Patent Venue After TC Heartland: Regular and Established Places of Business, Foreign Defendants, and Transfer

Douglas Harless sold supercomputers from his house in the Eastern District of Texas for about seven years. His employer's own sales map put him there. His price quotes listed his home phone, with a local area code, as his office line. The systems tied to his work brought in more than $345 million. When Raytheon sued his employer for patent infringement in that district, the Federal Circuit held that the house was not a place of business of the company, and it ordered the case transferred out of the Eastern District. In re Cray Inc., 871 F.3d 1355 (Fed. Cir. 2017).

Five years later, a different employee of a different company worked from a home in the Western District of Texas. He had two oscilloscopes, a soldering iron, a logic analyzer, and a pile of demonstration boards. The district court held that venue was proper, and the Federal Circuit refused to step in. A dissenting judge warned that the idea of an employer ratifying an employee's home "was not meant to be a leaky sieve." In re Monolithic Power Systems, Inc., 50 F.4th 157 (Fed. Cir. 2022).

So $345 million in sales did not make a place of business, and a soldering iron survived review. That contrast is a fair introduction to patent venue. The statute is one sentence long, it dates in substance from 1897, and it has to answer questions about remote work, content delivery networks, and Taiwanese parent companies. Damages exposure, trial speed, and the odds of a stay can all turn on which courthouse hears the case, so every word of that sentence has been litigated.

The short answer

• A domestic corporation can be sued for patent infringement in only two kinds of places. One is the single district where it "resides," which means its state of incorporation and, in a multi-district state, the district of its principal place of business or else its registered office. The other is a district where it has both committed acts of infringement and has a regular and established place of business. 28 U.S.C. § 1400(b); TC Heartland LLC v. Kraft Foods Group Brands LLC, 581 U.S. 258 (2017); In re BigCommerce, Inc., 890 F.3d 978 (Fed. Cir. 2018).
• Sales into a district are not enough. Neither is personal jurisdiction. The place must be physical, regular and established, and the defendant's own, and an employee or agent of the defendant must regularly conduct the defendant's business there.
• The plaintiff carries the burden. It must plead facts, and once venue is challenged it must prove them. In re ZTE (USA) Inc., 890 F.3d 1008 (Fed. Cir. 2018); Westech Aerosol Corp. v. 3M Co., 927 F.3d 1378 (Fed. Cir. 2019).
• Foreign corporations get no venue defense at all. They can be sued in any district where personal jurisdiction exists. In re HTC Corp., 889 F.3d 1349 (Fed. Cir. 2018). Their American subsidiaries get the full protection of § 1400(b), and that split shapes many complaints.
• A venue objection is easy to lose. Leave it out of a first Rule 12 motion or a first responsive pleading and it is waived. Sit on it tactically and a court may find it forfeited even without a Rule 12 waiver. In re Micron Technology, Inc., 875 F.3d 1091 (Fed. Cir. 2017).
• Proper venue does not end the fight. Section 1404(a) lets a court move a properly filed case for convenience, and the Federal Circuit reviews denials by mandamus with real frequency, as in In re Google LLC, No. 2026-157 (Fed. Cir. Oct. 6, 2026) (nonprecedential).
• Important questions remain open. They include whether performing one step of a method claim in a district counts as an act of infringement there, how far employee homes can go after Monolithic Power, whether a machine alone can ever be a place of business, and the correct timing rule for measuring venue.

Why patent cases have their own venue statute

Venue is the rule that says which federal district, among those with power over the parties, is the right one to hear a case. In most federal litigation venue is an afterthought, because the general statute, 28 U.S.C. § 1391, treats a corporate defendant as residing anywhere it is subject to personal jurisdiction. Patent law went its own way more than a century ago, and the reasons matter, because the Federal Circuit keeps returning to them.

The Judiciary Act of 1789 let a plaintiff sue where the defendant was an inhabitant or could be found. An 1887 amendment narrowed that to the defendant's home district, but In re Hohorst, 150 U.S. 653 (1893), arguably suggested the amendment did not reach patent cases, and lower courts split. Some let patent owners sue wherever a defendant could be served. Others required suit at home, which the sponsors of reform complained forced an inventor to chase an infringer across the continent even when its business was operating next door. Congress answered in 1897 with a patent-only statute that allowed suit where the defendant was an inhabitant or where it had committed acts of infringement and had a "regular and established place of business." One House sponsor said the point was to give jurisdiction where "a permanent agency transacting the business is located," and that isolated infringements would not do. Cray quotes that history at length.

The statute was therefore a compromise. It widened venue beyond the state of incorporation and narrowed it from the anywhere-you-can-serve-process rule. The Supreme Court has described it as "a restrictive measure, limiting a prior, broader venue," Stonite Products Co. v. Melvin Lloyd Co., 315 U.S. 561 (1942), and has said that "[t]he requirement of venue is specific and unambiguous; it is not one of those vague principles which, in the interest of some overriding policy, is to be given a 'liberal' construction." Schnell v. Peter Eckrich & Sons, Inc., 365 U.S. 260 (1961). Nearly every Federal Circuit venue decision cites one or both.

From Fourco to VE Holding to TC Heartland

Congress recodified the statute in 1948 as § 1400(b), swapping "resides" for "inhabitant." It also enacted a general venue statute defining a corporation's residence broadly "for venue purposes." In Fourco Glass Co. v. Transmirra Products Corp., 353 U.S. 222 (1957), the Supreme Court held that the general definition did not reach patent cases. Section 1400(b) was the "sole and exclusive" provision, and a corporation resided only in its state of incorporation.

That stood until 1988, when Congress rewrote § 1391(c) to say that "[f]or purposes of venue under this chapter" a corporation resided wherever it was subject to personal jurisdiction. In VE Holding Corp. v. Johnson Gas Appliance Co., 917 F.2d 1574 (Fed. Cir. 1990), the Federal Circuit treated those words as language of incorporation, because § 1400(b) sits in the same chapter. For the next 27 years the second half of § 1400(b) barely mattered. A company that sold a product nationwide "resided" nearly everywhere, and plaintiffs could pick from 94 districts. They increasingly picked the Eastern District of Texas, whose judges had built patent-specific rules and fast schedules, and whose Marshall courthouse became a fixture of patent litigation.

Congress amended § 1391 again in 2011, deleting "under this chapter" and adding a saving clause, "[e]xcept as otherwise provided by law." When Kraft sued TC Heartland in Delaware over flavored drink mixes, the defendant, organized and headquartered in Indiana with no meaningful local presence in Delaware beyond shipments into the state, argued that Fourco was still the law. The Federal Circuit said no in In re TC Heartland LLC, 821 F.3d 1338 (Fed. Cir. 2016). A unanimous Supreme Court, Justice Gorsuch not participating, reversed.

What TC Heartland decided and what it left alone

Justice Thomas's opinion is short and its logic is simple. Fourco definitively held that "resides" in § 1400(b) means the state of incorporation. Congress has never amended § 1400(b). When Congress means to change a settled reading through a related statute, it ordinarily says so clearly. The phrase "[f]or all venue purposes" is not materially different from the "for venue purposes" that Fourco already rejected, and the 2011 saving clause makes the case against displacement stronger. As for VE Holding, the Court noted that it rested almost entirely on the words Congress had deleted. The holding: "a domestic corporation 'resides' only in its State of incorporation for purposes of the patent venue statute."

Two footnotes define what the Court did not decide, and both still matter.

Unincorporated entities. The parties suggested in their briefs that TC Heartland was in fact an unincorporated entity. The Court confined its analysis to corporations and left that issue for remand. LLCs are everywhere in American business, and the Federal Circuit has since described LLC defendants as "incorporated" in Delaware, in Andra and in Comcast, without working through this reservation. Litigants commonly treat an LLC's state of organization as the equivalent of incorporation. A careful party should confirm how the forum's judges have handled the point and be ready to argue it.

Foreign corporations. The Court expressly declined to address the implications for foreign defendants or to express any opinion on Brunette Machine Works, Ltd. v. Kockum Industries, Inc., 406 U.S. 706 (1972). The Federal Circuit filled that gap a year later in HTC, discussed below.

Where a corporation "resides"

Delaware and other single-district states are easy. A Delaware corporation can always be sued for patent infringement in the District of Delaware, whatever else it does or does not do there. Multi-district states needed an answer, and BigCommerce supplied the test case. It was a Texas corporation headquartered in Austin, which sits in the Western District, and it was sued in the Eastern District. The district court reasoned that a corporation resides in every district of its home state.

The Federal Circuit disagreed, relying on the definite article in "the judicial district where the defendant resides," the contrast with "any district" in the 1897 text, and nineteenth-century decisions such as Shaw v. Quincy Mining Co., 145 U.S. 444 (1892). The rule from BigCommerce: a corporation resides only in the single district within its home state where it maintains its principal place of business, and failing that, the district of its registered office recorded in its corporate filings. Plaintiffs argued that modern businesses are too fluid for that kind of rule. The court called the argument "a non-starter" and sent the plaintiffs to Congress.

Two practical consequences follow. A Texas corporation headquartered in Dallas does not reside in Marshall or Waco. And a company incorporated in California, Texas, or New York has a single residence-based venue, not several, which should be checked against its filings before anyone drafts a complaint or an answer.

The second prong has two parts

If the defendant does not reside in the district, the plaintiff must show both that the defendant "has committed acts of infringement" there and that it "has a regular and established place of business" there. The statute joins them with "and," and Celgene Corp. v. Mylan Pharmaceuticals Inc., 17 F.4th 1111 (Fed. Cir. 2021), confirms the plaintiff must establish both. They do not need to be connected. The Federal Circuit has said that § 1400(b) requires no causal relationship between the place of business and the infringing acts. Monolithic Power so states in a footnote, citing its 2018 Google order. A company with a Texas sales office and an accused product built and sold entirely from Oregon can therefore satisfy both parts if the accused product is also sold in that Texas district.

Acts of infringement, and the one-step fight

For venue purposes, the plaintiff's allegation of infringement generally suffices, because "the issue of infringement is not reached on the merits in considering venue requirements." In re Cordis Corp., 769 F.2d 733 (Fed. Cir. 1985). Sales and offers in the district are straightforward. Inducement and contributory infringement raise their own location questions, which interact with the rules explained in Indirect and Divided Patent Infringement: Inducement, Contributory Liability, and the Akamai Problem.

Method claims are where the current fight lies. A method is infringed only when every step is performed, and NTP, Inc. v. Research In Motion, Ltd., 418 F.3d 1282 (Fed. Cir. 2005), held that a process is not used within the United States unless each step is performed here. Defendants say the same logic must apply to a district: if all steps must occur in the country for there to be infringement in the country, then all must occur in the district for there to be an act of infringement in the district. Software and network method claims make this more than academic, because the steps of a distributed method may run on a phone in Tyler, a load balancer in Dallas, and a database in Virginia.

Judge Gilstrap rejected the all-steps argument in SEVEN Networks, LLC v. Google LLC, 315 F. Supp. 3d 933 (E.D. Tex. 2018), holding that an allegation that the defendant practiced at least one step in the district could support venue. Western District of Texas decisions followed. Another Eastern District judge went the other way in AML IP, LLC v. Bath & Body Works Direct, Inc., No. 4:22-cv-216-SDJ, 2024 WL 3825242 (E.D. Tex. Aug. 13, 2024), applying NTP to the district level.

Comcast tried to force the question by petitioning for mandamus from a Marshall ruling in a case brought by Sandpiper CDN on five method patents. The Federal Circuit denied the petition in In re Comcast Cable Communications, LLC, No. 2026-104 (Fed. Cir. Dec. 9, 2025) (nonprecedential). It did not touch the merits. It held that Comcast had not shown a post-judgment appeal would be inadequate and that waiting would "allow the issue to percolate." The full court then denied rehearing en banc on February 25, 2026. So the question remains open, the split remains live, and the answer will likely come from an appeal after final judgment. Any defendant raising the argument should preserve it clearly, and any plaintiff relying on one-step allegations should plead the whole method as far as the facts allow, because a judgment resting on an unresolved venue theory can be vacated on appeal.

Cray's three requirements, read closely

Cray arose in the wave of motions that followed TC Heartland. Judge Gilstrap had denied transfer and, for the benefit of other litigants, proposed a four-factor test built around physical presence, the defendant's representations, benefits received, and targeted interactions with the district. He also said a fixed physical location was not a prerequisite. Judge Lourie's opinion granted mandamus and replaced that framework with one tied to the statute's words:

"(1) there must be a physical place in the district; (2) it must be a regular and established place of business; and (3) it must be the place of the defendant."

If any one fails, venue is improper. Cray also insists that these are aids to reading the statute and do not replace it, and that "no one fact is controlling." Federal Circuit law governs, because § 1400(b) is unique to patents.

A physical place. The statute cannot refer to "a virtual space" or to "electronic communications from one person to another." A website does not create venue anywhere. The place need not be a formal office or store, but it must be a physical, geographic location from which the defendant's business is carried out.

Regular and established. "Regular" excludes sporadic activity. "Established" requires permanence: a semiannual trade show display does not qualify, and five continuous years did in an older case Cray cites. A business can move, but for a meaningful time it must be stable. Here Cray adds a detail that has aged into the key to remote-work cases: if an employee can move out of the district at his or her own instigation without the defendant's approval, that cuts against treating the home as the defendant's place.

Of the defendant. This is where most fights land. "Employees change jobs," Cray observes, so the place must belong to the company, and "the defendant must establish or ratify the place of business." Cray then lists considerations that have become the working checklist in every venue dispute.

Employees and home offices

Three decisions set the boundaries, and read together they tell a defendant and a plaintiff exactly what evidence to look for.

Cray is the anchor. Mr. Harless received reimbursement for phone, internet, and travel, and administrative support from Minnesota, and customers saw his local number. But Cray did not own, lease, or rent any part of his home, did not choose its location, stored no inventory or demonstrations there, and did not condition his job on living there. Nothing suggested the company cared where he lived. He and his colleague were "free to live where [they] chose." The court quoted the Fourth Circuit: venue lies where the defendant has a place of business, "not where the defendant's employee owns a home in which he carries on some of the work that he does for the defendant."

The older contrast is Cordis, which Cray took care to limit without overruling. There two sales representatives kept their employer's literature and tens of thousands of dollars of its products at home and took inventory directly to clients, and the company paid a Minnesota secretarial service that answered its phone in the company's name and appeared in the phone book. Cray characterized that as a business that depended on employees being physically present in the district and had affirmatively acted to make operations permanent there.

Celgene pressed Cray further. After two years of venue discovery in New Jersey, the plaintiff had found 17 employees living in the state out of a workforce in the tens of thousands, a roster, business cards with home addresses, LinkedIn profiles, a job posting, and two storage lockers rented by sales employees in their own names to hold samples. The Federal Circuit affirmed dismissal. It also doubted an "aggregate-place" theory that would add up scattered locations into a district-wide presence: the plaintiff cited no case endorsing it, and the homes and lockers lumped together still would not have been "of the defendant." That opinion is the best single resource for what does not suffice.

Then came Monolithic Power. The defendant, a Delaware corporation, had four full-time remote employees in the Western District of Texas. Judge Albright found that the company viewed a presence there as important, given a history of soliciting employees in Austin to support local manufacturing customers. He also relied on equipment the company had supplied to one engineer, Jason Bone, for validation testing at home for at least one in-district customer. The per curiam panel denied mandamus, describing the facts as possibly "idiosyncratic" and emphasizing that it was not reaching the merits. Judge Lourie dissented, arguing the case could not be distinguished from Cray and Celgene.

The practical lesson is narrower than some commentary suggests. Monolithic Power does not hold that four engineers with lab gear create venue. It holds that a defendant who loses that argument in the district court usually has to wait for an appeal after final judgment. Plaintiffs should still treat it as a road map, and defendants as a warning, because a venue ruling on these facts may stick for the life of the case.

Worked example: the remote sales and engineering team

Suppose Harrow Sensor Works, a Delaware corporation headquartered in Minneapolis, makes industrial vibration sensors. It has moved to a remote-first model. A competitor wants to sue it in the Eastern District of Texas, where Harrow has sold sensors to a Tyler refinery for years.

Variation one. A single account manager has lived in Plano since before she was hired. Harrow reimburses her cell phone and internet and lists no Plano location anywhere. She could move to Oklahoma tomorrow with a phone call. On those facts, Cray controls, and venue is improper.

Variation two. Harrow's job posting says the account manager "must reside in the Dallas/Fort Worth area," her offer letter says the same, and when she told Harrow she might move, a vice president said the role would need to be refilled locally. Harrow also ships her a pallet of demonstration units, which she delivers to customers from her garage. Now the Cray considerations are lining up. Harrow conditioned employment on residence and stores material to be distributed from the place. Discovery into that garage becomes dangerous for Harrow, though it would still argue the garage itself is not its place, since it has no possessory right over it.

Variation three. Harrow pays a monthly stipend expressly for "a dedicated home lab," provides oscilloscopes and test rigs, and assigns the engineer to a named customer in the district. It puts no location requirement in writing. This is the Monolithic Power zone. A district court could go either way, and Harrow should assume that an adverse ruling would not be corrected until after trial.

Variation four. Harrow takes a two-desk coworking membership in Plano in its own name, lists it on its contact page, and sends staff there three days a week. Physical place, regular, established, and of the defendant. A flexible lease does not change much here, since Google rejects any requirement of a traditional real property interest. Venue is likely proper if an act of infringement is alleged in the district, whatever the desk has to do with the accused product.

The human resources manager who wrote that job posting probably did not know she was making a venue decision. For the steps a company can take before suit, and for how each side should build this record, see the companion Choosing or Challenging Venue in a Patent Case: A Practical Guide.

Servers, racks, and the cloud

The biggest post-Cray question was whether equipment alone could be a place of business. To see why, it helps to know what the equipment does.

A content delivery network keeps copies of popular files close to the people who want them, because data crossing fewer network hops arrives faster and costs less to carry. Google Global Cache servers are Google-owned machines that sit inside an internet service provider's data center. When a viewer in, say, Tyler requests a popular video, the provider's routers try to serve it from the cache in its own network before going back to Google's central facilities. The provider supplies rack space (a slot in a standard equipment cabinet), power, network ports, and IP addresses.

Google had such servers in two Eastern District of Texas facilities, hosted by Cable One and Suddenlink. Judge Gilstrap held in SEVEN Networks that they were a regular and established place of business, while another Eastern District decision, Personal Audio, LLC v. Google, Inc., 280 F. Supp. 3d 922 (E.D. Tex. 2017), had said the opposite. The Federal Circuit first declined to intervene in 2018 and let the issue percolate, over a dissent from denial of rehearing en banc in In re Google LLC, 914 F.3d 1377 (Fed. Cir. 2019). By 2020 the conflict had ripened, and it granted the writ. In re Google LLC, 949 F.3d 1338 (Fed. Cir. 2020).

The decision has two halves, and defendants often quote only the second.

Rack space can be a "place." Google argued that a place requires ownership or a leasehold in real property. The court rejected that. A defendant "who operates a table at a flea market may have established a place of business," and leased shelf space or rack space can qualify, as district courts had found. A fixed geographic location that the defendant possesses or controls is enough for the first requirement.

But a place of business needs a person. The court then looked to 28 U.S.C. § 1694, the patent service statute. It began life as the second sentence of the same 1897 section and allows service on the "agent" conducting business at the regular and established place. Reading the two together, along with the congressional reference to a "permanent agency," the court held that the phrase "requires the regular, physical presence of an employee or other agent of the defendant conducting the defendant's business at the alleged 'place of business.'"

The internet service providers did not fit. Agency requires a right of interim control, and Google had none over how network access was provided. Installation was a one-time event for each server. Maintenance, such as flipping a switch or replacing a card under step-by-step instructions, may have suggested agency but was only ancillary to the business. The court drew the line at "production, storage, transport, and exchange of goods or services."

Part IV of the order is easy to overlook. The court said it was not deciding whether a regular and established place of business will always require a human, or whether "a machine could be an 'agent.'" It noted that such a theory would have to reckon with service on a machine under § 1694, and it left unaddressed what might be inferred from section 18(c) of the America Invents Act, which says an automated teller machine is not a regular and established place of business for suits on covered business method patents. Judge Wallach's concurrence raised the further possibility that end users who generate data Google monetizes might be its agents. Those remain unanswered. Autonomous kiosks, robotic warehouses, and unattended retail lockers are the natural test cases.

Worked example: a streaming company's edge network

Suppose Thistledown Streaming, a Delaware company headquartered in San Jose, contracts for rack space in a colocation facility in Waco. Colocation means a landlord provides space, power, and cooling while the tenant owns the machines. It also runs its application on a major cloud provider, whose nearest region happens to be in another district.

The cloud deployment is the easy part. Thistledown rents computing capacity that the provider can place and move as it wishes, and Thistledown has no possession of any identifiable spot. Whatever else might be said, virtual machines look like Cray's "virtual space," and the provider's staff are not Thistledown's agents.

The Waco rack is closer to Google. It is a place. Standing alone, without an employee or agent regularly there, it is not a place of business. Three changes could alter the analysis. If Thistledown stations its own technician at the facility every weekday, the regular-presence requirement looks met. If it hires the landlord's "remote hands" staff to perform repairs, Google says that is ancillary maintenance. If Thistledown rents a locked cage where its staff also stage and ship set-top boxes to customers, it has moved into storage and transport of goods, which Google lists as conducting the business.

The upshot for infrastructure planners is that venue exposure now follows people more than machines, though a company should not assume that will always remain so.

Third parties and corporate affiliates

Google makes agency law central, and three further cases apply it. The plaintiff, as the party asserting agency, has the burden.

Outsourced call centers. American GNC sued ZTE USA in the Eastern District, relying on a Plano call center operated by a contractor with more than sixty representatives dedicated to ZTE and at least two ZTE supervisors on site. The magistrate judge put the burden on the defendant, and the district judge treated the center as established in partnership with the contractor. ZTE vacated. An arm's-length service contract does not automatically make a vendor's site the defendant's place, and the district court had to ask who owned or leased the space and equipment, what the signage said, whether ZTE chose the location, and whether the vendor needed permission to move or stop.

Franchised dealerships. StratosAudio sued Volkswagen and Hyundai distributors in Waco, relying on independent dealerships. The franchise agreements imposed staffing, inventory, training, signage, and tool requirements, and warranty work was reimbursed. Judge Albright found agency. The Federal Circuit granted mandamus in In re Volkswagen Group of America, Inc., 28 F.4th 1203 (Fed. Cir. 2022), because those provisions set constraints but gave no right of interim control over how cars were sold or warranty repairs performed. Contract disclaimers of agency were relevant but not dispositive. The court drew on franchise cases from tort law, relying on the Restatement (Third) of Agency's distinction between setting service standards and giving interim instructions. It did not reach Texas's statute barring distributors from operating dealerships.

Corporate family members. In Andra Group, LP v. Victoria's Secret Stores, LLC, 6 F.4th 1283 (Fed. Cir. 2021), one affiliate operated stores in the Eastern District, including one in Plano, while the parent, the website operator, and the brand company had no physical presence there. The accused technology was on the website. The Federal Circuit affirmed dismissal of the non-store defendants. A store manager's testimony that she hired and fired on her own undercut the agency theory, and accepting web returns in stores was one discrete task. On ratification, it set a threshold: where related companies maintain corporate separateness, one's place of business is not imputed to another, citing Cannon Manufacturing Co. v. Cudahy Packing Co., 267 U.S. 333 (1925). Shared branding and a "Find a Store" feature did not change that. Celgene added that disregarding the corporate form is a regional-circuit question and a demanding one, and that shared marketing and overlapping officers showed, in the court's phrase, "collaboration, not commonality."

These cases give corporate structure real consequences. A plaintiff who wants the district where the retail arm operates must show that the retail arm itself commits the alleged infringement, which is often hard when the accused feature lives on a website run by a sister company.

Burden, pleading, and timing

Burden. ZTE holds, as a matter of Federal Circuit law, that on a defendant's motion the plaintiff bears the burden of establishing proper venue under § 1400(b). The court relied on the statute's restrictive purpose and on the consistent practice of regional circuits before 1982. On this point patent cases differ from some regional practice, including the disagreement noted in Wright and Miller, which ZTE cites.

Pleading. Westech rejects the bare recital. Alleging that the defendant "has one or more regular and established places of business" in the district, with sales representatives and revenue on top, was a legal conclusion the court need not accept. The plaintiff must plead facts that show a place, consistent with the plausibility cases Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007), and Ashcroft v. Iqbal, 556 U.S. 662 (2009). For general technique, see Drafting a Complaint That Survives a Motion to Dismiss.

Standard of review. The Federal Circuit reviews the question of proper venue under § 1400(b) de novo on appeal, as Westech says. Mandamus is another matter, because the petitioner must show a clear and indisputable right to relief.

Timing. Must the place exist when the claim accrued, when the suit was filed, or both? Google noted in a footnote that the regional circuits "appear to be split," comparing an Eleventh Circuit decision looking to filing with a Seventh Circuit decision asking whether a place existed when the claim accrued and suit followed within a reasonable time. It did not need to decide, because Google's servers were present at both times. Closing an office shortly before suit is therefore no guarantee, and closing one shortly after suit should be presumed useless.

Venue discovery. "[W]here issues arise as to jurisdiction or venue, discovery is available to ascertain the facts bearing on such issues," the Supreme Court said in Oppenheimer Fund, Inc. v. Sanders, 437 U.S. 340 (1978), a note the Annotated Patent Digest places at the head of its discussion of § 1400(b) discovery. Courts vary in how much they allow. Celgene involved two years of it, which illustrates both the value of discovery to plaintiffs and the reason defendants should seek limits.

Waiver and forfeiture

Under Fed. R. Civ. P. 12(h)(1), an improper venue defense is waived if it is omitted from a first Rule 12 motion when it was available, or if it is left out of both any Rule 12 motion and the responsive pleading. The Supreme Court's older statement of the principle is that venue is a privilege, and "[i]t may be lost by failure to assert it seasonably, by formal submission in a cause, or by submission through conduct." Neirbo Co. v. Bethlehem Shipbuilding Corp., 308 U.S. 165 (1939).

TC Heartland created a question nobody had faced: what about defendants who had answered or moved before May 22, 2017, when VE Holding made any objection futile? Micron, which arose from Harvard's suit in Massachusetts, supplied a two-part answer. The defense was not "available" under Rule 12(g)(2) before the decision, because binding precedent barred a district court from granting it, so there was no Rule 12 waiver. But Rule 12 is not the only route to losing venue. Under Dietz v. Bouldin, Inc., 579 U.S. 40 (2016), and 28 U.S.C. § 1406(b), courts have inherent power to find forfeiture for untimeliness or consent, to be used with caution. The court flagged "a defendant's tactical wait-and-see bypassing of an opportunity to declare a desire for a different forum" as an obvious starting point and noted that it had denied mandamus where objections came close to trial.

In re Oath Holdings Inc., 908 F.3d 1301 (Fed. Cir. 2018), then held that Federal Circuit law governs waiver and forfeiture of venue under § 1400(b), and granted a writ against a district court that had refused to follow Micron. Oath's answer had admitted venue but reserved the right to challenge it based on any change in the law, and it moved within 21 days of the decision. The Federal Circuit found no sound basis for forfeiture on those facts. Compare BigCommerce, where a motion nine days after TC Heartland in a two-month-old case made forfeiture implausible.

The transition is over, so the modern rule is simple. A defendant should raise improper venue in its first Rule 12 motion or, if it does not move, in its answer, and then press the point. Litigating the merits while holding the objection in reserve invites a forfeiture finding. Other late-stage traps, such as filing a counterclaim or seeking relief on the merits first, are fact-specific, and a defendant should move for dismissal or transfer before asking the court for anything substantive. For deadline consequences generally, see Missed Federal Deadlines: Rule 6, Excusable Neglect, and the Deadlines a Court Cannot Excuse.

Foreign defendants: no venue defense at all

The alien-venue rule, as the Federal Circuit calls it, is one of the oldest in federal practice. The 1789 Act restricted suits only against "an inhabitant of the United States," so a noninhabitant could be sued in any district. Hohorst applied that understanding to a German steamship company under the 1887 Act. Brunette then applied it to patent cases, holding that a Canadian corporation could not use § 1400(b) as a shield, because "suits against aliens are wholly outside the operation of all the federal venue laws, general and special."

HTC Corporation of Taiwan argued that the 2011 amendments changed this. The new § 1391(c)(3) says "a defendant not resident in the United States may be sued in any judicial district," under a heading about residency, and § 1391(c)(2) defines residence for an "entity" without excluding foreign ones. If a foreign company resides somewhere under (c)(2), and § 1400(b) is exclusive, then perhaps it has a venue defense after all. The Federal Circuit rejected this in HTC. Congress made one clear change in 2011, giving permanent resident aliens who are natural persons a venue defense. It said nothing comparable about foreign corporations. And HTC's reading would open a venue gap, leaving some foreign infringers suable nowhere, which Brunette said Congress does not generally intend.

Three points follow in practice.

Personal jurisdiction becomes the only geographic limit. The foreign defendant may challenge jurisdiction, including through state-specific contacts analysis under cases like Ford Motor Co. v. Montana Eighth Judicial District Court, 592 U.S. 351 (2021), and Daimler AG v. Bauman, 571 U.S. 117 (2014). But Fed. R. Civ. P. 4(k)(2) serves as a federal long-arm statute for federal claims against a defendant not subject to jurisdiction in any state, measuring contacts with the United States as a whole. See Synthes (U.S.A.) v. G.M. Dos Reis Jr. Ind. Com. de Equip. Medico, 563 F.3d 1285 (Fed. Cir. 2009); M-I Drilling Fluids UK Ltd. v. Dynamic Air Ltda., 890 F.3d 995 (Fed. Cir. 2018). The topic is developed in Personal Jurisdiction Over Online and Foreign Defendants: From Zippo to Ford Motor.

Each defendant is analyzed separately. Andra requires venue to be shown as to each defendant, and § 1391(c)(3) says that joining a nonresident defendant is disregarded in deciding where the action may be brought against others. In HTC itself the district court found venue improper as to the American subsidiary and proper as to the Taiwanese parent, and the plaintiffs then dropped the subsidiary and stayed in Delaware against the parent.

Transfer remains available. A foreign company has no venue objection, but it can still seek § 1404(a) transfer. It must identify a district where the action "might have been brought" without its consent, and Hoffman v. Blaski, 363 U.S. 335 (1960), teaches that a defendant's after-the-fact willingness to be sued somewhere does not count. The 2011 amendment now also allows transfer to a district to which all parties have consented, which helps only when the plaintiff agrees.

Worked example: the Taiwanese parent and the Oregon subsidiary

Suppose Bluewater Imaging owns a patent on camera modules. Kestrelview Optics Co., a Taiwanese company, makes the modules and sells them to phone makers, and its wholly owned subsidiary, Kestrelview USA, Inc., an Oregon corporation in Portland, handles American sales from a single office. Bluewater wants a Texas jury.

Against Kestrelview USA, Texas is unavailable unless the subsidiary has a regular and established place there. It resides only in Oregon. Against the Taiwanese parent, every district is open as a venue matter, subject to personal jurisdiction, for which Bluewater will point to deliberate shipments through national distributors, and in the alternative to Rule 4(k)(2).

Bluewater therefore faces a choice. It can sue only the parent in Marshall and accept whatever difficulties come with discovery from and collection against a foreign company. Or it can sue both in Oregon, where the subsidiary is at home. Suing both in Texas would invite the subsidiary's dismissal or transfer under § 1406(a) and a likely severance fight, and the parent could argue that the case belongs with its subsidiary in Oregon under § 1404(a). The parent would need to show the case might have been brought in Oregon against it, which it can, since an alien may be sued in any district, subject to jurisdiction there.

A corporate group that sells through an American arm should expect exactly this analysis to be run against it, and should understand that the arm's protection does not extend upward.

Hatch-Waxman venue

Under the Hatch-Waxman Act, submitting an abbreviated new drug application (ANDA, the generic drug maker's filing that relies on the brand's safety data) seeking approval before a listed patent expires is itself an act of infringement under 35 U.S.C. § 271(e)(2). The Supreme Court has called this an artificial act. Before TC Heartland, it hardly mattered where it occurred, since a generic company was suable wherever personal jurisdiction lay, and Acorda Therapeutics Inc. v. Mylan Pharmaceuticals Inc., 817 F.3d 755 (Fed. Cir. 2016), had held that planned nationwide marketing supported specific jurisdiction. Brand companies assumed they could keep suing in Delaware and New Jersey.

Valeant Pharmaceuticals North America LLC v. Mylan Pharmaceuticals Inc., 978 F.3d 1374 (Fed. Cir. 2020), ended that. Section 1400(b) speaks in the past tense ("has committed"). So venue must rest on past acts, and for a domestic generic filer, those acts occur "only in districts where actions related to the submission of an Abbreviated New Drug Application" occur, not everywhere future sales are contemplated. The court did not catalog which preparatory acts count, and it noted without deciding the district court's suggestion that an act may occur in Maryland, where the FDA received the filing. As to a foreign Mylan affiliate, it reversed the venue dismissal, because foreign defendants can be sued anywhere.

Celgene tightened it further. Sending the required paragraph IV notice letter into New Jersey and receiving it at the patent owner's headquarters is not part of the act of infringement, because the statute and regulations treat the submission and the notice as distinct events, and the letter comes after the infringing submission. The notice affects the timing of the 45-day window for suit and the 30-month stay under 21 U.S.C. § 355(j), but it does not create the cause of action.

For brand companies the consequence is often protective suits, as Valeant itself filed in West Virginia, and a much closer look at where the ANDA was prepared and who submitted it. Hatch-Waxman Litigation: The Orange Book, Paragraph IV Certifications, and the 30-Month Stay and Running an ANDA Litigation: A Practical Guide cover the regulatory framework and case management. Whether Valeant carries over to biosimilar filings under the BPCIA has not been decided in the decisions reviewed here, and the statutory structure differs enough that a party should research the point directly.

Declaratory judgment actions run on different rails

An accused infringer who sues first for a declaration of noninfringement or invalidity is not bringing a "civil action for patent infringement," so § 1400(b) does not apply. VE Holding said it was long established that such an action, "the mirror image of a suit for patent infringement," is governed by the general venue statutes. That passage is among the parts of VE Holding that TC Heartland did not disturb on its own terms. Section 1391(b) therefore governs, and the plaintiff's real constraint is personal jurisdiction over the patent owner.

The Federal Circuit used to apply a nearly categorical rule that patent owners could not be haled into a forum on the basis of infringement letters alone, drawn from Red Wing Shoe Co. v. Hockerson-Halberstadt, Inc., 148 F.3d 1355 (Fed. Cir. 1998). It has since cut that back. Trimble Inc. v. PerDiemCo LLC, 997 F.3d 1147 (Fed. Cir. 2021), upheld jurisdiction over a patentee who exchanged 22 communications with the accused infringer, and Apple Inc. v. Zipit Wireless, Inc., 30 F.4th 1368 (Fed. Cir. 2022), reversed a dismissal, rejecting a reading of Red Wing as a bright-line rule. A foreign patentee with no designated agent can be reached in the Eastern District of Virginia under 35 U.S.C. § 293.

Competing suits raise the first-to-file rule, which generally favors the first case but yields to sound reasons of convenience or fairness. Genentech, Inc. v. Eli Lilly & Co., 998 F.2d 931 (Fed. Cir. 1993); Micron Technology, Inc. v. MOSAID Technologies, Inc., 518 F.3d 897 (Fed. Cir. 2008); Electronics for Imaging, Inc. v. Coyle, 394 F.3d 1341 (Fed. Cir. 2005). A full treatment is in Declaratory Judgment Actions: Jurisdiction, Discretion, and the Race to the Courthouse.

Multiple defendants and the customer problem

Venue must be proper for each defendant, and § 1400(b) has no provision that lets a court hear claims against one company because venue exists for another. Stonite is the old authority: two Pennsylvania corporations residing in different districts could not be joined in one because a general statute would have allowed it. Whether "pendent venue" can ever bring a patent claim along with a properly venued claim is treated with caution in the Annotated Patent Digest, and a party relying on it should find authority in its own district. The exclusive character of § 1400(b) makes it a weak foundation.

Joinder law compounds this. Since 2011, 35 U.S.C. § 299 bars joining unrelated accused infringers based only on allegations that each infringes the same patent, except in cases involving § 271(e)(2) acts. It requires the same transaction or series of transactions relating to the same accused product or process. Applying Rule 20 to a case filed before that statute took effect, In re EMC Corp., 677 F.3d 1351 (Fed. Cir. 2012), rejected a "not dramatically different" test for joining independent defendants under Rule 20.

Plaintiffs sometimes sue a manufacturer's customers because the customers are in the desired district. The customer-suit exception offers a counter: claims against mere resellers may be stayed in favor of the manufacturer's case. Katz v. Lear Siegler, Inc., 909 F.2d 1459 (Fed. Cir. 1990); In re Nintendo of America, Inc., 756 F.3d 1363 (Fed. Cir. 2014) (severing and staying retailer claims and transferring the rest).

Multidistrict centralization under 28 U.S.C. § 1407 can move pretrial proceedings but does not change the trial venue, which the transferee court cannot assign to itself. Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach, 523 U.S. 26 (1998).

Remedies when venue is wrong

Under 28 U.S.C. § 1406(a), the court "shall dismiss, or if it be in the interest of justice, transfer" to a district where the case could have been brought. The Federal Circuit's writs usually leave that choice to the district court, as in Cray, Google, and Volkswagen. Dismissal is typically without prejudice. The six-year damages lookback in patent cases means a dismissal can still cost a plaintiff a period of damages, so plaintiffs usually ask for transfer in the alternative and should say where.

Section 1404(a): a proper venue can still be the wrong place

Everything above concerns whether the plaintiff was allowed to file where it did. 28 U.S.C. § 1404(a) asks something different: "[f]or the convenience of parties and witnesses, in the interest of justice," should the case go elsewhere? As the Supreme Court said in Atlantic Marine Construction Co. v. United States District Court, 571 U.S. 49 (2013), the two are separate. Section 1406(a) and Rule 12(b)(3) apply only when venue is "wrong" or "improper," and a forum-selection clause does not make it so. Such a clause is enforced through § 1404(a). Because convenience is not patent-specific, regional circuit law governs. TS Tech says so, and ZTE contrasted it with § 1400(b) for this reason. Forum-selection and first-filed issues outside the patent setting are developed in Forum Selection, Transfer, and Parallel Proceedings: Choosing and Keeping the Court You Want.

The Fifth Circuit's framework and the road to Waco

Most of the law comes from Texas because most of the fights did. In In re Volkswagen of America, Inc., 545 F.3d 304 (5th Cir. 2008) (en banc), a family sued over a Dallas car crash in Marshall. The en banc court granted mandamus and held that the movant must show that the transferee venue is "clearly more convenient," a burden it described as reflecting the deference owed to a plaintiff's choice. It adopted four private and four public factors, along with the 100-mile rule from In re Volkswagen AG, 371 F.3d 201 (5th Cir. 2004), under which witness inconvenience grows with distance beyond 100 miles.

The Federal Circuit, applying Fifth Circuit law, promptly used that framework in patent cases, starting with In re TS Tech USA Corp., 551 F.3d 1315 (Fed. Cir. 2008). In re Genentech, Inc., 566 F.3d 1338 (Fed. Cir. 2009), rejected a requirement that the movant identify "key witnesses" and warned against mechanically applying the 100-mile rule to witnesses from Europe. Then came a run of cases aimed at manufactured ties to Texas: In re Hoffmann-La Roche Inc., 587 F.3d 1333 (Fed. Cir. 2009), where 75,000 pages of documents were converted and sent to a Texas lawyer's office before suit, In re Zimmer Holdings, Inc., 609 F.3d 1378 (Fed. Cir. 2010), and In re Microsoft Corp., 630 F.3d 1361 (Fed. Cir. 2011), where a United Kingdom-run company had an unstaffed office in Tyler. Presence that is "recent and ephemeral" counts for little.

After TC Heartland and Judge Albright's 2018 appointment, plaintiffs found a new home in Waco. Between 2020 and 2023 a remarkable line of mandamus orders followed.

• In re Apple Inc., 979 F.3d 1332 (Fed. Cir. 2020). Uniloc against Apple. Chief Judge Prost, over Judge Moore's dissent, held that a transfer motion "should unquestionably take top priority," faulted the district court for having "barreled ahead on the merits" with a Markman ruling while the motion was pending, and held that Apple's general presence in Austin, untethered to the suit, could not drive the local-interest factor.
• In re Samsung Electronics Co., 2 F.4th 1371 (Fed. Cir. 2021). A North Carolina patent owner formed a Texas LLC a month before suit and assigned it rights only for parts of Texas, then argued the case could not have been brought in California because the new entity's rights did not extend there. The court, saying it was "not bound by a plaintiff's efforts to manipulate venue," looked through the arrangement. The same order rejected the categorical discounting of party witnesses.
• In re Hulu, LLC, No. 2021-142, 2021 WL 3278194 (Fed. Cir. Aug. 2, 2021) (nonprecedential). Categorical rejection of prior-art witnesses was an abuse of discretion, and speed of a court's own scheduling order did not establish court congestion.
• In re Juniper Networks, Inc., 14 F.4th 1313 (Fed. Cir. 2021). Eleven party witnesses in California against one in the Western District, and a Waco office with two employees opened a few months before the complaints. The convenience of witnesses is "probably the single most important factor." The local interest factor looks to the events behind the suit. A court's aggressive scheduling order does not show congestion.
• In re Apple Inc., 52 F.4th 1360 (Fed. Cir. 2022). Requiring 30 more weeks of merits fact discovery and re-briefing before ruling on transfer was a clear abuse. The decision "must proceed expeditiously as the first order of business."
• In re Google LLC, 58 F.4th 1379 (Fed. Cir. 2023). Jawbone, a newly formed patent assertion company with no competing product, could not rely on faster time to trial. It was "a clear abuse of discretion to accord this factor any weight" on that record.

How the factors now work

Transfer does not change the applicable state law in a diversity case under Van Dusen v. Barrack, 376 U.S. 612 (1964), and in a patent case appeals still go to the Federal Circuit under 28 U.S.C. § 1295. That reduces the stakes of the familiarity factor and makes judges and juries the true differences between forums.

The Fifth Circuit has been refining the test

Because the law applied is the Fifth Circuit's, its non-patent decisions bind. Three deserve attention.

Planned Parenthood declined to give much weight to the sources-of-proof factor when evidence was electronic and equally accessible, and said district courts are "better placed" to evaluate their own docket efficiency. That sits in some tension with the Federal Circuit's skepticism of Texas trial schedules.

TikTok is a useful corrective. A Chinese software company sued in Waco over video-editing code, in a copyright and trade secret case. TikTok had a 300-person Austin office, none of whose people worked on the accused function. The court granted the writ, observing that "[p]laintiffs are permitted to engage in a certain amount of forum-shopping" but that § 1404(a) protects against the most blatant versions. Source code restricted behind a security clearance was not equally accessible in Texas merely because it was electronic. The court also noted, in a point the Federal Circuit later adopted in SAP, that a judge may not use progress made during a pending motion as a reason to deny it.

In re Clarke, 94 F.4th 502 (5th Cir. 2024), arose from the opposite posture, a district court granting transfer of a case about prediction markets, and raised the bar for movants: "a movant must show (1) that the marginal gain in convenience will be significant, and (2) that its evidence makes it plainly obvious ... that those marginal gains will actually materialize in the transferee venue." It also held that court congestion alone cannot support transfer. Clarke carries a practice lesson: the case had already been shipped to the District of Columbia on the day of the order, which made relief more complicated. A party opposing transfer should ask for a stay of the transfer order immediately.

A denial and a grant, side by side

SAP shows that the writ is not automatic. SAP sought to move a case from Marshall to Sherman within the Eastern District and asked for random reassignment. The Federal Circuit agreed that the district court had erred in relying on a closed companion case and on the case's "smoothly proceeding" to trial. It denied the petition anyway, because SAP had not identified physical evidence, custodians, or unwilling witnesses in Sherman, and its local office had little connection to the case. Monolithic Power likewise denied a § 1404(a) challenge, noting this was not a case with "only one correct outcome."

At the other end sits Google (2026), in which the University of Southern California sued Google in the Western District of Texas over Earth, Maps, and Street View. The district court found that the witness factor favored transfer slightly and all others were neutral. A nonprecedential order granted the writ: "At bottom, this case is about a California university suing a Northern California company for products designed and developed in Northern California based on patents invented by a resident of Northern California." A prior art CD kept in California counted, even though software can be copied. Quality assurance testing on an accused product in Texas, unconnected to the accused functions, could not match California's interest.

Delaware and other districts

Other circuits use their own factors, and the "clearly more convenient" wording belongs to the Fifth Circuit. In Delaware, the Third Circuit's Jumara v. State Farm Insurance Co., 55 F.3d 873 (3d Cir. 1995), controls. In re Link_A_Media Devices Corp., 662 F.3d 1221 (Fed. Cir. 2011), granted mandamus when a Delaware judge treated the defendant's incorporation and the plaintiff's choice as effectively dispositive, on a record where the defendant said Delaware had no other connection to the case. Incorporation in Delaware is sufficient for venue but not an answer to convenience. Delaware has its own distinct feature in the funding and ownership disclosure orders of Chief Judge Connolly, which the Federal Circuit declined to disturb on mandamus in In re Nimitz Technologies LLC, No. 2023-103 (Fed. Cir. Dec. 8, 2022) (nonprecedential), and it later affirmed related sanctions in Backertop Licensing LLC v. Canary Connect, Inc., 107 F.4th 1335 (Fed. Cir. 2024). Patent assertion plaintiffs weigh that when choosing a forum.

Why mandamus treats §§ 1404 and 1406 differently

A strange feature of this field is that a defendant wrongly kept in an improper venue often has a weaker route to immediate review than one denied a discretionary convenience transfer. HTC explains why. Inconvenience cannot be undone after trial and is usually harmless on appeal, so a § 1404(a) denial has no other adequate remedy, and in the Fifth Circuit that first mandamus requirement is treated as satisfied, as TikTok says citing In re Radmax, Ltd., 720 F.3d 285 (5th Cir. 2013). A violation of § 1400(b), by contrast, can be raised on appeal from final judgment, and the judgment vacated.

The court therefore grants § 1406 writs only in narrow circumstances that serve proper judicial administration, drawing on Schlagenhauf v. Holder, 379 U.S. 104 (1964), and La Buy v. Howes Leather Co., 352 U.S. 249 (1957): basic, undecided, recurring legal questions producing conflicting results. Each of Cray, Micron, ZTE, BigCommerce, Google, and Volkswagen fit. Monolithic Power and Comcast did not. Google itself shows the rhythm, with the court declining in 2018 to let the issue develop and granting on the same facts in 2020. The standard from Cheney v. United States District Court, 542 U.S. 367 (2004), applies throughout. Judge Lourie's argument, that a case tried in the wrong place and then retried wastes everyone's resources, has not carried the court. For the general law of the writ, see Interlocutory Appeals and Mandamus: Getting Appellate Review Before Final Judgment.

Waco, Marshall, Wilmington, and the politics of case assignment

Venue doctrine is written around districts, but plaintiffs choose judges. The Waco Division of the Western District of Texas had effectively one active district judge, and a plaintiff filing there could count on Judge Albright, who had been a patent trial lawyer and had issued scheduling orders plaintiffs liked. By one practitioner's estimate in a 2022 interview in Westlaw's Intellectual Property Daily Briefing, around a quarter of pending patent cases nationwide were in his court.

Pressure built. In November 2021, senators wrote to the Chief Justice about the concentration of patent cases in single-judge divisions, and the Chief Justice called for study of assignment practices in patent cases, according to the Judicial Conference's later account. On July 25, 2022, then-Chief Judge Orlando Garcia ordered that new patent cases filed in Waco be randomly assigned among twelve judges across the district. The court issued a further order on patent case assignment on May 30, 2024, so the current version should be checked. On March 12, 2024, the Judicial Conference announced a policy strengthening random assignment. By its terms it addresses civil actions seeking to bar or mandate state or federal actions, and it does not bind courts to any patent-specific rule. And Judge Albright, who had drawn so much of that docket, resigned from the bench effective August 31, 2026.

Marshall, Delaware, and other active patent courts remain. What the Waco story shows is that assignment rules can move faster than doctrine. A plaintiff's analysis today should treat the judge as a variable that can be checked but not guaranteed. A defendant's analysis should ask whether a local rule or general order has changed since the last case it fought there.

Forum choice also interacts with the Patent Office. The Patent Trial and Appeal Board's discretionary denial practice weighs the schedule of the parallel court case, so a fast docket can affect whether an inter partes review is instituted at all. That practice has shifted repeatedly and needs a current check. See Inter Partes Review Strategy: Discretionary Denial, Estoppel, and the Parallel District Court Case.

For a broader view of why forum selection has become so consequential, Daniel Klerman and Greg Reilly's "Forum Selling" (Southern California Law Review) and J. Jonas Anderson's "Court Competition for Patent Cases" (University of Pennsylvania Law Review) remain the leading academic treatments. The Federal Circuit's own citations to Wright and Miller's Federal Practice and Procedure and to Moore's Federal Practice are the usual starting point in treatises, and Robert A. Matthews Jr.'s Annotated Patent Digest, chapter 36, collects district court venue decisions by topic. Robert Tapparo's "The Supreme Court and the Federal Circuit Turn Patent Infringement Venue Jurisprudence Upside Down" (American University Business Law Review) provides a useful account of the immediate aftermath of TC Heartland.

What remains unsettled

What this means in practice

For a patent owner choosing a court. Start from the defendant's corporate filings and work outward, entity by entity. List where it resides. Then look for a real place, with real people, doing the defendant's own business. Publicly available information often answers most of this, and a complaint that pleads those facts survives Westech. Consider whether the foreign parent is the better defendant. Then run the § 1404(a) analysis against yourself, because a Delaware-incorporated California company sued in Texas on the basis of a modest Austin office will likely file a transfer motion with a credible chance of success, however faultless the venue.

For an accused infringer. The first decision after service is the venue decision, and the clock is short. Pull the facts before anything goes on file, and do not admit venue carelessly in the answer. Decide whether you want § 1406, § 1404, or both, and file promptly, since transfer motions are supposed to come first and delay undermines them. Build declarations with specifics about witnesses, their knowledge, and where documents and code are actually accessed. Patent Venue and Transfer Checklist: A Practical Checklist sets out the sequence, and Patent Venue Toolkit: Venue Discovery, Rule 12(b)(3) Motions, and Section 1404 Transfer Papers has working papers.

For a business owner. Where you put people has legal effects you may not have priced. A leased desk, a job posting requiring residence near a customer, a stipend for a home lab, or a warehouse with your own staff can open a courthouse door far from home. That does not mean a company should organize its workforce around lawsuits, but it does mean the facilities and human resources teams should know the question exists. Keeping accurate records of what each location actually does is cheap and will matter if a complaint arrives. If a demand letter arrives first, Managing Willfulness Risk After a Patent Notice Letter: A Practical Guide explains other reasons to act early.

For the broader procedural context of filing, see Comprehensive Guide to Patent Infringement Litigation: From Summary Judgment Denial to Post-Trial, Patent Litigation Toolkit: A Roadmap and Research Guide, and the general treatment of federal jurisdiction in Federal Subject-Matter Jurisdiction: Arising Under, Diversity, Supplemental Claims, and the Defect That Never Waives.

This article is general information about a complicated and moving area of law and does not advise anyone on a particular dispute. If you are deciding where to sue or facing a complaint in a distant court, an early conversation with patent litigation counsel usually costs far less than a venue objection lost by delay.

Related Documents

Related Guides

• Choosing or Challenging Venue in a Patent Case: A Practical Guide: the step-by-step companion for choosing a forum or challenging one, from footprint analysis through mandamus.
• Moving or Keeping a Case in Your Chosen Forum: A Practical Guide: general transfer and remand strategy outside the patent setting.
• Running an ANDA Litigation: A Practical Guide: venue and case management for generic drug litigation.
• Filing or Defending a Declaratory Judgment Action: A Practical Guide: the accused infringer's first-strike option and its limits.
• Recovering From a Missed Federal Court Deadline: A Practical Guide: what to do if a venue or response deadline has already been missed.

Related Articles

• Forum Selection, Transfer, and Parallel Proceedings: Choosing and Keeping the Court You Want: forum-selection clauses, first-filed disputes, and transfer generally.
• Interlocutory Appeals and Mandamus: Getting Appellate Review Before Final Judgment: the general law of the writ that governs venue petitions.
• Personal Jurisdiction Over Online and Foreign Defendants: From Zippo to Ford Motor: the limit that replaces venue for foreign defendants.
• Hatch-Waxman Litigation: The Orange Book, Paragraph IV Certifications, and the 30-Month Stay: the regulatory background to Valeant and Celgene.
• Declaratory Judgment Actions: Jurisdiction, Discretion, and the Race to the Courthouse: mirror-image suits and the general venue statute.
• Inter Partes Review Strategy: Discretionary Denial, Estoppel, and the Parallel District Court Case: how the forum's schedule can affect Patent Office review.
• Indirect and Divided Patent Infringement: Inducement, Contributory Liability, and the Akamai Problem: where inducement and divided infringement occur.
• Federal Subject-Matter Jurisdiction: Arising Under, Diversity, Supplemental Claims, and the Defect That Never Waives: the subject-matter jurisdiction rules that sit beside venue.
• Missed Federal Deadlines: Rule 6, Excusable Neglect, and the Deadlines a Court Cannot Excuse: the consequences of missing a Rule 12 deadline.

Related Checklists

• Patent Venue and Transfer Checklist: A Practical Checklist: a desk reference for venue analysis, Rule 12(b)(3) motions, and transfer.
• Venue and Transfer Checklist: A Practical Checklist: the general federal venue checklist.

Related Toolkits

• Patent Venue Toolkit: Venue Discovery, Rule 12(b)(3) Motions, and Section 1404 Transfer Papers: footprint worksheets, venue interrogatories, motion outlines, and declarations.
• Forum Strategy Toolkit: Clauses, Transfer Motions, and First-Filed Briefing: forum clauses and first-filed briefing.
• Interlocutory Review Toolkit: 1292(b) Certifications, Rule 54(b) Judgments, and Mandamus Petitions: templates for mandamus and other interlocutory review.