Sonos v. Google: When a Long-Pending Utility Patent Becomes Unenforceable

Judge Alsup wiped out a $32.5 million verdict by holding that Sonos's zone-scene patents were equitably unenforceable for prosecution laches. Then the Federal Circuit reversed on prejudice. A roadmap to the limits of the continuation game.

By Lidiia Levitska  ·  March 22, 2026
Networked wireless speakers in a modern living room
The dispute concerned patents claiming the grouping of networked speakers into saved 'zone scenes.' Shutterstock
Educational content, not legal advice. This article explains general legal concepts. It does not create an attorney–client relationship. For your specific situation, consult a licensed attorney.

Sonos, Inc. v. Google LLC, No. 3:20-cv-06754-WHA, 2023 WL 6542320 (N.D. Cal. Oct. 6, 2023), is the rare modern decision in which a district court took a jury verdict away on the ground that the asserted utility patents were equitably unenforceable for prosecution laches. After a jury awarded Sonos just over $32.5 million, having found that Google’s redesigned product infringed claim 1 of one of the two asserted “zone scene” patents (it found the other not infringed), Judge William Alsup granted Google’s post-trial motion and held both patents unenforceable, finding that Sonos had waited roughly thirteen years (until a competitor’s product was on the market) to inject claims reading on that product into a long-running chain of continuations. On August 28, 2025, in Google LLC v. Sonos, Inc., No. 24-1097 (Fed. Cir. Aug. 28, 2025) (nonprecedential), a panel of Judges Lourie and Prost and Chief District Judge Bumb (sitting by designation), in an opinion by Judge Lourie, reversed the laches ruling on the prejudice element and also reversed the district court’s separate invalidity holding, reviving the verdict. The two decisions together form the most instructive recent treatment of whether and how a patent owner can lose its rights by keeping an application family alive too long.

At a glance

  • District court: Sonos, Inc. v. Google LLC, No. 3:20-cv-06754-WHA, 2023 WL 6542320 (N.D. Cal. Oct. 6, 2023) (Alsup, J.)
  • Patents: U.S. Patent Nos. 10,469,966 and 10,848,885, the “zone scene” patents, claiming priority through a continuation chain to a 2007 nonprovisional application
  • Trial result: Jury verdict of just over $32.5 million for Sonos on claim 1 of the ‘885 patent (the jury found the ‘966 patent not infringed); Judge Alsup then set aside the verdict, holding both patents unenforceable for prosecution laches and the asserted claims invalid as anticipated, on the theory that a 2019 amendment to the specification was new matter that cost the claims their 2006/2007 priority date
  • Appeal: Google LLC v. Sonos, Inc., No. 24-1097 (Fed. Cir. Aug. 28, 2025) (nonprecedential; Lourie, J.), reversing both the prosecution-laches holding (on the prejudice prong) and the invalidity holding as to the zone scene patents, and affirming the separate invalidity of U.S. Patent No. 10,779,033 (the “Direct Control” patent) as obvious
  • Doctrine: Prosecution laches (an equitable defense rendering a patent unenforceable where the patentee’s delay in prosecution is unreasonable, inexcusable, and prejudicial)

The continuation strategy and the district court’s response

The substance of the technology is almost beside the point. Sonos’s asserted patents claimed the ability to create and save overlapping groups of networked speakers (“zone scenes”) so that the same speaker could belong to more than one saved group. What mattered to Judge Alsup was the timeline. The asserted claims issued in 2019 and 2020 but claimed priority back through a chain of continuation applications to a nonprovisional filed in 2007 (itself claiming benefit of a 2006 provisional). In the court’s telling, the specific claim language covering overlapping zone scenes did not appear until 2019, after Google had developed and shipped the accused products.

That sequence is what triggered the equitable defense. Prosecution laches, as the Supreme Court recognized in Woodbridge v. United States, 263 U.S. 50 (1923), and Webster Electric Co. v. Splitdorf Electrical Co., 264 U.S. 463 (1924), and as the Federal Circuit elaborated in Symbol Technologies, Inc. v. Lemelson Medical, Education & Research Foundation, 277 F.3d 1361 (Fed. Cir. 2002), and Hyatt v. Hirshfeld, 998 F.3d 1347 (Fed. Cir. 2021), permits a court to refuse enforcement where a patentee’s “unreasonable and unexplained” delay in prosecution causes prejudice. Judge Alsup found both elements. The delay, he reasoned, was unreasonable and inexcusable: Sonos had let the industry develop the feature and only then drafted claims to capture it, in what the court characterized as a “case of the industry leading with something new and, only then, an inventor coming out of the woodwork to say that he had come up with the idea first.” The prejudice, the court held, flowed from Google’s investment in the accused technology during the years of delay.

The opinion was notable precisely because it applied prosecution laches to a post-1995 application family. The 1995 GATT amendments changed patent term from seventeen years post-issuance to twenty years from the earliest priority date, largely eliminating the “submarine patent” abuse (prosecutors deliberately dragging out applications to delay the term clock) that animated the classic laches cases. Many practitioners had assumed that the new term regime made prosecution laches largely a historical artifact. Judge Alsup disagreed, holding that the doctrine survives the GATT transition and can still police continuation practice even where the patent term itself is not artificially extended.

The enforceability of long-pending utility patents

The Sonos litigation crystallizes a tension at the heart of continuation practice. Continuations are a legitimate, statutorily sanctioned tool. A patentee is entitled under 35 U.S.C. § 120 to keep a family pending and to draft new claims, within the bounds of the original written description, to read on later-emerging products. The Federal Circuit has repeatedly said that adding claims to cover a competitor’s product is not, by itself, inequitable: Kingsdown Medical Consultants, Ltd. v. Hollister Inc., 863 F.2d 867 (Fed. Cir. 1988) (en banc), and Liebel-Flarsheim Co. v. Medrad, Inc., 358 F.3d 898, 909 (Fed. Cir. 2004), which the Sonos panel invoked for the proposition that “it is not improper for an applicant to broaden his claims during prosecution in order to encompass a competitor’s products, as long as the disclosure supports the broadened claims.” The question is when ordinary continuation strategy curdles into the kind of “egregious misuse of the statutory patent system” that Symbol Technologies and Hyatt require for laches.

Judge Alsup’s answer leaned heavily on the relationship between the delay and the development of the accused product. But that framing is also where the district court was most exposed on appeal. There is a doctrinal difference between two propositions: (1) the patentee delayed, and during the delay the accused infringer built a business it would not have built had the claims issued earlier; and (2) the patentee delayed, and the accused infringer relied on the absence of the claims. The first is a true intervening-rights theory of prejudice; the second is closer to a windfall complaint. To establish prejudice, the defendant must show that it (or others) invested in, designed around, or commercialized the technology in reliance on the patent’s absence during the period of unreasonable delay: a causal link, not merely a coincidence of timing.

That distinction is what undid the district court on appeal. The Federal Circuit, in its August 2025 nonprecedential decision, declined to resolve whether a thirteen-year delay was unreasonable and inexcusable and instead limited its analysis to prejudice, finding it absent for two independent reasons. First, the panel found that Google’s only prejudice argument below was that it “began investing in its products by at least 2015,” and that Google had offered no evidence, testimony or otherwise, to support that assertion or to show that it was caught unawares that Sonos may already have invented the functionality. Without evidence, Google could not carry its burden. Second, and more pointedly, the court reasoned that even accepting a 2014 or 2015 investment date, Sonos’s 2007 nonprovisional application became public when it issued as U.S. Patent No. 8,483,853 on July 9, 2013, before any of Google’s asserted investments, and that “Google cannot be prejudiced by incorporating into its products a feature that was publicly disclosed in a patent application prior to its investment.” Once the invention was public, the theory that Google relied on its absence collapsed. The panel reversed both the unenforceability ruling and the written-description invalidity ruling on the zone scene patents, while affirming the separate invalidity of the Direct Control patent as obvious.

Open questions

The decisions leave the doctrine in an unsettled posture, deliberately so. Several questions remain live:

  • Does prosecution laches retain real force after GATT? Judge Alsup said yes; the Federal Circuit neither embraced nor rejected that view, resolving the appeal on prejudice and declining to opine on the delay prong. The viability of laches against post-1995 continuation families remains formally open at the appellate level.
  • What quantum of delay is “unreasonable”? Thirteen years was enough for the district court but was never tested on appeal. There is no bright line, and the totality-of-circumstances standard offers little predictability.
  • Can publication of the parent application ever defeat prejudice as a categorical matter? The panel’s reliance on the 2013 publication suggests a near-categorical defense for accused infringers whenever the priority application disclosed the invention before their development: an inference that could substantially blunt the doctrine if read broadly.
  • Because the opinion is nonprecedential, how much guidance does it actually provide? District courts are not bound by it, and the absence of a precedential holding on the delay prong means the next case could come out differently on similar facts.

Implications

  • For patent owners running continuation families: Document the technical and strategic reasons for each claim amendment, and be prepared to explain delay in terms of ordinary prosecution dynamics rather than waiting for a competitor’s product to crystallize. Publication of the application early in the chain is now a meaningful shield against later prosecution-laches attacks.
  • For accused infringers: Prosecution laches is alive, but the prejudice element is demanding. A timeline showing that the patentee added claims after your product launched is not enough; you must tie your investment to reliance on the patent’s absence, and that theory is vulnerable if the invention was publicly disclosed before you built.
  • For litigators: Treat prosecution laches as a fact-intensive equitable defense to be developed through prosecution-history discovery and investment timelines, not as a legal argument that can be won on the face of the file wrapper.
  • For prosecutors: The decision is a reminder that the written-description and new-matter constraints of § 112 sit alongside laches as an independent line of attack; the district court also invalidated the zone scene claims on a new-matter theory. But that ground failed too. The Federal Circuit reversed it, holding that the 2007 specification’s description of a “Morning Scene” and an overlapping “Evening Scene” already conveyed possession of overlapping zone scenes, so the 2019 amendment did not need to be reached. Both of Google’s post-trial grounds were undone by what the original disclosure actually said.

Frequently asked questions

What is prosecution laches? Prosecution laches is a judge-made equitable defense that renders a patent unenforceable when the patentee’s delay in prosecuting the application was unreasonable and inexcusable under the totality of the circumstances and caused prejudice to the accused infringer or the public. It traces to Supreme Court decisions from the 1920s and was revived for the continuation era by the Federal Circuit in Symbol Technologies (2002) and Hyatt (2021).

Did Sonos ultimately lose its patents? No. The Federal Circuit reversed Judge Alsup’s prosecution-laches ruling on the zone scene patents, finding no evidence of prejudice, and separately reversed his holding that those claims were invalid for lack of written description, reviving the $32.5 million verdict. On the laches question the panel did not decide whether Sonos’s delay was unreasonable; it resolved that issue on prejudice alone. Sonos did not win everything: the panel affirmed the invalidity of a separate asserted patent, the “Direct Control” patent, as obvious.

Does adding claims to cover a competitor’s product make a patent unenforceable? Not by itself. The Federal Circuit has long held that drafting continuation claims to read on a competitor’s product is a legitimate part of prosecution. Unenforceability requires the additional showings of unreasonable, inexcusable delay and prejudice (an egregious misuse of the patent system), not merely the act of claiming a competitor’s product within the scope of the original disclosure.

Authorities and sources

Related guides

Lidiia Levitska
About the Author

Lidiia Levitska

International Intellectual Property Attorney

Lidiia Levitska focuses on intellectual property dispute resolution, policy, and advisory work across international institutions and government bodies. From 2021 to 2025 she served at the World Intellectual Property Organization (WIPO), managing arbitration cases and overseeing compliance with the Uniform Domain-Name Dispute-Resolution Policy (UDRP), and earlier led IP policy research as a Senior Policy Officer at the American Chamber of Commerce in Ukraine. She holds an LL.M. in International Intellectual Property Law from Chicago-Kent College of Law and an M.A. in Information Technology Law from the University of Tartu, and was admitted to the Ukrainian Bar in 2019.

More about Lidiia →