Table of Contents >> Show >> Hide
- Why This Case Is Getting So Much Attention
- Background: What Debra Duke Alleged
- The TCPA Issue in Everyday Language
- The Discovery Fight: Why the Call Logs Mattered
- What the July 2025 Order Required
- Privacy Arguments Did Not Carry the Day
- Why This Follow-Up Is Bigger Than One Lawsuit
- What Businesses Can Learn from Duke
- Consumer Takeaways: What If You Receive Wrong-Number Robocalls?
- Analysis: The Real Message Behind the Follow-Up
- Practical Experiences and Lessons Related to the Duke Follow-Up
- Final Takeaway
- SEO Tags
Note: This article is for informational and editorial purposes only. It summarizes public allegations, court orders, and broader TCPA compliance lessons. It is not legal advice, and no court finding of TCPA liability against American Express is stated here.
Why This Case Is Getting So Much Attention
The follow-up on Debra Duke v. American Express Company matters because it sits at the intersection of three things Americans understand very well: cell phones, unwanted calls, and the special kind of irritation that arrives when a prerecorded voice wants to discuss someone else’s account. The case is pending in the U.S. District Court for the District of Arizona and centers on allegations under the Telephone Consumer Protection Act, commonly called the TCPA.
At its core, the lawsuit alleges that Debra Duke received prerecorded calls and text messages from or on behalf of American Express that were intended for another person. Duke says she was not an American Express customer in connection with those calls, did not know the intended recipient, and repeatedly told representatives they had the wrong number. According to the complaint, the calls continued anyway. That is the sort of fact pattern that can make TCPA lawyers sit up straight, spill their coffee, and start looking very carefully at call logs.
The case has become especially noteworthy because of a July 2025 discovery order. The court granted Duke’s second motion to compel and ordered American Express to produce records reflecting attempts by the company or its vendors to place prerecorded or artificial voice calls to certain wrong-party or “never call” numbers during the class period. In plain English: the court wanted the data, not a shrug wrapped in technical complexity.
Background: What Debra Duke Alleged
Debra Duke filed the class action complaint in March 2023. She alleged that American Express violated the TCPA by making prerecorded calls to consumers’ phone numbers without first obtaining prior express consent. Her own story, as pleaded, is strikingly simple: beginning in August 2022, she allegedly received calls and text messages meant for someone named Melissa Mae Treso, a person Duke said she did not know.
Duke claimed she had owned her cell phone number since at least 2012. She also alleged that before and during the calling period, she reached live employees, explained that the number was wrong, and asked for the calls to stop. The complaint describes several specific call dates and numbers, including calls from numbers alleged to be owned or operated by American Express. It also alleges that the communications caused annoyance, nuisance, wasted time, invasion of privacy, stress, and interference with the normal use of her phone.
The proposed class in the complaint was framed around people in the United States who, during the relevant period, received prerecorded voice calls on their cellular phones without the required consent. The complaint also sought certification under Federal Rule of Civil Procedure 23, which is the gateway for turning one person’s claim into a class action. That is where discovery became the engine of the case.
The TCPA Issue in Everyday Language
The TCPA restricts certain calls made with an automatic telephone dialing system or an artificial or prerecorded voice to cell phones unless the call is made for an emergency purpose or with prior express consent of the called party. The phrase “called party” is important. In wrong-number cases, a business may intend to reach a customer who once gave a phone number, but the call may actually reach a different person who never consented.
That distinction is the legal equivalent of mailing a birthday cake to the wrong address and then insisting the frosting had permission. Consent is not a magic sticker permanently attached to a number in all circumstances. A company must be able to show why it believed it had consent to call the person who actually received the call, especially when prerecorded or artificial voice technology is involved.
For businesses, the problem is not merely one awkward call. Large-scale outbound calling programs can generate thousands or millions of records. If even a small percentage of those calls reach wrong parties, the potential exposure can grow quickly because the TCPA provides statutory damages that can be significant on a per-call basis. That is why call logs, consent records, wrong-number notations, vendor data, and suppression practices become so important.
The Discovery Fight: Why the Call Logs Mattered
The most important follow-up development is not a final judgment. It is a discovery order. That may sound less exciting than a courtroom showdown, but in class action litigation, discovery can be the moment when the chessboard flips over and everyone suddenly remembers where the pieces were hiding.
Duke sought information that would help identify whether a class could be certified. Specifically, she wanted records showing calls made with prerecorded or artificial voice technology to phone numbers associated with wrong-party or “never call” notations. That information goes directly to Rule 23 issues such as numerosity and commonality. Numerosity asks whether there are enough potential class members to justify class treatment. Commonality asks whether the claims share common questions of law or fact.
American Express resisted, arguing that the requests were vague, overbroad, burdensome, disproportionate, and raised privacy concerns. The company also maintained that it could not readily determine certain call frequencies or systematically identify everyone who may have received calls because of customer error. The court, however, was not persuaded that those objections justified withholding the requested discovery.
The court emphasized that plaintiffs in putative TCPA class actions often need outbound call data to test whether class certification is appropriate. A defendant cannot effectively say, “You have not proven a class exists,” while simultaneously blocking the records needed to examine whether a class exists. That argument may look clever in a conference room, but courts tend to prefer evidence over circular logic.
What the July 2025 Order Required
The July 11, 2025 order granted Duke’s second motion to compel. The court directed American Express to produce records reflecting each attempt by American Express or a vendor on its behalf to place a prerecorded or artificial voice call to a phone number not identified as a landline, where the company’s records reflected at least one wrong-party or “never call” notation associated with that number during the class period.
The court also set deadlines for the next stage of the case, including class certification briefing and discovery completion. At the time of that order, all discovery was set to be completed by December 19, 2025, with dispositive motions due after a final ruling on class certification. Publicly indexed later docket snippets indicate that some class certification and mediation deadlines were later extended into 2026, but the publicly available July 2025 order remains the key document shaping the follow-up analysis.
One of the most important parts of the order involved the court’s assessment of American Express’s burden argument. American Express submitted information suggesting that compliance could require searching roughly 100 terabytes of data, take 16 weeks, involve five employees, and cost about $100,000. The court did not treat those numbers as automatically decisive. Instead, it compared that burden presentation with testimony suggesting the company could search call systems for wrong-number results and then identify which numbers received prerecorded calls.
That contrast mattered. In discovery disputes, courts do not simply ask whether compliance is annoying. Nearly all meaningful discovery is annoying. The question is whether the information is relevant, proportional, and reasonably accessible under the circumstances. Here, the court found the requested information relevant and concluded that American Express had not shown enough to block production.
Privacy Arguments Did Not Carry the Day
American Express also raised privacy concerns, arguing that the requested discovery could involve customer account data. The court drew a distinction between the company’s internal need to analyze account records and the actual information Duke sought to receive. Duke was not asking for financial account details, reasons for calls, or sensitive customer history. She was seeking phone numbers and call counts connected to wrong-party or “never call” notations.
The court also noted that a protective order was already in place, meaning confidential information could be handled under restrictions. That is an important practical point for TCPA litigation. Privacy matters, but privacy concerns do not automatically defeat discovery when narrower data can be produced under confidentiality protections.
For compliance teams, the lesson is straightforward: do not assume that “privacy” is a universal stop sign. Courts may ask whether the requested information is truly private, whether it can be redacted, whether a protective order exists, and whether the requesting party needs the information to test a legally relevant issue.
Why This Follow-Up Is Bigger Than One Lawsuit
The Duke follow-up has gained attention because it speaks to a broader reality in consumer calling litigation. Large organizations often rely on multiple platforms, outside vendors, account systems, consent databases, suppression tools, and call disposition codes. That architecture may be normal from an operations standpoint, but it can become a headache in litigation if no one can easily explain what happened, when it happened, and why it happened.
In a TCPA wrong-number case, the most dangerous phrase may be: “Our system cannot determine that.” Courts may hear that as a business choice rather than a legal excuse. If a company chooses to operate a high-volume calling program using prerecorded or artificial voice technology, it should expect to maintain records that can show consent, call purpose, number source, revocation, wrong-party reports, vendor activity, and suppression history.
The follow-up also matters because plaintiffs’ lawyers can use discovery rulings like this one as a roadmap. If outbound call logs are discoverable in one TCPA putative class action, similar plaintiffs may argue they are discoverable in others. That does not mean every court will order identical production, but it does mean defendants need precise, well-supported, technically credible objections.
What Businesses Can Learn from Duke
1. Wrong-Number Data Should Be Treated as High-Risk Data
Wrong-party and “never call” notes are not just customer service breadcrumbs. They are litigation-relevant signals. If a consumer tells a company, “You have the wrong number,” that statement should trigger a clear workflow. The number should be investigated, suppressed where appropriate, and documented. A wrong-number report that sits ignored in a system is like a smoke alarm with the battery removed: quiet, but not safe.
2. Vendor Calls Are Still Your Problem
The order covered calls placed by American Express or vendors on its behalf. That is a critical reminder. Outsourcing calls does not outsource risk. Companies should ensure vendor contracts require TCPA compliance, accurate call records, prompt suppression, audit rights, and preservation of data needed for litigation.
3. Burden Objections Need Evidence, Not Drama
Saying discovery is burdensome is easy. Proving it is another matter. Courts expect specific explanations: what systems must be searched, what fields exist, what queries are possible, how long the work would take, what alternatives exist, and why a narrower production would or would not solve the problem. A dramatic number like “100 terabytes” may sound impressive, but if a corporate witness describes a more targeted search process, the burden argument can lose its sparkle quickly.
4. Class Certification Discovery Is Often Crucial
In class actions, discovery before certification is not a side quest. It is often the main road. Plaintiffs need data to show whether class members can be identified and whether common issues predominate. Defendants need the same data to challenge certification. A company that lacks organized records may find itself fighting with fog instead of facts.
Consumer Takeaways: What If You Receive Wrong-Number Robocalls?
Consumers who receive repeated wrong-number prerecorded calls should keep detailed records. Save call dates, times, phone numbers, voicemails, text messages, screenshots, and any notes from conversations with live representatives. If you tell a company it has the wrong number, write down when you said it and whom you spoke with, if known.
Consumers can also file complaints with the FCC or FTC. Regulators do not resolve every individual complaint like a private customer service desk, but complaint data helps agencies identify patterns and support enforcement priorities. In other words, your complaint may not make the robocall gremlin disappear instantly, but it can help regulators find where the gremlin lives.
It is also wise to avoid giving personal information to callers unless you are certain the call is legitimate. If a call claims to involve a debt or financial account that is not yours, contact the company through an official number from its website or account materials rather than relying on a number left in a voicemail.
Analysis: The Real Message Behind the Follow-Up
The broader message of Duke v. American Express Company is not that every TCPA plaintiff automatically wins discovery battles. The message is more practical: when call data is central to class certification, courts may require defendants to produce it, especially when the defendant controls the systems where the evidence lives.
This case also shows how deposition testimony can change the momentum of litigation. According to the court’s order, testimony from American Express’s corporate representative raised questions about whether the data production matched the process the court and plaintiff understood had been used. That discrepancy helped fuel the second motion to compel. In complex litigation, a deposition answer can be a tiny hinge that swings a very large door.
For companies, the safest path is to align legal arguments with operational reality. If the legal team says a search is impossible, the technical team should not later testify that a similar search is available. If the company produces a dataset, it should be able to explain exactly how the dataset was created, what it includes, what it excludes, and why. Courts do not expect perfection, but they do expect candor and clarity.
Practical Experiences and Lessons Related to the Duke Follow-Up
One practical experience from cases like this is that the first wrong-number complaint is often the cheapest moment to fix the problem. When a consumer says, “You are calling the wrong person,” the business has a choice. It can treat that statement as a minor inconvenience, or it can treat it as a compliance event. The second approach is much safer. A well-trained representative should know how to document the report, escalate it, suppress the number when appropriate, and confirm that vendors receive the update.
Another experience is that call-center systems often grow like a garage full of mystery cables. One team owns the dialer. Another owns customer records. A vendor owns a platform. Compliance owns policies. Legal owns litigation strategy. Everyone assumes someone else can pull the perfect report. Then a subpoena arrives, and the perfect report turns out to be five databases, three acronyms, two retired employees, and one spreadsheet named “final_final_v7.” The Duke follow-up is a warning that courts may not be sympathetic when operational messiness prevents access to central evidence.
Compliance teams should run test drills before litigation happens. Pick a sample phone number and ask: Where did this number come from? When was consent obtained? Was consent revoked? Was the number reassigned? Did anyone report it as wrong? Did a vendor call it? Was a prerecorded voice used? Can we produce the call history in a readable format? If the answer requires archaeology, the company has work to do.
Businesses should also preserve context, not just raw numbers. A call log without consent history is incomplete. A consent record without call outcomes is incomplete. A wrong-party note without suppression action is incomplete. In TCPA litigation, the story lives in the connection between records. The best compliance programs build those connections before anyone asks for them.
From the consumer side, the experience is more personal. Repeated wrong-number calls can feel absurd at first, then invasive, then exhausting. People should not have to become unpaid call-routing consultants for major companies. When a consumer takes the time to say, “That person is not here,” the system should listen. A wrong-number report is not a suggestion box comment. It is a compliance signal with legal consequences.
The Duke follow-up also teaches lawyers a practical lesson: discovery requests should be targeted, explainable, and connected to class certification elements. Plaintiffs who can explain why a dataset matters to numerosity, commonality, predominance, or ascertainability are more likely to get traction. Defendants who can propose reasonable alternatives, sampling methods, phased production, or protective measures may fare better than defendants who simply say “too hard” and hope the court brings tissues.
Finally, this case underscores a simple truth: in high-volume calling, documentation is not paperwork for paperwork’s sake. It is the safety net. When documentation is clean, companies can defend themselves with confidence. When documentation is scattered, every discovery hearing becomes a weather report: cloudy, with a chance of sanctions, extensions, and expensive data pulls.
Final Takeaway
Debra Duke v. American Express Company remains important because it highlights how a single wrong-number calling allegation can expand into a major discovery fight with class action implications. The court’s July 2025 order did not decide whether American Express violated the TCPA, but it did make clear that outbound call records tied to wrong-party and “never call” numbers can be central to evaluating class certification.
For companies, the message is blunt but useful: if you place prerecorded or artificial voice calls at scale, your data systems must be ready for scrutiny. For consumers, the case is a reminder to document unwanted calls and speak up when a caller has the wrong number. For lawyers, it is another example of how discovery can shape the entire trajectory of a TCPA class action.
In short, the Duke follow-up is less about one phone number and more about accountability in automated calling. When technology makes it easy to dial, the law may require companies to prove they dialed the right people for the right reasons. That is not just good litigation hygiene. It is basic respect for the tiny rectangle buzzing in everyone’s pocket.