Toolkit · Intellectual Property

Litigating Patent Infringement: A Roadmap and Research Guide

Type
Toolkit
Jurisdiction
National
Date
2025-02-11
Topic
Intellectual Property

This toolkit is a stage-by-stage roadmap and curated research guide to United States patent infringement litigation, written for the digital law library at UpLaw. It follows a case from pre-suit diligence and venue selection, through pleadings, claim construction, discovery, proof of infringement, validity attacks in district court and at the PTAB, damages, trial, and appeal to the Federal Circuit, and closes with the cross-border dimension of modern patent wars. Each stage explains the governing statutes and controlling cases, flags the circuit nuances and edge cases that decide real disputes, and links to the deeper UpLaw guides and checklists that treat each step in full. Worked hypotheticals illustrate how the doctrine actually bites. This is general legal information, not legal advice; confirm current rules and authority before acting.

Litigating Patent Infringement: A Roadmap and Research Guide

Intellectual Property > Patent Litigation | Published 11 February 2025 - Updated 1 May 2025 | UpLaw - uplaw.us

A field map for the whole arc of a United States patent case — from the diligence you owe a court before filing, through claim construction and the parallel administrative track at the PTAB, to damages, trial, and the specialized appeal that reviews it all. Use it to orient a matter and then jump to the deep-dive resource for the stage in front of you.

Keywords: patent litigation, patent infringement, claim construction, Markman hearing, TC Heartland venue, inter partes review, PTAB, doctrine of equivalents, reasonable royalty, Georgia-Pacific, willful infringement, Federal Circuit appeal, 35 U.S.C. § 271, design patent damages, FRAND

Why patent cases sit in a category of their own

Most civil litigation asks a fact-finder to sort out what happened. A patent case asks it to do that and something stranger: to decide what a set of carefully lawyered words means, whether a machine or a line of code falls inside those words, and whether the government should have granted the words in the first place. Layered on top is a procedural architecture found nowhere else in the federal system. The trial happens in an ordinary district court, but the substantive law comes almost entirely from one appellate court, and the validity of the patent-in-suit can be re-decided, under a different burden of proof, by an administrative tribunal running on its own clock. A litigator who treats a patent case as generic commercial litigation with a technical overlay will lose to one who understands the machine.

Federal courts have exclusive jurisdiction over any civil action arising under the patent laws. 28 U.S.C. § 1338(a); see Gunn v. Minton, 568 U.S. 251 (2013). Appeals do not fan out to the regional circuits the way they do in most fields; they funnel into a single tribunal, the United States Court of Appeals for the Federal Circuit, which holds exclusive jurisdiction over appeals in cases with a patent claim or compulsory patent counterclaim. 28 U.S.C. § 1295(a)(1). And because so much validity practice now runs in parallel before the Patent Trial and Appeal Board (PTAB) at the U.S. Patent and Trademark Office, a litigation plan that ignores the administrative track is a plan with a hole in it. For the surrounding federal-civil scaffolding on which everything here rests, anchor on the Federal Civil Litigation roadmap and resource guide and, for a narrative walk through a single company's patent fight, the patent-focused field guide to federal litigation for small businesses.

This roadmap is written for four readers at once: the patent owner weighing whether to enforce, the accused company sizing up its exposure, the in-house lawyer riding herd on outside counsel, and the generalist litigator handed a patent matter and a short runway. It is a map, not the territory. At each stage it names the controlling authority and then points to the UpLaw resource that treats the topic in depth.

The shape of a patent case: eleven decision points

  1. Pre-suit diligence — Rule 11 investigation, claim charts, standing, and validity exposure.
  2. Venue and forum — where the case can be filed, where it should be, and whether the ITC is a better door.
  3. Pleadings and early motions — the complaint, contentions, the answer, and the fights that set the tempo.
  4. Claim construction — the Markman hearing that fixes what the claims mean.
  5. Discovery — source code, financials, protective orders, and dueling experts.
  6. Infringement — literal, equivalents, indirect, divided, and the territorial limit.
  7. Validity and the PTAB — invalidity in court versus inter partes review, and how the two collide.
  8. Damages — lost profits, reasonable royalty, apportionment, marking, and enhancement.
  9. Trial — making unseeable technology legible to a jury while the judge keeps the equitable issues.
  10. Appeal — the Federal Circuit, its standards of review, and the road to certiorari.
  11. The global case — parallel foreign litigation, the Unified Patent Court, and FRAND.

Two research spines run the length of the case. The first is the public docket: filings and orders on PACER (pacer.uscourts.gov) and their free mirror on CourtListener / RECAP (courtlistener.com), where you can pull protective-order models, Markman orders, and damages verdicts from analogous cases. The second is the PTAB's own record, searchable through the Board's end-to-end system. The substantive statutes to keep at your elbow are 35 U.S.C. §§ 271 (infringing acts), 281 (remedy and standing), 282 (defenses and the presumption of validity), 284 (damages and enhancement), 285 (fees), and 287 (marking), plus the venue and jurisdiction provisions in 28 U.S.C. §§ 1338, 1400(b), 1404(a), and 1295.

Stage 1 — Before the complaint: diligence, standing, and the Rule 11 problem

Patent litigation is expensive enough that the most important decisions are made before a complaint exists. Federal Rule of Civil Procedure 11 requires a good-faith, evidence-backed basis for every infringement allegation, and in patent practice that basis takes a specific form: a claim chart that places each element (each "limitation") of an asserted claim next to the corresponding feature of the accused product or process. The exercise is not a formality. The all-elements rule means that if the accused product lacks even one claimed limitation — literally and under the doctrine of equivalents — there is no infringement of that claim, full stop. A chart with an empty cell is a case with a hole.

Courts enforce the diligence duty with real teeth. In Source Vagabond Systems Ltd. v. Hydrapak, Inc., the Federal Circuit affirmed a $200,000 Rule 11 sanction against a patent owner's counsel who pressed a claim construction that the intrinsic record could not bear. 753 F.3d 1291, 1298 (Fed. Cir. 2014). The lesson is that pre-suit claim construction is not optional; you must read the claims the way a court will and then decide whether the accused product still reads on them.

Standing is the second gate. Only a party holding the patent's exclusionary rights may sue, and the chain of title has to be clean and complete. 35 U.S.C. § 281. Assignments recorded in the wrong order, an inventor whose rights were never actually conveyed, or a licensee without all substantial rights can each sink a case at the threshold — sometimes years in, after the merits have been briefed. Confirm title before you draft, not after you file. The employee invention-assignment guide explains where chains of title break and how present-assignment language repairs them.

The third piece of diligence is a candid look at the patent's own validity. A patent you are about to assert is a patent an opponent will attack, and the cheapest time to find the killing prior art is before you have spent a fortune. Read the file history, run the prior-art search you would run if you were the defendant, and price the risk that the claims fall in an inter partes review. If the asserted claims cover software or a business method, stress-test them against subject-matter eligibility under the Alice/Mayo framework before you commit; a § 101 problem can end a case at the pleadings. Alice Corp. v. CLS Bank International, 573 U.S. 208 (2014); Mayo Collaborative Services v. Prometheus Laboratories, Inc., 566 U.S. 66 (2012). The Section 101 strategy guide for software and business-method inventions maps that terrain.

For the accused company, this stage is where earlier caution pays. A documented freedom-to-operate analysis or a competent non-infringement or invalidity opinion, obtained before the accused product launched, is the classic tool for blunting a later willfulness claim. Willfulness now turns on the infringer's subjective state of mind at the time of the challenged conduct, so contemporaneous good-faith diligence matters. Halo Electronics, Inc. v. Pulse Electronics, Inc., 579 U.S. 93 (2016). Build that record early — see the freedom-to-operate analysis guide and its companion FTO search checklist.

A patent owner is often tempted to open with a demand letter. Do so with care. A letter specific enough to communicate a real threat of suit can hand the recipient a live controversy and let it race to its chosen forum with a declaratory-judgment action for non-infringement and invalidity. 28 U.S.C. § 2201. Treat the enforcement decision as the business decision it is, using the checklist for evaluating a new civil case and the fuller early-case-assessment framework, and if you do write, write to the standards in the demand-letter basics guide.

Worked example (hypothetical). Northwind Cells, a hypothetical maker of silicon-anode battery packs for delivery drones, believes a competitor's new pack infringes its patent on a layered anode structure. Before filing, counsel charts each of the three independent claims against a teardown of the rival's cell, confirms that Northwind's founder assigned the patent to the company in 2019 (and that the assignment was recorded), and commissions a prior-art search that turns up a Japanese application counsel decides is distinguishable but not trivially so. Counsel drafts a demand letter but, wary of a declaratory-judgment race to a distant district, sends a narrow notice that preserves marking damages without threatening imminent suit. Only then does the enforcement decision go to the board.

Anchor authorities: Fed. R. Civ. P. 11; 35 U.S.C. §§ 271, 281; 28 U.S.C. § 2201; Source Vagabond Sys. Ltd. v. Hydrapak, Inc., 753 F.3d 1291 (Fed. Cir. 2014); Halo Elecs., Inc. v. Pulse Elecs., Inc., 579 U.S. 93 (2016); Alice Corp. v. CLS Bank Int'l, 573 U.S. 208 (2014).

Stage 2 — Choosing the battlefield: venue, forum, and transfer

Where a patent case is filed shapes almost everything downstream: how fast it moves, which local patent rules govern the calendar, how the assigned judge handles claim construction, and how a jury pool is likely to lean. For a decade, plaintiffs exploited a permissive reading of the venue statute to concentrate cases in the Eastern District of Texas. The Supreme Court closed that door in TC Heartland LLC v. Kraft Foods Group Brands LLC, holding that for a domestic corporation, "resides" in the patent venue statute means the state of incorporation, and nothing broader. 581 U.S. 258 (2017). Under 28 U.S.C. § 1400(b), a domestic corporate defendant may be sued for patent infringement only (1) where it is incorporated, or (2) where it has committed acts of infringement and has a "regular and established place of business."

The Federal Circuit then put content into that second prong. In In re Cray Inc., the court held that a "regular and established place of business" requires a physical, geographical place in the district; a place that is regular and established, not sporadic; and a place of the defendant, not merely of an employee. 871 F.3d 1355 (Fed. Cir. 2017). A home office a sales rep happens to keep is usually not enough; a leased facility with the company's name on the door usually is. The practical result was a redistribution of the national patent docket, with Delaware (where much of corporate America is incorporated) and the Western District of Texas absorbing large shares.

Venue that is proper is not always venue that is convenient. A defendant can move to transfer under 28 U.S.C. § 1404(a) to a more convenient district, and the Federal Circuit polices § 1404 rulings aggressively through mandamus, repeatedly ordering transfers out of plaintiff-friendly divisions where the operative facts and witnesses sit elsewhere. Forum analysis therefore runs on two tracks at once: is venue proper, and if so, will it hold against a transfer motion?

A patent owner should also weigh a different door entirely. The U.S. International Trade Commission hears patent-based complaints under Section 337 of the Tariff Act, 19 U.S.C. § 1337, and its signature remedy is an exclusion order barring infringing imports at the border — powerful leverage against a defendant that manufactures abroad. The ITC moves fast, does not award damages, and requires proof of a domestic industry, so it complements rather than replaces a district-court suit. For the strategic interplay among district court, the ITC, and foreign forums, see the cross-border patent litigation guide.

Worked example (hypothetical). Northwind's rival is incorporated in Delaware, assembles its packs in Malaysia, and imports them through the Port of Long Beach. Post-TC Heartland, Northwind cannot simply pick a favorable Texas division; it files in Delaware, where venue is secure because the defendant resides there. Because the packs are imported, Northwind also files a parallel Section 337 complaint at the ITC, aiming an exclusion order at the supply chain while the Delaware damages case proceeds on a slower track.

Anchor authorities: 28 U.S.C. §§ 1400(b), 1404(a); 19 U.S.C. § 1337; TC Heartland LLC v. Kraft Foods Grp. Brands LLC, 581 U.S. 258 (2017); In re Cray Inc., 871 F.3d 1355 (Fed. Cir. 2017).

Stage 3 — Framing the fight: pleadings, contentions, and early motions

The complaint must do more than announce a grievance; it must plausibly allege that an identified product infringes identified claims. A patent complaint that merely parrots the statute will not survive a motion to dismiss, but one that ties the accused product to the claim language with enough specificity to make infringement plausible on its face will. See Fed. R. Civ. P. 8(a)(2); CODA Development S.R.O. v. Goodyear Tire & Rubber Co., 916 F.3d 1350 (Fed. Cir. 2019); K-Tech Telecommunications, Inc. v. Time Warner Cable, Inc., 714 F.3d 1277 (Fed. Cir. 2013). Do not forget the mechanics that decide cases on procedure rather than merits: proper service under Rule 4 and the ninety-day service window of Rule 4(m), and the defendant's twenty-one-day clock to respond under Rule 12(a). For the full choreography of federal filings, see the comprehensive guide to federal civil litigation filings.

Many districts overlay the Federal Rules with local patent rules that force early, binding disclosures. The patent owner must serve infringement contentions — element-by-element charts for each accused product — often within weeks of the scheduling conference; the accused infringer answers with invalidity contentions identifying prior art and grounds. These contentions discipline the case: they narrow the asserted claims, lock in theories, and are amendable only for good cause. Read the local rules of your assigned court before the scheduling conference, because the contention calendar, not the Federal Rules default, will govern your first hundred days.

The answer typically raises the affirmative defenses catalogued in 35 U.S.C. § 282 — non-infringement, invalidity, unenforceability (for example, inequitable conduct), license, and exhaustion — and usually pairs them with declaratory-judgment counterclaims for non-infringement and invalidity, so that the accused infringer keeps control of the validity question even if the patent owner later tries to drop the patent.

Early motion practice frequently decides momentum. Three moves recur:

  • A § 101 motion to dismiss. Because subject-matter eligibility is often a legal question resolvable on the intrinsic record, a defendant can attack software, diagnostic, or business-method claims at the pleadings under Alice and Mayo, sometimes ending the case before discovery. The Section 101 eligibility checklist walks the two-step.
  • A motion to transfer under § 1404(a), discussed above, which can relocate the entire fight.
  • A motion to stay pending PTAB review. If the defendant has filed an inter partes review petition, it will usually move to stay the district-court case. Courts weigh whether a stay will simplify the issues, the stage of the litigation, and prejudice to the patent owner; district courts possess broad discretion to manage their dockets in this posture. Ethicon, Inc. v. Quigg, 849 F.2d 1422 (Fed. Cir. 1988); see VirtualAgility Inc. v. Salesforce.com, Inc., 759 F.3d 1307 (Fed. Cir. 2014).

A patent owner who needs to stop ongoing sales may seek a preliminary injunction, but the bar is high. The accused infringer can usually defeat the motion by raising a substantial question of invalidity or non-infringement, Altana Pharma AG v. Teva Pharmaceuticals USA, Inc., 566 F.3d 999 (Fed. Cir. 2009), and even at final judgment, injunctive relief is governed by the four-factor equitable test of eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006), which ended the old presumption that a prevailing patentee automatically gets an injunction. For the small-business vantage on these opening moves, see the stage-by-stage guide to federal civil litigation for small businesses.

Anchor authorities: Fed. R. Civ. P. 8, 12, 4(m); 35 U.S.C. § 282; CODA Dev. S.R.O. v. Goodyear Tire & Rubber Co., 916 F.3d 1350 (Fed. Cir. 2019); eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006); VirtualAgility Inc. v. Salesforce.com, Inc., 759 F.3d 1307 (Fed. Cir. 2014).

Stage 4 — What the words mean: claim construction and the Markman hearing

Claim construction is the fulcrum of a patent case. Because the claims define the metes and bounds of the patent right, deciding what a disputed term means often decides infringement and validity in one stroke — which is why a favorable construction so frequently triggers settlement or teed-up summary judgment. The Supreme Court settled the allocation of that decision in Markman v. Westview Instruments, Inc., holding that claim construction is a question for the judge, not the jury. 517 U.S. 370 (1996). Hence the "Markman hearing," a mini-trial on meaning that most patent cases stage after contentions and before the close of expert discovery.

The interpretive method comes from the en banc decision in Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005). Courts read a disputed term as a person of ordinary skill in the art would have understood it at the time of the invention, giving primacy to the intrinsic evidence — the claims themselves, the specification, and the prosecution history — and turning to extrinsic evidence such as expert testimony, treatises, and dictionaries only as a secondary aid. The default is the term's ordinary and customary meaning, which yields only when the patentee acted as a lexicographer, defining the term, or clearly disavowed claim scope. Thorner v. Sony Computer Entertainment America LLC, 669 F.3d 1362 (Fed. Cir. 2012). Statements the patentee made to distinguish prior art — in prosecution, and now even in an IPR — can narrow the claims by disclaimer. Aylus Networks, Inc. v. Apple Inc., 856 F.3d 1353 (Fed. Cir. 2017).

Two further doctrines live at this stage. First, a court cannot duck a genuine dispute about a term's scope by reciting that it carries its "plain meaning"; it must actually resolve the disputed scope. See O2 Micro International Ltd. v. Beyond Innovation Technology Co., 521 F.3d 1351 (Fed. Cir. 2008). Second, a claim that fails to inform a skilled artisan of its scope with reasonable certainty is invalid as indefinite under 35 U.S.C. § 112. Nautilus, Inc. v. Biosig Instruments, Inc., 572 U.S. 898 (2014). Indefiniteness is often litigated inside the Markman process, because construing a term and deciding whether it can be construed at all are two faces of the same inquiry.

Appellate review of claim construction has a wrinkle worth knowing at trial. The ultimate construction is a legal conclusion reviewed de novo, but any subsidiary findings of fact — for instance, what a skilled artisan understood a technical term to mean based on expert testimony — are reviewed only for clear error. Teva Pharmaceuticals USA, Inc. v. Sandoz, Inc., 574 U.S. 318 (2015). That allocation rewards a patentee who builds a factual record around a favorable construction, because facts are stickier on appeal than pure questions of law.

Worked example (hypothetical). In the Northwind case, everything turns on the claim term "substantially continuous barrier layer." Northwind argues the term covers any layer that blocks lithium migration across at least 90% of the interface; the defendant argues "continuous" means gap-free and that its layer, which is deliberately perforated, falls outside. The specification describes an embodiment with "microscopic interruptions," and during prosecution Northwind distinguished a prior-art reference precisely because that reference's layer had large gaps. The court, following Phillips, reads "substantially continuous" to permit microscopic interruptions but not the defendant's engineered perforations — a construction that narrows Northwind's literal-infringement theory and pushes it toward the doctrine of equivalents.

Anchor authorities: 35 U.S.C. § 112; Markman v. Westview Instruments, Inc., 517 U.S. 370 (1996); Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005) (en banc); Teva Pharms. USA, Inc. v. Sandoz, Inc., 574 U.S. 318 (2015); Thorner v. Sony Computer Entm't Am. LLC, 669 F.3d 1362 (Fed. Cir. 2012); Nautilus, Inc. v. Biosig Instruments, Inc., 572 U.S. 898 (2014).

Stage 5 — Building the record: discovery in patent cases

Patent discovery is heavier and more technical than most civil discovery, and it runs on the ordinary Federal Rules — Rules 26, 30, 33, 34, 36, and 37 — with patent-specific pressure points bolted on. The document universe includes engineering design records, product requirements, and above all source code, produced under a stringent protective order that typically confines review to a standalone, non-networked computer in a secure room, with printing limits and an "attorneys' eyes only" tier. Damages discovery pulls in sales data, cost accounting, licensing histories, and comparable-license agreements. Prior-art discovery reaches inventor notebooks, conception and reduction-to-practice records, and third-party subpoenas to prior users.

Two features distinguish the patent discovery calendar. First is the fact-versus-expert sequence: local patent rules and the scheduling order stagger fact discovery, opening expert reports, rebuttal reports, and expert depositions, and missing an expert-disclosure deadline can preclude a theory under Rule 37(c)(1). See the guide to Rule 37(c)(1) preclusion of undisclosed evidence. Second is the centrality of experts under Federal Rule of Evidence 702: infringement, invalidity, and damages each typically require their own expert, and each expert's methodology will be tested by a Daubert motion before trial.

Because the discovery toolkit is shared across all federal litigation, lean on the specialized library rather than reinventing it. Start with the federal discovery toolkit and the refresher on mastering federal discovery tools, rules, and pitfalls. Open the phase with the Rule 26 initial-disclosures and discovery-plan checklist; respond to document demands with the Rule 34 response checklist; and manage the depositions of inventors, engineers, and experts with the federal deposition toolkit and the comprehensive guide to taking depositions. Privilege is acute in patent cases, because opinion-of-counsel and prosecution communications sit near the line and a willfulness or inequitable-conduct fight can put them in play; log them carefully using the privilege-log checklist.

Anchor authorities: Fed. R. Civ. P. 26, 30, 33, 34, 36, 37; Fed. R. Evid. 702; Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).

Stage 6 — Proving (or defeating) infringement

Liability turns on 35 U.S.C. § 271, and the analysis is more textured than "did they copy it."

Direct infringement under § 271(a) requires that the accused product or process satisfy every limitation of an asserted claim — the all-elements rule again — either literally or under the doctrine of equivalents. Equivalence asks whether any missing limitation is met by something that differs only insubstantially, often framed as performing substantially the same function in substantially the same way to reach substantially the same result. Warner-Jenkinson Co. v. Hilton Davis Chemical Co., 520 U.S. 17 (1997). The doctrine has a powerful check: prosecution history estoppel bars a patentee from recapturing, through equivalents, subject matter it surrendered by amendment or argument during prosecution. Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722 (2002). A narrowing amendment made to secure allowance presumptively surrenders the territory between the original and amended claim, and the patentee bears the burden to rebut that presumption.

Indirect infringement requires knowledge and intent. Induced infringement under § 271(b) makes liable one who actively encourages another's direct infringement, but only if the inducer knew of the patent and knew that the induced acts constituted infringement; willful blindness satisfies the knowledge element, but mere negligence does not. Global-Tech Appliances, Inc. v. SEB S.A., 563 U.S. 754 (2011). A defendant's good-faith belief that the patent is invalid is not a defense to inducement, the Supreme Court held in Commil USA, LLC v. Cisco Systems, Inc., 575 U.S. 632 (2015), because infringement and validity are distinct questions. Contributory infringement under § 271(c) reaches the sale of a component especially made for an infringing use that is not a staple article suitable for substantial non-infringing use.

Divided infringement is the trap that catches method claims performed by more than one actor. There is no direct infringement of a method unless a single entity performs, or is attributable with, all the steps; and there can be no inducement without an underlying act of direct infringement. Limelight Networks, Inc. v. Akamai Technologies, Inc., 572 U.S. 915 (2014). The Federal Circuit later widened attribution to reach steps performed by others when the accused party directs or controls that performance, or when the actors form a joint enterprise — an important escape hatch for patentees drafting around multi-party systems.

Several defenses and limits round out the picture. Patent exhaustion cuts off the patentee's rights in a particular article once it makes an authorized sale, and that first sale exhausts rights even when it happens abroad and even when the sale purports to impose post-sale restrictions. Impression Products, Inc. v. Lexmark International, Inc., 581 U.S. 360 (2017). Territoriality generally confines § 271 to acts in the United States, but § 271(f) reaches the supply of components from the United States for assembly abroad, and a patentee can recover foreign lost profits flowing from such domestic infringement. WesternGeco LLC v. ION Geophysical Corp., 585 U.S. 407 (2018).

Design patents follow a different infringement test. Rather than element-by-element claim matching, a design patent is infringed if an ordinary observer, familiar with the prior art, would find the accused design substantially the same as the patented design. Egyptian Goddess, Inc. v. Swisa, Inc., 543 F.3d 665 (Fed. Cir. 2008) (en banc). Design-patent cases carry their own damages rule and their own recent obviousness upheaval, treated in the design-patent litigation guide, the design patent toolkit, and the analysis of design-patent obviousness after LKQ v. GM. For the anatomy of the § 271 analysis as a standalone reference, see what counts as patent infringement and the practical guide to patent infringement claims and defenses.

Worked example (hypothetical). Because the court narrowed "substantially continuous," Northwind cannot show literal infringement, so it presses the doctrine of equivalents: the perforated barrier, it argues, blocks lithium migration in substantially the same way to substantially the same result. The defendant counters with Festo — during prosecution Northwind added "continuous" to overcome the Japanese reference, and that amendment presumptively surrendered non-continuous barriers. The equivalents theory now depends on whether Northwind can rebut the Festo presumption, a fact-intensive fight that will headline the Markman-to-summary-judgment stretch.

Anchor authorities: 35 U.S.C. § 271; Warner-Jenkinson Co. v. Hilton Davis Chem. Co., 520 U.S. 17 (1997); Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722 (2002); Global-Tech Appliances, Inc. v. SEB S.A., 563 U.S. 754 (2011); Commil USA, LLC v. Cisco Sys., Inc., 575 U.S. 632 (2015); Limelight Networks, Inc. v. Akamai Techs., Inc., 572 U.S. 915 (2014); Impression Prods., Inc. v. Lexmark Int'l, Inc., 581 U.S. 360 (2017); WesternGeco LLC v. ION Geophysical Corp., 585 U.S. 407 (2018); Egyptian Goddess, Inc. v. Swisa, Inc., 543 F.3d 665 (Fed. Cir. 2008).

Stage 7 — Attacking validity: district court and the PTAB

Validity is fought on two fronts under two different burdens, and the mismatch drives much of modern patent strategy.

In district court, an issued patent is presumed valid, and a challenger must prove invalidity by clear and convincing evidence. 35 U.S.C. § 282; Microsoft Corp. v. i4i Ltd. Partnership, 564 U.S. 91 (2011). The substantive grounds are familiar: anticipation under § 102 (a single prior-art reference disclosing every limitation); obviousness under § 103, analyzed through the Graham factual inquiries and the flexible, common-sense approach of KSR, tempered by objective indicia such as commercial success and long-felt need; and the disclosure requirements of § 112 — enablement, written description, and definiteness. The Supreme Court recently reaffirmed the demands of enablement, holding that a patent claiming an entire genus must enable a skilled artisan to make and use the full scope of the claimed invention. Amgen Inc. v. Sanofi, 598 U.S. 594 (2023). Graham v. John Deere Co., 383 U.S. 1 (1966); KSR International Co. v. Teleflex Inc., 550 U.S. 398 (2007). The obviousness analysis, which recurs in prosecution and litigation alike, is treated in depth in the comprehensive guide to obviousness under Section 103.

At the PTAB, an accused infringer can attack the same patent through inter partes review (IPR), and the burden is dramatically lighter: unpatentability need only be shown by a preponderance of the evidence. 35 U.S.C. §§ 311–319. That gap — clear-and-convincing in court, preponderance at the Board — is why defendants so often prefer to litigate validity administratively. IPR is limited to anticipation and obviousness grounds based on patents and printed publications, so it cannot reach § 101 or § 112 challenges, but within its lane it is formidable. The Supreme Court has twice blessed and shaped the mechanism: Oil States Energy Services, LLC v. Greene's Energy Group, LLC upheld IPR against Article III and Seventh Amendment challenges, 584 U.S. 325 (2018); and SAS Institute Inc. v. Iancu held that if the Board institutes review, it must decide the patentability of every claim the petitioner challenged, not a subset, 584 U.S. 357 (2018). The decision whether to institute is itself largely unreviewable. Cuozzo Speed Technologies, LLC v. Lee, 579 U.S. 261 (2016).

Coordinating the two tracks is a discipline of its own. A petitioner must file its IPR within one year of being served with the infringement complaint, 35 U.S.C. § 315(b), and the Board may exercise discretion to deny institution when a parallel district-court trial is set to conclude first, under the multi-factor Fintiv analysis. Apple Inc. v. Fintiv, Inc., No. IPR2020-00019 (P.T.A.B. Mar. 20, 2020) (precedential). Petitioners increasingly neutralize Fintiv by offering a broad estoppel stipulation not to relitigate in court the grounds raised or that could have been raised at the Board. Sotera Wireless, Inc. v. Masimo Corp., No. IPR2020-01019 (P.T.A.B. Dec. 1, 2020) (precedential).

The teeth of that stipulation come from statutory estoppel. Once the Board issues a final written decision, § 315(e) bars the petitioner from asserting in district court any ground it raised or reasonably could have raised in the IPR. The Federal Circuit read that language broadly in California Institute of Technology v. Broadcom Ltd., overruling earlier authority and holding that estoppel extends to all grounds a skilled searcher diligently could have found, not merely those actually presented. 25 F.4th 976 (Fed. Cir. 2022). A defendant who loses an IPR therefore loses far more than the specific art it cited; it loses the whole family of prior-art invalidity theories it could have brought. Because so much rides on sequencing, see the practical treatment in the patent-focused field guide and the survey of recent doctrinal shifts in pivotal patent decisions and their strategy fallout.

A separate route to unenforceability is inequitable conduct — proof that the patentee, with intent to deceive, withheld or misrepresented information material to patentability during prosecution. The modern standard, requiring but-for materiality and specific intent, is demanding by design, and pleading it requires Rule 9(b) particularity. The offensive and defensive mechanics live in inequitable conduct and the duty of candor and proving inequitable conduct in patent prosecution.

Worked example (hypothetical). The defendant files an IPR on Northwind's anode patent, citing the Japanese reference plus a U.S. patent it found later, and moves to stay the Delaware case. To beat a Fintiv denial, it offers a Sotera stipulation. The Board institutes, and eighteen months later upholds the claims: the prior art does not teach the specific layering sequence. The victory is expensive for the defendant, because § 315(e) and California Institute of Technology now estop it from running any obviousness theory in Delaware that it reasonably could have raised at the Board. Its validity case in court shrinks to the § 101 and § 112 grounds the PTAB could not hear.

Anchor authorities: 35 U.S.C. §§ 102, 103, 112, 282, 311–319, 315(e); Microsoft Corp. v. i4i Ltd. P'ship, 564 U.S. 91 (2011); Graham v. John Deere Co., 383 U.S. 1 (1966); KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007); Amgen Inc. v. Sanofi, 598 U.S. 594 (2023); Oil States Energy Servs., LLC v. Greene's Energy Grp., LLC, 584 U.S. 325 (2018); SAS Inst. Inc. v. Iancu, 584 U.S. 357 (2018); Cal. Inst. of Tech. v. Broadcom Ltd., 25 F.4th 976 (Fed. Cir. 2022).

Stage 8 — What it is worth: damages, royalties, and enhancement

A prevailing patentee is entitled to damages "adequate to compensate for the infringement, but in no event less than a reasonable royalty," together with interest and costs. 35 U.S.C. § 284. Two compensatory theories dominate.

Lost profits aim to put the patentee where it would have been absent infringement. The classic test comes from Panduit Corp. v. Stahlin Brothers Fibre Works, Inc.: the patentee must show (1) demand for the patented product, (2) the absence of acceptable non-infringing substitutes, (3) manufacturing and marketing capacity to meet the demand, and (4) the profit it would have made. 575 F.2d 1152 (6th Cir. 1978). Lost profits are available only to a patentee that actually competes in the market; a non-practicing entity is confined to a royalty.

Reasonable royalty is the floor and the fallback. Courts most often build it through the hypothetical negotiation the parties would have conducted at the moment infringement began, evaluated against the fifteen Georgia-Pacific factors. 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). The royalty is the product of a rate and a base, and the modern law of damages is largely the law of getting that base right.

Apportionment is the discipline that keeps royalties honest. Damages must be tied to the incremental value of the patented feature, not the entire product, unless the patented feature drives consumer demand for the whole under the narrow entire market value rule. The Federal Circuit has repeatedly enforced this limit: a patentee is entitled only to damages attributable to the infringing features, VirnetX, Inc. v. Cisco Systems, Inc., 767 F.3d 1308 (Fed. Cir. 2014); damages theories must be tied to the facts and cannot rest on arbitrary rules of thumb, Commonwealth Scientific & Industrial Research Organisation v. Cisco Systems, Inc., 809 F.3d 1295 (Fed. Cir. 2015); and in multi-component products, the royalty base should generally begin from the smallest salable patent-practicing unit, LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51 (Fed. Cir. 2012). For standard-essential patents, apportionment folds in the further requirement that the royalty reflect only the value of the invention, not the value the standard itself conferred, and must respect any FRAND commitment. Ericsson, Inc. v. D-Link Systems, Inc., 773 F.3d 1201 (Fed. Cir. 2014). The economics of valuing a patent — reasonable-royalty construction, comparable licenses, and the role of the damages expert — are developed in the guide to what an IP case is worth and, from the licensing side, in monetizing a patent through licensing.

Marking can quietly erase years of damages. Under 35 U.S.C. § 287, a patentee that sells patented articles (or licenses others to) generally cannot recover damages for the period before it either marked the products with the patent number — physically or virtually — or gave the infringer actual notice. The marking duty runs to licensees too, and a patentee bears the burden to show compliance once the accused infringer identifies unmarked products. Arctic Cat Inc. v. Bombardier Recreational Products Inc., 950 F.3d 860 (Fed. Cir. 2020). See the patent-marking guide and the constructive-notice checklist.

Enhancement and fees sit at the back end. For willful infringement, the court may increase damages up to treble under § 284; Halo Electronics replaced a rigid two-part test with a discretionary standard keyed to the infringer's subjective culpability at the time of the conduct, so that egregious cases can draw enhancement while good-faith cases do not. 579 U.S. 93 (2016). In exceptional cases the court may shift attorney's fees under § 285, and Octane Fitness, LLC v. ICON Health & Fitness, Inc. loosened that standard to reach cases that simply stand out from others in the strength of a party's position or the manner of litigation. 572 U.S. 545 (2014). One caution on timing: laches is no longer a defense to damages within the six-year limitations window of § 286. SCA Hygiene Products Aktiebolag v. First Quality Baby Products, LLC, 580 U.S. 328 (2017).

Design patents carry a uniquely potent remedy. Section 289 lets the holder recover the infringer's total profit on the article of manufacture bearing the infringing design — with no apportionment to the design's contribution — though the Supreme Court held that the relevant "article of manufacture" may be a component rather than the entire end product. Samsung Electronics Co. v. Apple Inc., 580 U.S. 53 (2017). Finally, remember the tail: prejudgment interest on the award under § 284 and a post-trial accounting for supplemental damages covering infringement between the verdict and final judgment. See SynQor, Inc. v. Artesyn Technologies, Inc., 709 F.3d 1365 (Fed. Cir. 2013).

Anchor authorities: 35 U.S.C. §§ 284, 285, 286, 287, 289; Panduit Corp. v. Stahlin Bros. Fibre Works, Inc., 575 F.2d 1152 (6th Cir. 1978); Georgia-Pacific Corp. v. U.S. Plywood Corp., 318 F. Supp. 1116 (S.D.N.Y. 1970); VirnetX, Inc. v. Cisco Sys., Inc., 767 F.3d 1308 (Fed. Cir. 2014); LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51 (Fed. Cir. 2012); Halo Elecs., Inc. v. Pulse Elecs., Inc., 579 U.S. 93 (2016); Octane Fitness, LLC v. ICON Health & Fitness, Inc., 572 U.S. 545 (2014); Samsung Elecs. Co. v. Apple Inc., 580 U.S. 53 (2017).

Stage 9 — Trial: making the invisible legible

If a case survives summary judgment and does not settle, it goes to trial — usually before a jury. The division of labor is distinctive. The judge owns claim construction and the equitable issues (inequitable conduct, equitable estoppel, and the injunction analysis under eBay), while the jury decides the fact-laden merits: infringement, the factual predicates of invalidity, willfulness, and the amount of damages. The Seventh Amendment secures the right to a jury on the legal issues, and a party must demand it under Rule 38(b) or waive it.

The craft of a patent trial is translation. A jury will not reward the side with the better patent; it will reward the side that makes a genuinely difficult technology feel understandable and its story feel fair. That is won in the pretrial motions as much as on the stand. Daubert motions under Rule 702 test whether each expert's methodology is reliable enough to reach the jury — a damages expert who ignores apportionment, or an infringement expert who papers over a missing limitation, is a target. Motions in limine and Rule 403 balancing keep out evidence whose unfair prejudice outweighs its probative value, a recurring battleground over copying narratives, the parties' relative size, and prior litigation. See the guide to Rule 403 and unfair prejudice. Proposed jury instructions and a verdict form tailored to the asserted claims are not clerical work; a well-built special verdict form can preserve issues for appeal and expose an inconsistent verdict.

Trial motions run alongside the evidence. A party may move for judgment as a matter of law under Rule 50(a) once the other side has been fully heard on an issue, and renew it post-verdict under Rule 50(b); a motion for a new trial lies under Rule 59(a). Willfulness is typically decided by the jury as a factual matter, but the decision whether to enhance damages, and by how much, stays with the judge under Halo. Keep the trial resources close: the general federal deposition toolkit feeds impeachment at trial, and the comprehensive federal litigation filings guide maps the post-trial motion practice.

Anchor authorities: Fed. R. Civ. P. 38, 50, 59; Fed. R. Evid. 403, 702; U.S. Const. amend. VII; 35 U.S.C. § 284.

Stage 10 — The appeal: the Federal Circuit and beyond

Almost every patent appeal lands in one place. The Federal Circuit holds nationwide, exclusive jurisdiction over appeals from final district-court decisions in cases with a patent claim or compulsory patent counterclaim, and over appeals from the PTAB. 28 U.S.C. § 1295(a)(1). The centralization was Congress's deliberate cure for the forum-shopping and doctrinal chaos of the pre-1982 era, and the practical effect is that Federal Circuit precedent, not the law of the regional circuit where the case was tried, supplies the substantive patent rules the district court had to follow. The court's specialized expertise is a recurring theme in the Supreme Court's own patent decisions. See Dickinson v. Zurko, 527 U.S. 150 (1999).

Timing and finality gate the appeal. With narrow exceptions — a denied or granted preliminary injunction may be appealed immediately, for instance — a party can appeal only from a final judgment that ends the litigation on the merits. See Amgen Inc. v. Amneal Pharmaceuticals LLC, 945 F.3d 1368 (Fed. Cir. 2020). The notice of appeal goes to the district court and must name the Federal Circuit as the destination.

The standards of review are where appeals are won and lost, and they vary by issue. The ultimate claim construction is reviewed de novo, but subsidiary fact findings underlying it are reviewed only for clear error after Teva. Jury findings on infringement, the factual predicates of invalidity, and damages are reviewed for substantial evidence — a deferential standard that makes a clean trial record precious. Discretionary rulings — evidentiary calls, the injunction decision, fee awards, enhancement — are reviewed for abuse of discretion. Appeals from the PTAB review legal conclusions de novo and factual findings for substantial evidence, but institution decisions are largely off-limits under § 314(d) and Cuozzo. A litigant unhappy with the Federal Circuit's answer may seek certiorari, and the Supreme Court has intervened with increasing regularity over the past two decades — nearly every landmark cited in this roadmap is a product of that oversight. Track the docket and opinions through the Federal Circuit and CourtListener, and verify that any case you rely on is still good law using the guide to finding, reading, and verifying judicial decisions.

Anchor authorities: 28 U.S.C. § 1295(a)(1); Teva Pharms. USA, Inc. v. Sandoz, Inc., 574 U.S. 318 (2015); Amgen Inc. v. Amneal Pharms. LLC, 945 F.3d 1368 (Fed. Cir. 2020); Dickinson v. Zurko, 527 U.S. 150 (1999).

Stage 11 — The global case: parallel wars and FRAND

Patents are territorial. A U.S. patent stops at the U.S. border, so a genuinely global product dispute means parallel litigation in multiple countries, each with its own courts, remedies, timelines, and appetite for injunctions. The strategic questions multiply: where to sue first, how the foreign judgments interact, and how to keep the cases from tripping over one another.

Two features of the current environment reward attention. First, Germany and the Unified Patent Court (UPC) move quickly and grant injunctions comparatively readily; Germany's traditional bifurcation of infringement and validity into separate proceedings can leave a defendant enjoined before its invalidity defense is heard, and the UPC now offers a single forum with effect across most of the EU. That speed makes Europe a favored opening front for a patentee seeking leverage. Second, standard-essential patents (SEPs) encumbered by FRAND (fair, reasonable, and non-discriminatory) licensing commitments generate the largest and most globally coordinated fights in technology, spilling into rate-setting trials, antitrust theories, and dueling anti-suit injunctions as courts in different countries jockey to be the one that sets a worldwide royalty. These dynamics are developed in the SEP and FRAND licensing guide for the 5G and IoT era and illustrated through marquee disputes in landmark patent wars.

The mechanics of running a cross-border case — serving foreign defendants, taking evidence abroad, and enforcing or resisting foreign judgments — have their own playbook. Start with the toolkit for litigating IP across borders, and when service on a foreign defendant is required, work the Hague Convention service checklist.

Anchor authorities: Patent Cooperation Treaty; Agreement on a Unified Patent Court; Hague Service Convention. External resources: WIPO, the Unified Patent Court, and the EPO.

A case through the whole machine

Pull the stages together with one more hypothetical, followed end to end.

Worked illustration (hypothetical). Cortado Systems holds a patent on a pressure-sensing valve used in commercial espresso machines. It suspects that Riverbend Appliance, a competitor, copied the valve. Stage 1: Cortado charts each claim limitation against a Riverbend teardown, confirms clean title, and, mindful of declaratory-judgment risk, sends only a narrow notice letter to start the marking clock. Stage 2: Riverbend is incorporated in Delaware and imports its machines, so Cortado files in Delaware and, because of the imports, opens a parallel ITC complaint aimed at an exclusion order. Stage 3: Riverbend answers with non-infringement and invalidity counterclaims, moves to dismiss on § 101 (denied — the valve is plainly not an abstract idea), files an IPR, and moves to stay, offering a Sotera stipulation to head off a Fintiv denial. Stage 4: At the Markman hearing the court construes "responsive to a pressure differential" narrowly, reshaping Cortado's infringement theory. Stage 5: Discovery produces Riverbend's valve source code under a source-code protective order and its cost data for the damages case. Stage 6: Because the construction defeats literal infringement, Cortado runs the doctrine of equivalents and fends off a Festo estoppel argument. Stage 7: The PTAB upholds the claims; § 315(e) and California Institute of Technology now bar Riverbend from re-running its prior-art theories in court, and its validity case narrows. Stage 8: The jury awards a reasonable royalty apportioned to the valve rather than the whole machine, and finds willfulness, teeing up enhancement for the judge under Halo. Stage 9: Post-trial, the court denies Riverbend's Rule 50(b) motion and modestly enhances the award. Stage 10: The Federal Circuit affirms the construction de novo and the damages verdict on substantial evidence. Stage 11: Meanwhile, parallel suits over the same valve family proceed in the UPC and Germany. That is the machine, start to finish.

How to work a patent problem in this library

A research method beats a reading list. When a patent matter lands on your desk, orient with this toolkit, then descend to the stage-specific resource: the Section 271 anatomy for the infringement read, the obviousness guide and Section 101 checklist for validity, and the IP-damages guide for exposure. For background on how patents come to exist at all — the disclosure bargain, prosecution, and the statutory requirements you will later attack or defend — see how U.S. patents work and the prosecution roadmap, which is the mirror image of this litigation map. Then pull the analogous filings from PACER and CourtListener, read the assigned judge's standing orders and the district's local patent rules, and pressure-test your theory against the parallel PTAB record before you commit a client's money to any of it.

Related UpLaw Resources

Related Guides

Related Checklists

Related Toolkits

Related Articles

This toolkit is general legal information, not legal advice, and it creates no attorney-client relationship. Patent statutes, local patent rules, PTAB practice, and Federal Circuit and Supreme Court precedent change; verify current authority and consult qualified patent litigation counsel before acting on any matter.