Inc. v. McKesson Corp., No. 23-1226 (U.S. June 20, 2025); Kauffman v. CallFire, Inc., 141 F. Supp. 3d 1044 (S.D. Cal. 2015); Nunes v. Twitter, Inc., 194 F. Supp. 3d 959 (N.D. Cal. 2016); Cunningham v. Montes, 378 F. Supp. 3d 741 (W.D. Wis. 2019); League of Women Voters of New Hampshire v. Kramer, No. 24-cv-73-SM-TSM (D.N.H. Mar. 26, 2025); and Connor v. ServiceQuik Inc., No. 1:24-cv-02286-CNS-NRN (D. Colo. Oct. 8, 2025).
This post is general information about the Telephone Consumer Protection Act and related laws as of November 15, 2025. It is not legal advice, and reading it does not create an attorney-client relationship. Courts and the FCC change the rules in this area often, and the outcome of any claim depends on its facts. Past results do not guarantee future outcomes.