The campaign at a glance

Respondent	Pandora Media, LLC — streaming radio, a SiriusXM subsidiary
Legal theory	California Unruh Civil Rights Act (Civ. Code §§ 51–52): Pandora allegedly used claimants' age, sex, and gender data to target advertising — a theory descended from Liapes v. Facebook
Lead claimant counsel	Keller Postman (Warren Postman) — the firm that built the mass-arbitration playbook
Scale	~41,000 pre-arbitration notices; 23,821 demands filed at AAA plus 2,562 attempted; 26,383 total in AAA disclosure data (01-25-0003-3428) 12
Duration	~18 months of pre-arbitration and court fighting; 208 days inside AAA
Outcome	A single AAA Process Arbitrator dismissed every pending claim in one order on January 28, 2026; AAA closed the matter February 10, 2026 12

The story

The claimants spent a year fighting to get into arbitration. Pandora's own Process Arbitrator then needed one order to end all of it.

The theory came from a live edge of California law: Liapes v. Facebook (2023) held that a platform's demographic ad-targeting — showing insurance ads only to younger, male users — could state a claim under the Unruh Act, which carries $4,000 minimum statutory damages per violation ⁶⁷. A wave of Unruh ad-targeting mass filings followed; Consovoy McCarthy filed 19,541 demands on the same theory against Sega in April 2024 ⁸. Keller Postman aimed it at Pandora: the free tier is ad-supported, the ads are targeted, and the targeting allegedly used protected characteristics.

What happened next inverted every assumption about forced arbitration. Pandora — whose terms of service compel individual arbitration — did not rush to its chosen forum. The claimants had to petition the Los Angeles Superior Court for an order compelling Pandora to arbitrate ². They mostly won. And then the forum they fought to enter dismissed them wholesale — not on the merits of any individual claim, but through the one role AAA created specifically for mass filings.

Timeline

Date	Event
2024–2025	Mass pre-arbitration notices served on behalf of ~41,000 claimants alleging Unruh Act violations in Pandora's ad targeting 2
2025-01-17	Petition to compel arbitration filed in LA Superior Court on behalf of ~26,000 petitioners; amended February 14 2
5 months
2025-06-18	Court grants the petition in part: arbitration compelled for most petitioners, denied as to 2,216 2
2025-07-04	23,821 claimants file demands with AAA; the AAA opens a mass arbitration 2
2025-07-15	The filing lands in AAA's consumer docket as case 01-25-0003-3428 — 26,383 demands in the disclosure data, lead attorney Warren Postman 1
5 months
2025-12-15	Claimants' counsel moves to add 2,562 claimants with identical claims 2
2026-01-28	A Process Arbitrator issues an order dismissing all of the pending claims 2
2026-02-10	AAA closes the matter 12

How the campaign died

Meet the Process Arbitrator. When mass arbitration broke AAA's one-case-at-a-time machinery, AAA built a gatekeeper: a single arbitrator appointed to decide campaign-level questions before any merits arbitrator hears any individual claim. Under the Mass Arbitration Supplementary Rules — substantially expanded effective January 2024 — the Process Arbitrator rules on whether filing requirements were met, conditions precedent, fee disputes, and which demands get to be part of the mass filing at all ³⁴⁵. The role exists because coordinated filings exist. It is administrative plumbing with case-dispositive power.

One decision-maker, campaign-level stakes. In court, a defect across 26,000 claims gets litigated through motions, appeals, and certification fights measured in years. At AAA, the same question goes to one arbitrator whose threshold ruling binds the entire filing. That concentration cuts both ways — it is how respondents get fee schedules enforced against them at eight figures, and it is how 26,383 claims can end in a single order. The mechanics that make mass arbitration fast are the same mechanics that make it mortal.

The asymmetry that defines the category. Same corporate family, same forum, same quarter: Hattis's fee campaign against SiriusXM closed in a global settlement covering ~30,700 claimants; Postman's Unruh campaign against Pandora closed with everyone dismissed. Across AAA generally, only 1–2% of mass-arbitration cases end in awards — the action is in settlements (59–77% of closures) and, increasingly, in threshold process rulings like this one ¹⁰. The Process Arbitrator is now among the highest-leverage decision points in aggregate litigation, and almost nobody outside the caption ever reads their orders.

What it cost to lose

The claimant side's sunk inputs are the barrier-to-entry story in reverse: acquiring and screening ~41,000 claimants, a year of state-court motion practice just to reach the forum, AAA initiation fees on a five-figure filing, and eighteen months of a national firm's time — recovered only if the campaign survived the gate. It didn't. Whether any path remains — refiling the 2,216 court-track claimants, curing and refiling at AAA, or challenging the order — is not publicly known ².

The operator's takeaway

None of this is an argument against the category — it is the category's actual rulebook, written one order at a time. Three lessons travel:

1. The gate has rules, and they are knowable. Process Arbitrators decide filing requirements, conditions precedent, and inclusion — every one a question a campaign can answer before demand one is drafted. Wallrich died for lack of evidence of arbitration agreements ¹¹; that is an intake discipline, not an act of God. The firms that survive threshold review are the ones whose files could have survived it on the day each client signed.
2. The same concentration pays when the paperwork holds. Identical one-arbitrator mechanics produced DoorDash's $9.5 million fee order against the respondent ¹² and the SiriusXM global settlement one column over. The gate is not stacked against claimants; it is stacked against defective filings — in either direction.
3. Price the gate. A dismissal probability belongs in campaign underwriting the way loss ratios belong in insurance — a number you carry, not a surprise you absorb. (Our simulator's privacy preset ships with a 35% dismissal default; this case is why.)

Sidebar · The inverse case: when the gate cuts the other way

The same firm knows the other side of this door. In Wallrich v. Samsung, Keller Postman filed ~50,000 AAA demands and AAA billed Samsung $4.125 million in initial administrative fees; a district court ordered Samsung to arbitrate and pay. In 2024 the Seventh Circuit reversed — claimants' counsel had not produced evidence of an arbitration agreement for the people they claimed to represent ¹¹. Set that against Abernathy v. DoorDash (2020), where a federal judge ordered DoorDash to pay $9.5 million in fees to arbitrate with 5,010 couriers, telling the company it was hoisted on its own arbitration clause ¹².

Fees and filing integrity are the twin levers of every process fight. When the paperwork holds, the respondent pays millions to face the claims. When it does not — Samsung, and perhaps Pandora — the campaign dies at the threshold. The gate is the game.

Open questions

Reporting on this campaign runs into a wall the parties built: the order is confidential, and no one has moved to make it public. Here is what we still don't know — and would ask. Our line is open to Keller Postman and to Pandora alike.

1. What did the January 28 order actually hold? The Process Arbitrator's powers run to threshold questions — filing requirements, preconditions, who belongs in the filing — and AAA's own 2025 data shows merits dismissals in consumer cases are vanishingly rare. Which threshold ground ended 26,383 claims?
2. Why did nobody seek review? Vacatur — the narrow court challenge to an arbitral ruling — runs on a clock of roughly three months. We can find no petition filed anywhere. A firm that believed 26,000 claims were wrongly dismissed had a path; what does declining it mean?
3. Did the December motion to add 2,562 claimants trigger the review that ended the campaign six weeks later?
4. What happens to the claimants — the dismissed thousands, and the 2,216 whose claims the court kept out of arbitration back in June 2025?
5. How should firms price this risk? Process-arbitrator dismissal is now a campaign-level exposure every operator carries. What does a well-built filing file look like on the day it's tested?