Does the TCPA Apply If a Person Dials Each Call by Hand?

Plenty of businesses assume that having an employee click to send each call or text keeps them clear of the Telephone Consumer Protection Act, and others assume the opposite, that any automated platform is illegal. Both miss the point. After the Supreme Court’s 2021 decision in Facebook v. Duguid, the autodialer ban depends on whether the equipment generates phone numbers, not on how much a person touches the keyboard. A company sued over how it places calls usually has more defenses than the demand letter lets on.

Does it matter whether a person or the software places the call?

Less than most people think. The statute and the Supreme Court both key the autodialer rules to what the equipment can do, not to who pressed the button.

The TCPA’s autodialer provision makes it unlawful to call a cell phone without consent “using any automatic telephone dialing system or an artificial or prerecorded voice,” under 47 U.S.C. § 227(b)(1)(A). An automatic telephone dialing system, or autodialer, is defined as equipment with the capacity “to store or produce telephone numbers to be called, using a random or sequential number generator” and to dial them. The “random or sequential number generator” is the whole ballgame.

In Facebook, Inc. v. Duguid, the Court held that “a necessary feature of an autodialer under §227(a)(1)(A) is the capacity to use a random or sequential number generator to either store or produce phone numbers to be called.” The plaintiff had argued that the dividing line should be human intervention, that a system which dials on its own is an autodialer and one a person operates by hand is not. The Court rejected that framing, noting in the same opinion that “all devices require some human intervention,” from programming a phone to answer texts automatically to commanding a computer to generate and dial numbers at random. So dialing by hand isn’t a magic phrase, and neither is running an automated platform. The question a court asks is narrower: can this equipment generate the numbers it calls?

That cuts both ways, and the defense side is usually the better side of it. Most business calling and texting runs off a known list, customers, patients, leads who filled out a form, numbers a salesperson typed in. Equipment that only dials numbers a human or a customer supplied generally can’t generate numbers at all, so it falls outside the definition no matter how automated the sending feels.

If our platform could autodial but we called a set list, are we liable?

Not for that call, in the courts that have looked closely at the statute’s wording. Having capable equipment in the building is not the same as using it to break the law.

Section 227(b)(1)(A) doesn’t ban owning an autodialer. It bans making a call that uses one. The Third Circuit took that word seriously in Panzarella v. Navient Solutions, Inc., holding that a call breaks the rule only when it actually employs the equipment’s capacity to generate numbers randomly or sequentially. The calls there went to a curated borrower list, so even assuming the system could generate numbers, the claim failed. The court read the statute to reach calls that actually use an autodialer’s number-generating function, not to punish a business for the features its software happens to ship with.

Panzarella binds only the Third Circuit, and other courts frame the capacity-versus-use question differently. But the point survives the differences. A company that dials a specific, known list, whether a person clicks each number or a system fires them in order, is a long way from the random-number robodialing Congress wrote the statute to stop. The practical lesson for a defendant is to prove two things, not one: what the platform can do, and what it actually did on the calls in the complaint. A plaintiff who can’t show the equipment generated the numbers, or that the challenged calls used that function, has a thin autodialer case.

What manual dialing does not fix

Businesses get this part wrong all the time. The dial-by-hand defense answers the autodialer claim. That’s all it answers. Two of the most common TCPA theories don’t depend on dialing equipment at all, and a company that leans on manual dialing while ignoring them is defending the wrong case.

The first is the prerecorded and artificial-voice ban. A separate part of the same subsection bars a business “to initiate any telephone call to any residential telephone line using an artificial or prerecorded voice to deliver a message without the prior express consent of the called party,” under 47 U.S.C. § 227(b)(1)(B), and the parallel cell-phone rule reaches artificial and prerecorded voices too. The Court in Duguid pointed out that Congress prohibited those voice calls “irrespective of the type of technology used.” A live person can press dial on every single call, and if what plays when the phone picks up is a recording or a synthesized voice, the autodialer defense does nothing. The FCC has already said the TCPA’s voice restrictions reach today’s AI-generated voices, so businesses running AI voice agents for outbound calls carry this exposure whether or not a human initiated the call.

The second is Do Not Call. The private claim under 47 U.S.C. § 227(c)(5) belongs to someone who got more than one telephone solicitation within a year in violation of the FCC’s do-not-call rules, and those rules never cared how the numbers were dialed. They turn on whether a number is on the Registry, whether the recipient asked the company to stop, and the hours the calls went out. Dialing by hand changes none of that.

The honest read for a business reader: the manner of dialing is a real defense, but it’s the one companies most often overstate, and treating it as a cure-all is a mistake. The claims that survive a manual-dialing argument are often the harder ones to beat.

Where manual dialing helps, and where it doesn’t

Claim theory Where it comes from Exposure per call or text Does dialing by hand help?
Autodialed call or text to a cell phone 47 U.S.C. § 227(a)(1), (b)(1)(A)(iii), (b)(3) $500, or up to $1,500 if willful or knowing Yes. The claim needs equipment that generates numbers and a call that used that capacity
Artificial or prerecorded voice, including AI voice 47 U.S.C. § 227(b)(1)(A) and (B) $500, or up to $1,500 if willful or knowing No. It applies no matter the technology or who placed the call, and consent is the defense
Do Not Call Registry or stop-request calls and texts 47 U.S.C. § 227(c)(5) Up to $500, or up to $1,500 if willful or knowing No. It comes down to the Registry and opt-outs, not the dialer

Each $500 figure is a floor unless the recipient’s actual loss is greater, and for a willful or knowing violation a court may, in its discretion, raise the award to as much as three times that amount. No one is entitled to any particular amount, because these figures are statutory caps on what a court may award.

What to document about how your calls go out

When a demand letter or complaint lands, the manner-of-dialing defense depends on records you can mostly gather before anyone sues. Put a litigation hold on everything below the moment a claim arrives.

  1. Get your dialing or texting vendor’s technical description in writing: how numbers enter the system, and whether the platform can generate telephone numbers at all. A short vendor declaration covering that ground is far easier to obtain while the relationship is friendly, and it can anchor a motion later.
  2. Preserve the platform logs, campaign settings, and the consent record for every number named in the claim, under the hold.
  3. Pull the opt-out and internal do-not-call history for those numbers, with the date each request came in and when the number was suppressed.
  4. Pull the National Do Not Call Registry scrub records for the campaigns involved and the version dates of the Registry data used, since manual dialing won’t answer a Registry claim.
  5. Collect any recorded or AI-generated audio that played on the calls, because a prerecorded-voice theory ignores the dialer and looks only at what the recipient heard.

The firm’s work defending TCPA class actions in California often starts with exactly these records, because how the numbers entered the system usually decides the autodialer count. For the deeper question of what counts as an autodialer at all, this post’s companion on the autodialer definition after Duguid walks through the case law, and for text campaigns the discussion of peer-to-peer texting platforms covers how manual sending plays out in practice.

Questions businesses ask about manual dialing and the TCPA

If our agents click to dial every call, are we safe from the TCPA?

Not on its own. Clicking to dial each call is strong evidence against an autodialer claim, because it shows the calls went to chosen numbers instead of randomly or sequentially generated ones. But it does nothing for a prerecorded-voice or AI-voice claim, which applies no matter who places the call, and nothing for a Do Not Call claim, which looks at the Registry and opt-outs. Click-to-dial narrows your exposure. It doesn’t erase it.

Is our texting platform an autodialer just because it sends messages automatically?

Usually not for that reason alone. The test is whether the platform can store or produce numbers using a random or sequential number generator, not how automatically it sends. A platform that only messages numbers your customers gave you generally doesn’t meet that test. The answer depends on the platform’s actual capabilities, which is why a written description from the vendor matters.

Does owning an autodialer make every call we place illegal?

No. The statute bars making a call using an autodialer, not owning one. A business can have capable equipment and still place calls that don’t use the number-generating function at all, as the Third Circuit recognized in Panzarella. Proving what the system did on the specific calls, not just what it could do, is the defense.

Does manual dialing help against a prerecorded-voice claim?

No. The artificial and prerecorded-voice ban is a separate rule that applies regardless of the dialing technology or who started the call. If a recording or synthesized voice plays when the call connects, the defense is prior express consent, not the fact that a person dialed.

Defending a claim over how your calls are placed

If your company has been sued or received a demand letter over calls or texts sent through a dialer or texting platform, how the numbers entered that system and what played when the call connected often decide the case. Rosenberger + Kawabata represents businesses defending TCPA lawsuits and class actions through its business litigation practice, including claims involving autodialer allegations and prerecorded-voice calls. Contact Rosenberger + Kawabata online for a confidential consultation, or call (310) 894-6921.

Sources

The statute is the Telephone Consumer Protection Act at 47 U.S.C. § 227. The Supreme Court decision is Facebook, Inc. v. Duguid, 592 U.S. 395 (2021). The court of appeals decision on using an autodialer is Panzarella v. Navient Solutions, Inc., 37 F.4th 867 (3d Cir. 2022).

This post is general information about the Telephone Consumer Protection Act and related laws as of August 5, 2026. 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.

← Back to Blog

Let’s Talk.