A patent lawsuit is one of the longest, most expensive, and most technically demanding journeys in American civil litigation. It can run for years, consume millions of dollars in fees and expert costs, and turn on the precise meaning of a single word in a patent claim. Yet for all its complexity, a patent case follows a recognizable arc: it must be filed in the right court, pleaded adequately, the patent's claims must be construed, infringement and validity must be proved or disproved, damages must be quantified, and the loser may move to undo the verdict and ultimately appeal. This guide walks that entire arc from beginning to end, and pays special attention to the inflection point named in its title -- the denial of summary judgment, the procedural hinge on which a case swings from "decided on paper" to "tried to a jury."

To keep the abstractions grounded, we will follow a single invented dispute throughout. Helios Optics, Inc. ("Helios"), a fictional maker of fiber-optic sensors, owns U.S. Patent No. 9,XXX,XXX, which claims a temperature sensor that uses a particular arrangement of reflective coatings on an optical fiber. Helios believes that Aurora Instruments, LLC ("Aurora"), a competitor, sells a rival sensor that copies the patented design. Helios sues Aurora for patent infringement. As we move through each phase, watch how a single technical and legal question -- does Aurora's product fall within the scope of Helios's claims? -- gets refined, contested, and ultimately answered. (These parties are hypothetical; any resemblance to real companies is coincidental, and every scenario built around them is offered as an illustration, not as legal advice.)

This article is written so that a judge, a seasoned patent litigator, and a curious non-lawyer can all follow it. Every term of art is explained in plain English the first time it appears. Where the law rests on a particular case or statute, we cite it so that you can verify the proposition and dig deeper. For a foundational primer on the underlying right itself -- what a patent is and what it protects -- see our companion pieces on patent basics and utility patent basics, and for the threshold question of liability itself, what constitutes patent infringement.

What you will learn

This guide covers, in order: where a patent case can be brought (subject-matter jurisdiction and venue after TC Heartland and In re Cray); how a complaint must be pleaded now that the old fill-in-the-blank form is gone; how a court fixes the meaning of the patent claims (the Markman hearing and the Phillips methodology); how infringement is proved -- literally, under the doctrine of equivalents, and through indirect and divided-infringement theories; how an accused infringer attacks the patent's validity, including through parallel proceedings at the Patent Trial and Appeal Board; the standards for summary judgment and why a denial sets up trial; the mechanics of the jury trial, including willfulness; the law of patent damages, from reasonable royalties to lost profits to enhancement; post-trial motions under Rules 50 and 59; permanent injunctions and ongoing royalties under eBay; and the appeal to the U.S. Court of Appeals for the Federal Circuit. A FAQ and a list of related reading follow.

Part I -- Choosing the Forum: Jurisdiction and Venue

Federal courts have exclusive jurisdiction

Patents are creatures of federal law. The Constitution authorizes Congress to grant inventors exclusive rights for limited times (U.S. Const. art. I, § 8, cl. 8), and Congress has done so through the Patent Act, Title 35 of the U.S. Code. A claim "arising under" the patent laws may be heard only in federal district court, not in state court, under 28 U.S.C. § 1338(a). This is subject-matter jurisdiction -- the court's basic power to hear the type of dispute. So Helios cannot sue Aurora in state court for patent infringement; it must go to a United States District Court.

There is a wrinkle worth flagging for non-specialists. Not every dispute that mentions a patent "arises under" the patent laws. A contract dispute over a patent license, for example, may belong in state court. The Supreme Court refined this boundary in Gunn v. Minton, 568 U.S. 251 (2013), holding that a state-law legal-malpractice claim embedding a patent question did not confer federal jurisdiction because the patent issue, though disputed, was not "substantial" to the federal system as a whole. For our hypothetical, though, Helios is alleging straightforward infringement, so federal jurisdiction is secure. (For where licensing fits in the broader scheme, see how to license your patent.)

Personal jurisdiction over the defendant

Separate from subject-matter jurisdiction is personal jurisdiction -- the court's authority over the particular defendant. A court must have a constitutionally sufficient connection to the defendant, typically because the defendant has "minimum contacts" with the state where the court sits, consistent with International Shoe Co. v. Washington, 326 U.S. 310 (1945). If Aurora sells its accused sensors into a state, ships them there, or maintains operations there, personal jurisdiction usually follows. In patent cases, the Federal Circuit applies its own law to personal-jurisdiction questions because they are intimately tied to patent enforcement.

Venue: the seismic shift of TC Heartland

Venue is the question of which of the many proper courts the case may actually proceed in. For decades, patent plaintiffs enjoyed extraordinarily broad venue choices, and a remarkable share of all patent cases were filed in a handful of plaintiff-friendly districts -- most famously the Eastern District of Texas. That changed dramatically in 2017.

In TC Heartland LLC v. Kraft Foods Group Brands LLC, 581 U.S. 258 (2017), the Supreme Court held that the patent-specific venue statute, 28 U.S.C. § 1400(b), is the exclusive provision governing venue in patent infringement actions against domestic corporations, and that for venue purposes a domestic corporation "resides" only in its state of incorporation. The Court rejected the prior, far broader reading -- embodied in the Federal Circuit's VE Holding Corp. v. Johnson Gas Appliance Co., 917 F.2d 1574 (Fed. Cir. 1990) -- that had equated patent venue with the general corporate-residence rules of the general venue statute, 28 U.S.C. § 1391.

Section 1400(b) provides that a patent infringement suit may be brought either (1) where the defendant resides, or (2) where the defendant "has committed acts of infringement and has a regular and established place of business." After TC Heartland, the first prong means only the defendant's state of incorporation; where a corporation is incorporated in a multi-district state, it resides in the district of its principal place of business, or, failing that, the district of its registered office (In re BigCommerce, Inc., 890 F.3d 978 (Fed. Cir. 2018)). The second prong became hotly litigated, and the Federal Circuit gave it teeth in In re Cray Inc., 871 F.3d 1355 (Fed. Cir. 2017), holding that a "regular and established place of business" requires (1) a physical place in the district, (2) that is regular and established, and (3) that is the place of the defendant, not merely an employee's home office. Remote and work-from-home employees generally do not create venue unless the employer requires the location as a condition of employment, and even the pandemic-era surge in remote work did not loosen that rule (GreatGigz Solutions, LLC v. ZipRecruiter, Inc., No. 6:21-cv-00807 (W.D. Tex. Feb. 11, 2022)). Two further points matter in practice: the patent venue statute is "restrictive" and is "strictly construed," and the plaintiff bears the burden of establishing proper venue once challenged (In re ZTE (USA) Inc., 890 F.3d 1008 (Fed. Cir. 2018)).

For Helios and Aurora, this matters enormously. If Aurora is incorporated in Delaware and has its only facility in Massachusetts, Helios cannot simply pick a distant, plaintiff-friendly district. It can sue in Delaware (residence) or in the District of Massachusetts (if Aurora's facility there constitutes a regular and established place of business where infringing acts occurred). A misstep here can produce a motion to dismiss or transfer for improper venue, costing months. Note a strategic trap: counsel should not combine a motion to dismiss for improper venue under 28 U.S.C. § 1406(a) with a motion to transfer for convenience under § 1404(a) in the alternative without care, because the two rest on opposite premises -- § 1406 assumes venue is improper, while § 1404 assumes it is proper -- and courts have viewed a careless combined filing skeptically. (The practical fallout of TC Heartland -- the rise of Delaware and the Western District of Texas as patent hubs, and the procedural fights over transfer under § 1404(a) -- is a recurring theme in cross-border enforcement, which we discuss in our companion article on global patent litigation strategies.)

A note for foreign defendants: the venue restrictions of TC Heartland apply to domestic corporations. Foreign defendants may be sued in any judicial district under longstanding doctrine, and serving them raises its own challenges, addressed in our guides on serving legal documents to Chinese defendants in IP litigation and serving defendants in China under the Hague Service Convention.

Part II -- Pleadings: The End of Form 18

What a complaint must now allege

Once Helios has chosen a proper court, it files a complaint -- the document that starts the lawsuit and states the plaintiff's claims. For many years, patent plaintiffs could plead infringement using a bare-bones template called Form 18, an official form appended to the Federal Rules of Civil Procedure. Form 18 required little more than an allegation that the defendant made or sold a product that infringed the patent. It did not require the plaintiff to map the patent's claims onto the accused product.

Form 18 was abrogated effective December 1, 2015, when the Appendix of Forms was eliminated from the Federal Rules. Its demise was significant. With Form 18 gone, courts increasingly held that patent complaints must satisfy the ordinary "plausibility" pleading standard of Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007), and Ashcroft v. Iqbal, 556 U.S. 662 (2009). Under that standard, a complaint must contain enough factual matter to make the claim "facially plausible" -- enough to let the court draw a reasonable inference that the defendant is liable -- not merely possible.

In practice, that means Helios's complaint should do more than recite that Aurora "infringes the '___ patent." Good modern practice -- and what many district courts now expect -- is for the plaintiff to identify at least one asserted claim, identify the accused product by name or model, and explain, at least in general terms, how the accused product practices the elements of that claim. The Federal Circuit confirmed in Nalco Co. v. Chem-Mod, LLC, 883 F.3d 1337 (Fed. Cir. 2018), that a direct-infringement claim need not prove its case at the pleading stage, but it must give fair notice and plausibly allege that each claim limitation is met. The bar is not always exacting: for relatively simple technology, identifying the accused products by photographs and model numbers and alleging that they meet each claim limitation can suffice to provide fair notice (Disc Disease Solutions Inc. v. VGH Solutions, Inc., 888 F.3d 1256 (Fed. Cir. 2018)), though courts have suggested that holding may not stretch to complex inventions.

A practical point that bridges pleading and trial preparation: many active patent districts have local patent rules that require the plaintiff to serve detailed infringement contentions -- a near-claim-chart mapping each asserted claim to each accused product -- early in the case (for example, N.D. Cal. Patent L.R. 3-1 and E.D. Tex. P.R. 3-1), with the accused infringer reciprocating with invalidity contentions. These contentions discipline the parties' theories from the outset and become the backbone of the eventual expert reports.

Pleading indirect and willful infringement

If Helios also alleges indirect infringement -- that Aurora actively induced others to infringe (35 U.S.C. § 271(b)) or contributed to infringement by supplying a component (35 U.S.C. § 271(c)) -- the pleading bar is higher. The Supreme Court held in Global-Tech Appliances, Inc. v. SEB S.A., 563 U.S. 754 (2011), that induced infringement requires knowledge that the induced acts constitute patent infringement, and that "willful blindness" can satisfy that knowledge requirement. So a complaint for inducement must plausibly allege the defendant's knowledge of the patent and of the infringement (or willful blindness to it).

Helios may also want to plead willful infringement to set up a request for enhanced damages (discussed in Part VIII). To do so credibly, it typically must allege that Aurora knew of the patent -- often through a pre-suit notice letter -- and continued infringing anyway. Courts are split on whether the filing of the complaint itself can supply the knowledge needed to support post-suit willfulness, and some districts require allegations of "egregious" conduct to survive a motion to dismiss while others do not. A carefully drafted pre-suit notice letter is therefore a strategic asset; for how to write one that establishes knowledge without inviting a declaratory-judgment race, see our treatment of patent notice and demand-letter practice in writing a demand letter basics.

Aurora's answer and counterclaims

Aurora responds with an answer, admitting or denying each allegation, and almost always asserting affirmative defenses (non-infringement, invalidity, unenforceability) and counterclaims seeking declaratory judgments that the patent is not infringed and is invalid. Invalidity counterclaims are critical: even if Aurora's product reads on the claims, Aurora wins if it can show the patent should never have issued. We discuss the substance of invalidity in Part VI. (For the prosecution-side mirror image of these validity arguments -- how applicants overcome rejections before the patent ever issues -- see responding to patent office actions and overcoming obviousness rejections.)

Part III -- Claim Construction: The Markman Hearing

Why claim construction is the heart of the case

A patent ends with one or more numbered sentences called claims. The claims define the legal boundary of the invention -- the "metes and bounds" of what the patentee may exclude others from doing. Everything in a patent case ultimately turns on what the claims mean, because infringement asks whether the accused product falls inside those boundaries, and invalidity asks whether the prior art falls inside them too. The two inquiries are linked by a maxim worth memorizing: a claim is construed the same way for both, so a construction broad enough to capture the accused product is also broad enough to capture more prior art -- "that which infringes, if later, would anticipate, if earlier."

The process of authoritatively determining the meaning of disputed claim terms is called claim construction. The Supreme Court held in Markman v. Westview Instruments, Inc., 517 U.S. 370 (1996), that claim construction is a question of law for the judge, not the jury. Because the judge decides it, claim construction is typically resolved at a pretrial proceeding known as a Markman hearing. The judge's claim-construction rulings often decide the case: a favorable construction can make infringement obvious or non-infringement inevitable, and many cases settle once the constructions issue. Indeed, where the parties agree on the structure of the accused product and dispute only what a claim term means, the infringement question collapses entirely into claim construction and can be resolved on summary judgment (Duncan Parking Technologies, Inc. v. IPS Group, Inc., 914 F.3d 1347 (Fed. Cir. 2019)).

The Phillips methodology

How does a judge decide what a claim term means? The governing framework comes from the en banc Federal Circuit decision in Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005). Phillips instructs courts to give claim terms their "ordinary and customary meaning" as understood by a person of ordinary skill in the art (a hypothetical engineer or scientist with the typical training and experience in the relevant field) at the time of the invention, read in light of the patent itself.

Phillips establishes a hierarchy of evidence. The court looks first to intrinsic evidence -- the patent's claims, its specification (the descriptive portion that teaches how to make and use the invention), and its prosecution history (the record of back-and-forth between the applicant and the Patent Office). The specification is, in the Phillips court's words, "the single best guide to the meaning of a disputed term," and it is "usually dispositive." The prosecution history, while intrinsic, is generally "less useful" than the specification because it reflects ongoing negotiation rather than a finished statement. Only if intrinsic evidence leaves ambiguity should the court resort to extrinsic evidence -- dictionaries, treatises, and expert testimony -- which is less reliable because it was not created as part of the patent. As the Federal Circuit had earlier cautioned in Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576 (Fed. Cir. 1996), reliance on extrinsic evidence to vary the meaning supplied by the intrinsic record is improper; Phillips specifically warned against elevating dictionary definitions over the patentee's own usage in the specification.

There are two important guardrails. A court may not improperly import a limitation from a single embodiment in the specification into the claims -- describing one example of the invention does not necessarily narrow the claim to that example. Conversely, the patentee may act as its own lexicographer (defining a term to mean something special) or may disclaim claim scope through clear statements, in which case those definitions and disclaimers control. The threshold for both is exacting: the patentee must "clearly set forth a definition" or make a clear and unmistakable disavowal, and a mere implication is not enough (Thorner v. Sony Computer Entertainment America LLC, 669 F.3d 1362 (Fed. Cir. 2012)). Notably, statements a patent owner makes about claim scope in an inter partes review at the PTAB can themselves create a binding prosecution disclaimer in the district court litigation (Aylus Networks, Inc. v. Apple Inc., 856 F.3d 1353 (Fed. Cir. 2017)) -- a sharp reminder that the two proceedings are not hermetically sealed.

A specialized branch of claim construction governs means-plus-function claims under 35 U.S.C. § 112(f), in which a limitation is written as a "means for" performing a function. Such a limitation is construed to cover only the structure disclosed in the specification for performing that function, and its equivalents -- a deliberately narrowing rule that can sink an infringement theory if the specification fails to disclose corresponding structure.

The role of the appellate standard: Teva

For years it was assumed that claim construction was reviewed entirely fresh ("de novo") on appeal. The Supreme Court refined this in Teva Pharmaceuticals USA, Inc. v. Sandoz, Inc., 574 U.S. 318 (2015), holding that while the ultimate construction remains a legal question reviewed de novo, any subsidiary factual findings the trial judge makes about extrinsic evidence (for example, what a person of skill would have understood a technical term to mean) are reviewed only for "clear error." This gives careful trial-court fact-finding more durability on appeal and rewards a litigant who builds a robust extrinsic-evidence record before the trial judge.

Helios v. Aurora at the Markman hearing

In our hypothetical, the parties dispute the meaning of "reflective coating" in Helios's claim. Helios argues that the term covers any layer that reflects a meaningful portion of incident light, including the dielectric multilayer film Aurora uses. Aurora argues that the specification consistently describes a metallic coating and that the term must be limited to metallic films -- which Aurora's product lacks. The judge reviews the claims, combs the specification for whether "reflective coating" was used broadly or narrowly, checks the prosecution history to see whether Helios disclaimed non-metallic coatings to overcome a prior-art rejection, and hears competing expert testimony. Finding no clear lexicography and no unmistakable disclaimer, the judge construes "reflective coating" to mean "a layer that reflects incident light, whether metallic or dielectric." That construction keeps Aurora's product potentially within the claims and pushes the dispute toward trial. (For a deeper treatment of how claim language and figures interact -- a recurring problem in the design-patent context -- see the LKQ decision and design patent obviousness.)

Part IV -- Proving Infringement

The two-step infringement inquiry

Once the claims are construed, infringement is decided in two steps: (1) construe the claims (done at the Markman stage), and (2) compare the properly construed claims to the accused product or process. Step two is a question of fact, usually for the jury. The patentee bears the burden of proving infringement by a preponderance of the evidence -- meaning more likely than not -- and may do so through any probative evidence (Forest Laboratories, Inc. v. Abbott Laboratories, 239 F.3d 1305 (Fed. Cir. 2001)).

The cardinal rule is the all-elements rule (also called the all-limitations rule): the accused product must contain every element (limitation) recited in the asserted claim. Missing even one limitation defeats infringement of that claim (V-Formation, Inc. v. Benetton Group SpA, 401 F.3d 1307 (Fed. Cir. 2005)). A patentee proves infringement element-by-element, typically through a claim chart -- a table that lines up each claim limitation against the corresponding feature of the accused product -- supported by expert testimony from a person qualified as one of ordinary skill in the art.

Literal infringement

Literal infringement exists when the accused product contains each and every limitation of the claim exactly as construed. If Helios's claim requires "a reflective coating," and Aurora's sensor has a dielectric film that the court has construed to fall within "reflective coating," and the accused product also has every other claimed element (the optical fiber, the specified geometry, and so on), then Aurora literally infringes.

The doctrine of equivalents

What if the accused product does not literally meet a limitation but is only trivially different? Patent law would be hollow if a competitor could escape liability by making an insubstantial change. To prevent that, courts apply the doctrine of equivalents (DOE): an accused product can infringe even without literal correspondence if each missing limitation is met by an equivalent element. The doctrine traces to Graver Tank & Manufacturing Co. v. Linde Air Products Co., 339 U.S. 605 (1950). Courts apply two related tests: the insubstantial-differences test (is the difference between the claimed and accused element insubstantial to a person of skill?) and the function-way-result test (does the accused element perform substantially the same function, in substantially the same way, to achieve substantially the same result?). The function-way-result formulation tends to fit mechanical and electrical inventions; the broader insubstantial-differences inquiry can be more apt in the chemical arts (Mylan Institutional LLC v. Aurobindo Pharma Ltd., 857 F.3d 858 (Fed. Cir. 2017)).

The Supreme Court reaffirmed and disciplined the doctrine in Warner-Jenkinson Co. v. Hilton Davis Chemical Co., 520 U.S. 17 (1997). Warner-Jenkinson held that the doctrine of equivalents must be applied to each claim element individually, not to the invention as a whole, preserving the all-elements rule (DePuy Spine, Inc. v. Medtronic Sofamor Danek, Inc., 469 F.3d 1005 (Fed. Cir. 2006)). It also addressed prosecution history estoppel -- the principle that a patentee who narrowed a claim during prosecution to get the patent allowed cannot later recapture, through the doctrine of equivalents, the territory it surrendered.

The Supreme Court fleshed out estoppel in Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722 (2002). Festo rejected a rigid rule that a narrowing amendment surrenders all equivalents of the amended element. Instead, it created a rebuttable presumption that a narrowing amendment made for reasons of patentability surrenders all equivalents, which the patentee can overcome by showing, for example, that the equivalent was unforeseeable at the time of the amendment, that the rationale for the amendment bore only a tangential relation to the equivalent, or that there was some other reason the patentee could not reasonably have described the equivalent. Where the only barrier to an equivalents theory is estoppel, the question becomes purely legal and a court may resolve infringement (or non-infringement) on summary judgment -- as the Federal Circuit did in finding the "tangential" exception satisfied in Eli Lilly & Co. v. Hospira, Inc., 933 F.3d 1320 (Fed. Cir. 2019).

Two further limits matter. Claim vitiation bars using the doctrine in a way that would read a limitation out of the claim entirely. And the disclosure-dedication rule bars equivalents for subject matter that was disclosed in the specification but not claimed -- the patentee is deemed to have dedicated that unclaimed matter to the public (Eagle Pharmaceuticals Inc. v. Slayback Pharma LLC, 958 F.3d 1171 (Fed. Cir. 2020)). Because of these doctrines, accused infringers find it comparatively easy to win summary judgment of non-infringement (one missing element ends the inquiry), while patentees rarely win summary judgment of infringement under the doctrine of equivalents, which usually turns on contested fact (Leggett & Platt, Inc. v. Hickory Springs Manufacturing Co., 285 F.3d 1353 (Fed. Cir. 2002)).

Helios v. Aurora on infringement

Suppose Aurora has redesigned its product so that, instead of arranging its coatings in the exact stacking order the claim recites, it reverses two of the layers. Helios argues the reversed arrangement is an insubstantial variation that performs the same function (reflecting the relevant wavelengths) in the same way (constructive interference among thin films) to the same result (a temperature-dependent signal). Aurora counters that Helios narrowed this very limitation during prosecution to overcome a prior-art reference, triggering Festo estoppel. The resolution of that estoppel question -- a legal question the judge decides -- may determine whether Helios can even reach the jury on equivalents. If the judge finds the narrowing amendment bore only a tangential relation to the reversed-layer equivalent, Helios survives; if not, the doctrine-of-equivalents theory is dead and Helios must win on literal infringement alone.

Indirect and divided infringement

Beyond direct infringement, Helios may pursue induced infringement (Aurora encouraged customers to use the product in an infringing way, with knowledge under Global-Tech) and contributory infringement (Aurora sold a component especially made for an infringing use, with no substantial non-infringing use, under 35 U.S.C. § 271(c)). Indirect infringement always requires an underlying act of direct infringement by someone -- often the defendant's customers -- and is common where a patented method is performed only when the end user operates the device. Importantly, a defendant's good-faith belief that the patent is invalid is not a defense to inducement; the Supreme Court so held in Commil USA, LLC v. Cisco Systems, Inc., 575 U.S. 632 (2015), reasoning that infringement and validity are distinct questions.

A thorny variant is divided or joint infringement, which arises when no single actor performs every step of a method claim. The Supreme Court held in Limelight Networks, Inc. v. Akamai Technologies, Inc., 572 U.S. 915 (2014), that there can be no inducement without a single act of direct infringement somewhere. On remand, the en banc Federal Circuit (Akamai Technologies, Inc. v. Limelight Networks, Inc., 797 F.3d 1020 (Fed. Cir. 2015)) clarified that the acts of multiple actors may be attributed to one accused infringer for direct-infringement purposes where that infringer "directs or controls" the others' performance or where the actors form a joint enterprise. Drafting method claims that can be performed by a single actor -- and pleading control or a joint enterprise where they cannot -- is a recurring strategic concern, especially for networked and software inventions.

Part V -- Damages Theory Develops Alongside Liability

Although damages are tried alongside liability, the damages case is built throughout discovery, and we introduce it here because it shapes pretrial strategy. We return to the doctrinal detail in Part VIII. For now, understand that two principal categories of compensatory damages exist: lost profits (the profits the patentee would have earned but for the infringement) and a reasonable royalty (a hypothetical license fee), with the statute guaranteeing "in no event less than a reasonable royalty" under 35 U.S.C. § 284. Two statutory limits frame the entire damages period from the outset: under 35 U.S.C. § 286, no damages may be recovered for infringement occurring more than six years before the complaint, and under the marking statute, 35 U.S.C. § 287(a), a patentee that sells products embodying the patent but fails to mark them (or fails to give actual notice) cannot recover pre-notice damages at all. These rules make early attention to the marking question and the notice date essential; see our companion guides on patent marking requirements and patent and trade dress marking requirements. Both parties typically retain damages experts, whose reports and depositions become central battlegrounds, and whose methodologies are frequently attacked under the reliability standard of Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).

Part VI -- Invalidity and the PTAB Interplay

Validity is a defense and a counterclaim

A patent is presumed valid under 35 U.S.C. § 282, and an accused infringer challenging validity in district court must prove invalidity by clear and convincing evidence -- a heightened standard the Supreme Court confirmed in Microsoft Corp. v. i4i Ltd. Partnership, 564 U.S. 91 (2011). That is a meaningfully higher bar than the preponderance standard that governs infringement, and the gap between the two standards is one reason accused infringers increasingly route their best validity arguments to the PTAB, discussed below.

The principal invalidity grounds are:

  • Anticipation under 35 U.S.C. § 102: a single prior-art reference discloses every element of the claim, so the invention is not new.
  • Obviousness under 35 U.S.C. § 103: the claimed invention, though not identically disclosed, would have been obvious to a person of ordinary skill in light of the prior art. The framework comes from Graham v. John Deere Co., 383 U.S. 1 (1966), and was loosened in favor of a flexible, common-sense analysis in KSR International Co. v. Teleflex Inc., 550 U.S. 398 (2007). (We treat obviousness in depth in our companion piece, overcoming obviousness rejections.)
  • Lack of patent-eligible subject matter under 35 U.S.C. § 101: the claim is directed to an abstract idea, law of nature, or natural phenomenon without an "inventive concept," under Alice Corp. v. CLS Bank International, 573 U.S. 208 (2014), building on Mayo Collaborative Services v. Prometheus Laboratories, Inc., 566 U.S. 66 (2012). Eligibility is especially contested for software and business methods, which we cover in patent eligibility after Alice and in our broader survey of the legal protection of software.
  • Inadequate disclosure under 35 U.S.C. § 112: the claim fails the written-description, enablement, or definiteness requirements. The Supreme Court tightened the enablement requirement in Amgen Inc. v. Sanofi, 598 U.S. 594 (2023), and tightened definiteness in Nautilus, Inc. v. Biosig Instruments, Inc., 572 U.S. 898 (2014) (a claim is indefinite if, read in light of the specification and prosecution history, it fails to inform those skilled in the art about the scope of the invention with reasonable certainty).

A patent may also be held unenforceable for inequitable conduct -- a separate doctrine concerning misconduct before the Patent Office, sometimes called the "atomic bomb" of patent law, which we address in detail in inequitable conduct in patent prosecution and in finding evidence of inequitable conduct.

Inter partes review at the PTAB

Since the America Invents Act of 2011, an accused infringer like Aurora has a powerful alternative venue for attacking validity: inter partes review (IPR) before the Patent Trial and Appeal Board (PTAB), an administrative tribunal within the U.S. Patent and Trademark Office. In an IPR, a petitioner may challenge claims only on the grounds of anticipation or obviousness, and only based on patents and printed publications (35 U.S.C. §§ 311-319). Two features make IPR attractive to challengers. First, the burden of proof is a mere preponderance of the evidence, not clear and convincing -- so it is meaningfully easier to invalidate a patent at the PTAB than in court. Second, IPRs are relatively fast (a final written decision is due roughly one year after institution, under 35 U.S.C. § 316(a)(11)) and comparatively inexpensive next to a full district-court trial.

A petitioner must file an IPR petition within one year of being served with an infringement complaint (35 U.S.C. § 315(b)) -- a deadline triggered by a complaint dismissed with prejudice but not by one dismissed without prejudice. The PTAB decides whether to institute review; that institution decision is generally unreviewable (Cuozzo Speed Technologies, LLC v. Lee, 579 U.S. 261 (2016)). If the patent survives, broad estoppel under 35 U.S.C. § 315(e) bars the petitioner from later asserting in district court any ground it "raised or reasonably could have raised" in the IPR -- a much wider net than the narrower, "only-actually-raised" estoppel that applied to the now-sunset covered-business-method (CBM) program. The Supreme Court has shaped this regime repeatedly: SAS Institute Inc. v. Iancu, 584 U.S. 357 (2018) (the PTAB must decide all challenged claims, not a subset, once it institutes); and Thryv, Inc. v. Click-To-Call Technologies, LP, 590 U.S. 45 (2020) (time-bar determinations under § 315(b) are not appealable). When the PTAB issues a final written decision, that decision is itself appealable to the Federal Circuit, so the same patent's validity can be on appeal from two directions at once.

Coordinating the two proceedings: the stay decision

When an IPR and a district-court case proceed in parallel, the accused infringer often moves to stay the litigation pending the PTAB's decision. District courts possess inherent authority to manage their dockets and to grant such stays (Ethicon, Inc. v. Quigg, 849 F.2d 1422 (Fed. Cir. 1988)), and they weigh several factors: whether a stay will simplify the issues, the stage of the litigation, and whether a stay would unduly prejudice or tactically disadvantage the patent owner. Discretionary PTAB practice has also produced Fintiv considerations, under which the Board may decline to institute when a parallel district-court trial is imminent. The strategic choreography between the PTAB and the district court -- which forum decides validity first, how estoppel constrains later arguments, and how a stay affects damages exposure -- is one of the most consequential decisions in modern patent litigation. A challenger who wins at the PTAB may moot the entire district-court case; a challenger who loses may be estopped from raising its best invalidity arguments at trial. Timing cuts both ways: filing the petition early maximizes the odds of a stay, but waiting allows a more complete prior-art search before the estoppel net drops.

In our hypothetical, Aurora files an IPR asserting that Helios's claims are obvious over two earlier sensor patents and moves to stay the district-court case. The judge denies the stay because fact discovery is nearly complete and a trial date is set, leaving Aurora to fight validity on two fronts -- and to live with the prospect that any narrowing statement it makes about claim scope at the PTAB may bind it in court under Aylus.

Part VII -- Summary Judgment and Why Its Denial Sets Up Trial

The summary judgment standard

After discovery, either party may move for summary judgment under Federal Rule of Civil Procedure 56. Summary judgment is appropriate only where there is "no genuine dispute as to any material fact" and the moving party is entitled to judgment as a matter of law. The governing trilogy -- Anderson v. Liberty Lobby, Inc., 477 U.S. 242 (1986); Celotex Corp. v. Catrett, 477 U.S. 317 (1986); and Matsushita Electric Industrial Co. v. Zenith Radio Corp., 475 U.S. 574 (1986) -- instructs the court to view the evidence in the light most favorable to the non-moving party and to ask whether a reasonable jury could return a verdict for that party. A "genuine" dispute is one on which reasonable minds could differ; a "material" fact is one that could affect the outcome under the governing law.

In patent cases, summary judgment can resolve discrete issues: non-infringement (no reasonable jury could find an accused product meets a limitation), infringement (the accused product indisputably meets every limitation, often where the only dispute was claim construction), or invalidity (the prior art indisputably anticipates). But many patent disputes are riddled with competing expert opinions and credibility questions -- exactly the kind of "genuine disputes of material fact" that a court cannot resolve on paper, because weighing expert credibility is the jury's province.

Why denial channels the case to trial

When a court denies summary judgment, it is not declaring that the moving party will lose. It is holding only that the case presents genuine factual disputes that a fact-finder must resolve. For the litigants, denial is a pivotal moment -- the very hinge this guide is named for: the off-ramp of pretrial resolution is closed, and the case is now barreling toward a jury. Several consequences follow immediately.

First, the parties must reassess settlement. Denial often recalibrates each side's view of its odds and the cost of proceeding, and many cases settle in the weeks after a summary-judgment ruling once the realistic path to trial is clear. A defendant whose validity motion was denied now confronts the clear-and-convincing burden in front of a jury; a patentee whose infringement motion was denied now confronts the cost and uncertainty of a full trial.

Second, the parties pivot to intensive trial preparation. They file motions in limine to exclude prejudicial or improper evidence, and Daubert motions to exclude expert testimony that is unreliable or untethered to the facts -- a particularly important tool against a damages expert whose royalty theory ignores apportionment or rests on non-comparable licenses. They prepare the pretrial order, exchange final witness and exhibit lists, designate deposition testimony for witnesses who will not appear live, and submit competing proposed jury instructions and a verdict form.

Third, the denial itself becomes a roadmap. The court's order usually explains which factual disputes precluded summary judgment, telling each side precisely what it must prove or rebut at trial. A defendant whose non-infringement motion was denied learns exactly which limitations the court thinks a jury could find present, and can focus its trial story accordingly. Critically, a denial of summary judgment is generally not itself appealable and does not preserve the underlying legal arguments for appeal; to keep a sufficiency-of-the-evidence challenge alive, the losing party must raise it again at trial through a Rule 50 motion, as Part IX explains.

For Helios and Aurora, the court denies cross-motions for summary judgment: Aurora's expert says the accused coating is dielectric and outside the claim as a factual matter, while Helios's expert says it falls squarely within the construed term. That is a classic battle of experts, and a court may not pick a winner between dueling qualified experts on summary judgment. The judge sends infringement, validity, willfulness, and damages to the jury.

Part VIII -- The Jury Trial

Structure of a patent trial

A patent trial proceeds like other federal civil trials but with a heavy technical overlay. After jury selection (voir dire), each side delivers an opening statement -- a roadmap, not argument -- in which counsel must translate fiber optics and thin-film interference into a narrative a lay jury can follow. The plaintiff then presents its case-in-chief through direct examination of its witnesses, subject to cross-examination by the defense. The defendant then presents its case, the plaintiff may offer rebuttal, and both sides deliver closing arguments. The judge reads the jury instructions, and the jury deliberates.

In a patent trial, witnesses typically include inventors and company personnel (fact witnesses who explain the invention's development and the harm from infringement), technical experts (who walk the jury through the claims and either map them onto the accused product or pull them apart), and damages experts (who quantify the economic harm). Demonstrative exhibits -- animations, exploded-view models, side-by-side comparisons -- are essential to making the technology comprehensible, and must be vetted with the court in advance.

Issue one: infringement

The jury decides infringement under the constructions the judge already fixed; it is not free to revisit what the claim terms mean. The plaintiff must prove, by a preponderance, that the accused product meets every limitation of at least one asserted claim, literally or under the doctrine of equivalents. The defense attacks the proof element-by-element, often conceding most limitations and fighting hard on one or two where it has a genuine technical argument. Because the all-elements rule is unforgiving, the entire infringement verdict can turn on a single contested limitation -- in our case, the "reflective coating."

Issue two: validity

The same jury usually decides validity, applying the clear-and-convincing standard. The defense presents prior art and expert testimony to show anticipation or obviousness; the patentee responds with arguments that the references do not disclose every element, that a skilled artisan would not have combined them, and with objective indicia of non-obviousness (so-called secondary considerations) such as commercial success, long-felt but unmet need, failure of others, industry praise, and copying. There must be a nexus between those indicia and the claimed invention for them to carry weight. A jury that finds a claim infringed but invalid hands the defendant a complete win on that claim.

Issue three: willfulness and the Halo standard

If Helios wants enhanced damages, it must prove willful infringement. The Supreme Court overhauled the willfulness standard in Halo Electronics, Inc. v. Pulse Electronics, Inc., 579 U.S. 93 (2016). Halo discarded the rigid, two-part Seagate test (which had required objective recklessness plus subjective knowledge, proven by clear and convincing evidence) and replaced it with a more flexible inquiry committed to the district court's discretion. After Halo:

  • Willfulness is judged by the preponderance standard, not clear and convincing evidence.
  • The focus is on the infringer's subjective intent and culpability at the time of the challenged conduct, not on an attorney-constructed defense raised later in litigation. An infringer can no longer escape enhancement merely by mustering a reasonable but unsuccessful defense at trial (WBIP, LLC v. Kohler Co., 829 F.3d 1317 (Fed. Cir. 2016)).
  • The conduct that warrants enhancement is the kind of "egregious," "willful, wanton, malicious, bad-faith, deliberate" behavior characteristic of the "wanton and malicious pirate."

Practically, willfulness usually requires proof that the defendant knew of the patent and proceeded anyway. Pre-suit notice -- such as a demand letter from Helios, or citation of the patent during licensing talks -- is the classic way to establish knowledge. A defendant may defend by showing it had a good-faith belief in non-infringement or invalidity, sometimes supported by an opinion of counsel (a written legal opinion obtained before or during the conduct). Two statutory and doctrinal points govern that defense. First, under 35 U.S.C. § 298, an accused infringer's failure to obtain or present an opinion of counsel may not be used to prove willfulness -- abrogating earlier law that had treated the absence of an opinion as evidence of culpability. Second, an infringer that does rely on an opinion of counsel waives privilege over communications on that subject matter (In re EchoStar Communications Corp., 448 F.3d 1294 (Fed. Cir. 2006)), which is why accused infringers frequently move to bifurcate willfulness from liability to defer the privilege-waiver decision. (The parallel role of clearance opinions and good-faith belief in the trademark context is explored in the shield of good faith.)

Importantly, the jury finds the facts of willfulness (did the defendant act with the requisite culpability?), but whether and how much to enhance damages remains a discretionary decision for the judge under 35 U.S.C. § 284. A willfulness finding is a necessary predicate for enhancement, but it does not compel it -- the judge separately weighs whether the conduct is egregious enough (SRI International, Inc. v. Cisco Systems, Inc., 14 F.4th 1323 (Fed. Cir. 2021)).

Issue four: damages

The jury also fixes compensatory damages. Section 284 entitles a prevailing patentee to "damages adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer." The two main routes are lost profits and a reasonable royalty.

Lost profits and the Panduit test

Lost profits compensate the patentee for sales it would have made but for the infringer's presence in the market. To recover, the patentee must prove a reasonable probability that, but for the infringement, it would have made the infringer's sales (Minnesota Mining & Manufacturing Co. v. Johnson & Johnson Orthopaedics, Inc., 976 F.2d 1559 (Fed. Cir. 1992)). The most common framework is the four-factor test of Panduit Corp. v. Stahlin Bros. Fibre Works, Inc., 575 F.2d 1152 (6th Cir. 1978), which requires the patentee to show:

  1. Demand for the patented product;
  2. Absence of acceptable non-infringing substitutes in the market during the accounting period;
  3. The patentee's manufacturing and marketing capability to make the infringer's sales; and
  4. The amount of profit the patentee would have made.

Panduit is not the only path. The Federal Circuit confirmed en banc in Rite-Hite Corp. v. Kelley Co., 56 F.3d 1538 (Fed. Cir. 1995), that Panduit is not the sine qua non of lost profits and that any sound "but for" causation showing will do -- including the two-supplier market theory, under which a patentee that competes in a market with only one rival (the infringer) may recover by proving the two-supplier market and its own capacity. The second Panduit factor is often the decisive battleground: the availability of a non-infringing alternative can cap or eliminate lost profits, because the infringer will argue it would have made the sales lawfully with the substitute (Grain Processing Corp. v. American Maize-Products Co., 185 F.3d 1341 (Fed. Cir. 1999)). The substitute must have been available during the accounting period, not merely conceivable.

Lost profits can extend beyond the infringing units themselves. A patentee may recover for price erosion -- the lower prices it was forced to charge because of the infringer's competition -- and for convoyed sales of unpatented items sold together with the patented product, where the items together produce the utility of the invention and are functionally a single unit rather than merely sold together for convenience (Warsaw Orthopedic, Inc. v. NuVasive, Inc., 778 F.3d 1365 (Fed. Cir. 2015)). And in multi-component products, lost profits must still be apportioned to the patented contribution unless the patented feature drives demand for the whole. Helios, as a direct competitor selling its own sensors, has a plausible lost-profits theory; a patentee that does not practice its invention generally cannot recover lost profits at all and is limited to a royalty.

Reasonable royalty and the Georgia-Pacific factors

Every prevailing patentee is guaranteed at least a reasonable royalty -- the royalty a willing licensor and willing licensee would have agreed to in a hypothetical negotiation at the time infringement began. The canonical list of considerations is the fifteen Georgia-Pacific factors, from Georgia-Pacific Corp. v. United States Plywood Corp., 318 F. Supp. 1116 (S.D.N.Y. 1970), modified, 446 F.2d 295 (2d Cir. 1971). These factors include established royalties for the patent, rates paid for comparable patents, the nature and scope of the license, the commercial relationship of the parties, the profitability and advantages of the patented feature, the availability of non-infringing alternatives, the portion of profit attributable to the invention, and expert opinion. A reasonable royalty is usually expressed as a royalty base (the relevant sales) multiplied by a royalty rate, or as a lump sum.

Apportionment and the smallest salable patent-practicing unit

A central modern constraint on royalties is apportionment: damages must be tied to the value of the patented feature, not the entire product. As the Federal Circuit put it, the patentee is limited to "only those damages attributable to the infringing features" (VirnetX, Inc. v. Cisco Systems, Inc., 767 F.3d 1308 (Fed. Cir. 2014)). The entire market value rule -- which would base a royalty on the whole product's revenue -- is permissible only when the patented feature drives consumer demand for the entire product. Otherwise the patentee must apportion down to the smallest salable patent-practicing unit and further to the value of the patented feature within that unit (LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51 (Fed. Cir. 2012); Commonwealth Scientific & Industrial Research Organisation v. Cisco Systems, Inc., 809 F.3d 1295 (Fed. Cir. 2015)). Royalty theories that ignore apportionment, or that rely on truly non-comparable license agreements, are routinely struck under Daubert or reversed on appeal. (For standard-essential patents, apportionment intersects with FRAND obligations, addressed in standard-essential patents and FRAND licensing; for the underlying art of valuing a license, see how to license your patent.)

In our hypothetical, Helios's damages expert offers a reasonable-royalty opinion of 6% on the sensor units, apportioned to the value the patented coating adds over a conventional sensor, plus a lost-profits claim for the subset of sales Helios proves it would have captured. Aurora's expert counters with a much lower royalty, argues that an available non-infringing redesign caps lost profits at zero under Grain Processing, and moves under Daubert to exclude Helios's expert for failing to apportion the convoyed-sales component. The judge admits the testimony but flags its weaknesses for cross-examination -- a reminder that the line between admissibility (a Daubert question for the judge) and weight (a question for the jury) is fiercely contested in every damages case.

Enhanced damages and attorney's fees

If the jury finds willfulness, the judge may, in discretion, enhance the compensatory award up to three times under 35 U.S.C. § 284, guided by Halo and the older Read Corp. v. Portec, Inc., 970 F.2d 816 (Fed. Cir. 1992), factors (deliberateness of copying, the infringer's investigation and good-faith belief, litigation conduct, the duration of the misconduct, and the like). Separately, in exceptional cases, the court may award reasonable attorney's fees to the prevailing party -- patentee or accused infringer -- under 35 U.S.C. § 285. The Supreme Court relaxed the exceptional-case standard in Octane Fitness, LLC v. ICON Health & Fitness, Inc., 572 U.S. 545 (2014), defining an exceptional case as one that "stands out from others" with respect to the strength of a party's litigating position or the unreasonable manner in which the case was litigated, and lowered the proof standard to a preponderance in the companion case Highmark Inc. v. Allcare Health Management System, Inc., 572 U.S. 559 (2014), which also held that fee awards are reviewed only for abuse of discretion. Fee shifting is the principal financial deterrent against meritless suits brought by patent assertion entities.

The verdict

The jury returns a verdict, often on a detailed special-verdict form that asks separate questions for each claim: infringed? valid? willful? and a damages figure. Counsel may poll the jury to confirm unanimity. Suppose Helios's jury finds that Aurora literally infringes the asserted claim, that the claim is not invalid, that the infringement was willful, and awards $14 million as a reasonable royalty. That verdict is the launching point for the post-trial phase.

Part IX -- Post-Trial Motions

Rule 50: judgment as a matter of law (JMOL)

A jury verdict is not the last word. Under Federal Rule of Civil Procedure 50, a party may ask the court to override the jury when the evidence cannot reasonably support its verdict. There are two stages, and the sequence is a trap for the unwary:

  • Rule 50(a) -- a motion for judgment as a matter of law made during trial, after the opposing party has been fully heard but before the case goes to the jury. The standard is whether "a reasonable jury would not have a legally sufficient evidentiary basis" to find for that party.
  • Rule 50(b) -- a renewed motion for judgment as a matter of law (historically called JNOV, judgment non obstante veredicto) filed after the verdict, within 28 days of judgment. That deadline is non-extendable.

The interplay between the two is governed by a strict preservation rule that decides many appeals before they are briefed. A party may make a Rule 50(b) motion only on grounds it raised in a Rule 50(a) motion before the case went to the jury. A defendant who fails to move under Rule 50(a) on, say, lack of substantial evidence of infringement generally forfeits the ability to raise that argument under Rule 50(b) -- and on appeal. The Supreme Court enforced this rule unforgivingly in Unitherm Food Systems, Inc. v. Swift-Eckrich, Inc., 546 U.S. 394 (2006), holding that a party that fails to renew its motion under Rule 50(b) cannot obtain appellate review of the sufficiency of the evidence at all. The practical lesson is iron-clad: make the Rule 50(a) motion on every issue before the case goes to the jury, identify each ground with specificity, and renew it under Rule 50(b) within 28 days.

The substantive standard on a Rule 50 motion mirrors summary judgment: the court asks whether substantial evidence -- more than a scintilla, but viewed in the light most favorable to the verdict -- supports the jury's findings, and it may not reweigh the evidence or second-guess credibility. In a patent case, JMOL motions are filed on every issue: no reasonable jury could find infringement; the patent is invalid as a matter of law; the willfulness finding lacks support; the damages award is unsupported or legally excessive. Aurora will move for JMOL of non-infringement and invalidity, and to vacate the willfulness finding and the damages award.

Rule 59: new trial and remittitur

Under Rule 59, a party may seek a new trial -- on the ground that the verdict is against the great weight of the evidence, that prejudicial trial errors occurred (improper evidence, flawed jury instructions, attorney misconduct), or that the damages are excessive. The new-trial standard is more forgiving to the movant than JMOL because, unlike on a Rule 50 motion, the court may weigh the evidence and need not view it solely in the verdict-winner's favor. Where damages are excessive but liability is sound, courts often use remittitur: the court offers the plaintiff the choice of a reduced damages figure or a new trial on damages. Rule 59(e) also allows a motion to alter or amend the judgment. Best practice is to file the renewed Rule 50(b) motion and an alternative Rule 59 new-trial motion together, so that if the appellate court reverses a JMOL it can address the conditional new-trial ruling without a remand (Rule 50(c)-(d)).

The judge's discretionary rulings

After the verdict, the judge resolves the matters reserved for the bench: whether to enhance damages for willfulness under § 284, whether the case is exceptional so as to warrant attorney's fees under § 285, the award of pre-judgment and post-judgment interest and costs, and whether to grant an injunction. These rulings can dwarf the jury verdict in practical importance, and each carries its own standard of review on appeal -- abuse of discretion -- that makes the trial judge's reasoning, not the jury's, the focus of the appeal.

Part X -- Injunctions After eBay

The end of the automatic injunction

Historically, courts issued a permanent injunction almost automatically once infringement of a valid patent was found. That ended with eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006). The Supreme Court held that a patentee is not automatically entitled to an injunction; instead, under 35 U.S.C. § 283, which authorizes injunctions "in accordance with the principles of equity," the patentee must satisfy the traditional four-factor test used for injunctions generally. The patentee must show:

  1. That it has suffered an irreparable injury;
  2. That remedies at law (money damages) are inadequate to compensate for that injury;
  3. That, balancing the hardships between the parties, an equitable remedy is warranted; and
  4. That the public interest would not be disserved by an injunction.

eBay reshaped patent remedies. A patentee that practices its invention and competes directly with the infringer -- like Helios -- often can show irreparable harm (lost market share, lost goodwill, erosion of an exclusive position) and obtain an injunction. By contrast, a non-practicing entity (an owner that licenses but does not make products, sometimes pejoratively called a "patent troll") that already monetizes the patent through royalties frequently cannot show that money damages are inadequate, and so is often relegated to an ongoing royalty rather than an injunction. The Federal Circuit has required patentees to demonstrate a causal nexus between the infringement and the alleged irreparable harm -- the harm must flow from the patented feature, not merely from competition generally (Apple Inc. v. Samsung Electronics Co., 809 F.3d 1252 (Fed. Cir. 2015)).

Two practical features of patent injunctions deserve emphasis. First, an injunction must be carefully tailored: it binds only the parties and those in active concert with them (Fed. R. Civ. P. 65(d)), and it cannot sweep in "colorably different" redesigns, which must be tested separately in contempt proceedings (TiVo Inc. v. EchoStar Corp., 646 F.3d 869 (Fed. Cir. 2011) (en banc)). Second, the eBay framework applies in district court but not at the U.S. International Trade Commission, where Section 337 investigations can yield exclusion orders barring importation under a different public-interest standard (Spansion, Inc. v. International Trade Commission, 629 F.3d 1331 (Fed. Cir. 2010)) -- which is one reason the ITC remains an attractive forum for patentees seeking exclusionary relief.

Ongoing royalties as the alternative

When an injunction is denied, the court typically imposes an ongoing royalty for the infringer's continued sales (Bard Peripheral Vascular, Inc. v. W.L. Gore & Associates, Inc., 670 F.3d 1171 (Fed. Cir. 2012)). That ongoing rate may well exceed the jury's pre-verdict royalty rate, because the post-verdict infringer is now an adjudicated infringer of a patent already held valid and infringed, changing the leverage in the (now-counterfactual) negotiation. Courts may also stay an injunction pending appeal upon an appropriate showing and bond.

For Helios, a head-to-head competitor with a strong showing of lost market share and a clear causal nexus between that harm and the patented coating, the court grants a permanent injunction barring Aurora from making or selling the infringing sensor, tailored to allow Aurora to sell any genuinely redesigned, non-infringing product.

Part XI -- Appeal to the Federal Circuit

A single appellate court for patents

Almost all patent appeals go to one specialized court: the U.S. Court of Appeals for the Federal Circuit, which has nationwide, near-exclusive jurisdiction over appeals in patent cases under 28 U.S.C. § 1295. This centralization was designed to bring uniformity to patent law. A losing party must file a notice of appeal within 30 days of the final judgment under Federal Rule of Appellate Procedure 4(a) (60 days when the United States is a party). One quirk: on issues unique to patent law the Federal Circuit applies its own law, but on procedural and evidentiary questions not unique to patents (for example, the conduct of trial or the admission of non-expert evidence) it generally applies the law of the regional circuit from which the appeal arose.

Standards of review

What the Federal Circuit can change on appeal depends on the standard of review, which differs by issue -- and which is often the single most important strategic variable in framing an appeal, because the most deferential standards make reversal nearly impossible:

  • Claim construction is reviewed de novo as to the ultimate legal conclusion, but subsidiary factual findings about extrinsic evidence are reviewed for clear error (Teva, discussed above). Because so much turns on construction, this is often the most consequential issue on appeal, and a reversed construction can unravel an entire verdict and require a new trial.
  • Jury findings of fact -- infringement, anticipation, the factual predicates of obviousness, the facts of willfulness, and the amount of damages -- are reviewed under the deferential substantial evidence standard. The court asks only whether a reasonable jury could have reached the result, not whether it would have reached the same result itself.
  • Bench findings of fact -- for example, in a bench trial on inequitable conduct or on injunction factors -- are reviewed for clear error under Rule 52(a)(6), reversible only if the appellate court is left with a "definite and firm conviction" that a mistake was made.
  • Obviousness and patent eligibility are ultimately questions of law reviewed de novo, but they rest on underlying factual findings reviewed for substantial evidence (jury) or clear error (bench).
  • Discretionary rulings -- enhancement of damages, exceptional-case fee awards, the grant or denial of an injunction, evidentiary rulings, and new-trial decisions -- are reviewed for abuse of discretion (Halo; Highmark).

Because the standard of review is so outcome-determinative, appellate counsel often work hard to characterize a contested issue as a legal question (reviewed de novo) rather than a factual one (reviewed deferentially) -- for instance, by recasting a jury non-infringement finding as flowing from an erroneous claim construction.

Possible outcomes

The Federal Circuit may affirm, reverse, vacate, or remand (send the case back for further proceedings), often in combination. A reversal of claim construction frequently requires a new trial under the corrected construction. A finding that the damages methodology was unreliable may produce a remand for a new damages trial while leaving liability intact. A party that loses at the Federal Circuit may petition for rehearing en banc (review by the full court) or seek certiorari in the Supreme Court, which hears only a handful of patent cases each term.

If Aurora's IPR (Part VI) is still pending, the Federal Circuit may also be hearing an appeal from the PTAB on the same patent's validity. A PTAB decision invalidating the claims, once final and affirmed, can extinguish the district-court judgment even after a jury verdict and an affirmance on infringement -- a stark reminder of why the PTAB-litigation interplay deserves early, careful strategy. The patent that wins a $14 million verdict in court can still be erased by an administrative tribunal applying a lower burden of proof.

Key Takeaways

A patent infringement case is best understood as a sequence of gates, each of which can end or reshape the dispute. Venue under TC Heartland and In re Cray dictates where the fight happens. Adequate pleading after Form 18's demise -- and detailed infringement contentions under local patent rules -- gets the plaintiff through the courthouse door. Claim construction under Markman and Phillips fixes the meaning of the patent and frequently decides the case, with the Teva standard rewarding a strong trial-court record. Infringement is proved element-by-element, literally or under the disciplined doctrine of equivalents shaped by Warner-Jenkinson and Festo, with special rules for indirect and divided infringement. Validity can be attacked in court (clear and convincing evidence) or, often more effectively, at the PTAB (preponderance), making the stay-and-estoppel calculus pivotal. Summary judgment can dispose of issues, but its denial signals that genuine factual disputes must go to a jury, triggering intense trial preparation -- and it neither decides the case nor, by itself, preserves the loser's arguments for appeal. At trial, the jury decides infringement, validity, willfulness (under Halo), and damages (under Georgia-Pacific and Panduit, constrained by apportionment, price erosion, and convoyed-sales doctrine). After the verdict, Rule 50 and Rule 59 motions -- governed by the strict preservation rule of Unitherm -- enhancement, fee awards under Octane, and injunctions under eBay can transform the outcome. And the Federal Circuit, applying issue-specific standards of review, has the last judicial word short of the Supreme Court.

For any party facing this gauntlet, the lesson is that strategy must be integrated from day one. Venue choices affect transfer fights; pleading and marking choices affect damages exposure; claim-construction positions must be consistent with infringement and invalidity theories and with anything said at the PTAB; the PTAB decision must be coordinated with the trial calendar and the estoppel net; the damages model must respect apportionment or risk exclusion under Daubert; and every sufficiency argument must be teed up under Rule 50(a) or it is lost forever. Patent litigation rewards parties who think backward from the appeal and forward from the complaint at the same time.

Frequently Asked Questions

What does it mean when a court denies summary judgment in a patent case? It means the court has concluded that the case presents genuine disputes of material fact -- typically competing expert opinions about infringement, validity, or damages -- that a jury must resolve. Denial does not mean the moving party will lose; it simply removes the possibility of resolving those issues on paper and channels the case toward trial. The denial order usually explains which factual disputes remain, giving each side a roadmap for trial. Note that a denial is generally not appealable on its own, and it does not preserve the underlying legal arguments; to keep a sufficiency challenge alive, the loser must renew it at trial under Rule 50.

Why is claim construction so important? Because the claims define the legal boundary of the invention, the judge's interpretation of disputed claim terms often decides whether the accused product infringes and whether the patent is valid over the prior art. Under Markman v. Westview Instruments, the judge -- not the jury -- construes the claims, and a single favorable or unfavorable construction can effectively end the case. The same construction governs both infringement and validity, so a litigant cannot argue for a broad construction to capture the accused product and a narrow one to dodge the prior art.

How is willful infringement proven after Halo v. Pulse? After Halo Electronics, Inc. v. Pulse Electronics, Inc., willfulness is judged by a preponderance of the evidence and focuses on the infringer's subjective culpability at the time of the conduct -- usually whether the defendant knew of the patent (often through a notice letter) and infringed anyway. A reasonable but unsuccessful litigation defense no longer immunizes the infringer. The jury finds the willfulness facts, but the judge decides, in discretion, whether to enhance damages up to three times under 35 U.S.C. § 284. Under 35 U.S.C. § 298, the absence of an opinion of counsel cannot be used to prove willfulness.

What is the difference between lost profits and a reasonable royalty? Lost profits compensate a patentee that competes in the market for the sales and pricing it lost because of the infringer, and require a "but for" causation showing, often through the four-factor Panduit test (or a two-supplier-market theory) and frequently including price erosion and convoyed sales. A reasonable royalty is the hypothetical license fee a willing licensor and licensee would have negotiated, analyzed under the fifteen Georgia-Pacific factors and constrained by apportionment. Every prevailing patentee gets at least a reasonable royalty; only those who practice the patent and can prove causation can recover lost profits.

Can a patentee always get an injunction after winning? No. Since eBay Inc. v. MercExchange, a prevailing patentee must satisfy the traditional four-factor equitable test -- irreparable harm, inadequacy of money damages, balance of hardships, and the public interest -- and show a causal nexus between the patented feature and the harm. Direct competitors often qualify; non-practicing entities that already license the patent frequently do not and are instead awarded an ongoing royalty for continued infringement. The eBay test does not apply at the ITC, where exclusion orders follow a different standard.

Why do so many patent cases involve the PTAB? Inter partes review before the Patent Trial and Appeal Board lets an accused infringer challenge a patent's validity on anticipation and obviousness grounds under a lower preponderance standard, faster and cheaper than a district-court trial. A successful IPR can moot the litigation, but losing one triggers statutory estoppel that bars the challenger from re-raising any ground it raised or reasonably could have raised -- which is why coordinating the two proceedings, and watching the one-year filing deadline under § 315(b), is one of the most important strategic decisions in a modern patent case.

What happens if a defendant forgets to file a Rule 50(a) motion at trial? It can be catastrophic. Under Unitherm Food Systems, Inc. v. Swift-Eckrich, Inc., a party that does not move for judgment as a matter of law under Rule 50(a) before the case goes to the jury -- and then renew it under Rule 50(b) within 28 days -- forfeits any later challenge to the sufficiency of the evidence, both in the district court and on appeal. The argument that "no reasonable jury could have found infringement" is simply unavailable if it was not preserved at the right moment.

Related Articles

This article is provided by mclaw.io for general informational purposes only. It is not legal advice, does not create an attorney-client relationship, and may not reflect the most current legal developments. Patent litigation is fact-specific and procedurally complex; readers facing an actual or threatened dispute should consult qualified patent litigation counsel about their particular circumstances.