A patent infringement injunction is a court order barring an infringer from continuing to make, use, or sell the patented technology, and since the Supreme Court’s 2006 decision in eBay Inc. v. MercExchange, it is no longer automatic after a win at trial. A patent holder now has to prove a four-factor equity test, and judges have wide discretion to say no. Competing companies that practice their patents still get injunctions roughly 84% of the time. Entities that exist primarily to license patents almost never do.
The Four-Factor Test From eBay v. MercExchange
Before 2006, the Federal Circuit treated permanent injunctions as near-automatic once a patent was found valid and infringed. The Supreme Court rejected that approach in eBay Inc. v. MercExchange, L.L.C., holding that patent injunctions follow the same equitable principles as injunctions in any other area of law.1Justia U.S. Supreme Court. eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006) The patent statute gives courts discretion to grant injunctions “in accordance with the principles of equity” and “on such terms as the court deems reasonable,” rather than mandating them.2Office of the Law Revision Counsel. 35 U.S.C. 283 – Injunction
To get a permanent injunction, a patent holder must prove all four of these:
- Irreparable harm from the infringement, meaning injury money alone cannot fix.
- Inadequate legal remedies, meaning damages awarded at trial are not enough to make the patent holder whole.
- A balance of hardships in the patent holder’s favor, weighing continued infringement against the harm a product ban would inflict on the infringer.
- A public interest that would not be disserved by the injunction.
Justice Kennedy’s concurrence, joined by Justices Stevens, Souter, and Breyer, has become the most influential passage in post-eBay litigation. Kennedy warned that an entire industry had developed around firms that use patents “not as a basis for producing and selling goods but, instead, primarily for obtaining licensing fees,” and that injunctions in those cases can become tools to “charge exorbitant fees” through hold-up tactics.1Justia U.S. Supreme Court. eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006) Lower courts have leaned on that passage heavily, particularly when the patented invention is a small component of a larger product.
Who Actually Wins an Injunction After Trial
The eBay decision did not eliminate permanent injunctions. Empirical studies covering 2006 through 2013 found courts granted contested requests about 72.5% of the time. But the split is dramatic. Companies that compete directly with the infringer in the same market succeeded roughly 84% of the time. Patent assertion entities obtained injunctions in only about 16% of cases.
The gap tracks what each factor actually asks. A competitor that makes and sells products can point to lost sales, eroded pricing, and damage to its market position. A licensing entity that doesn’t sell anything has a much harder time arguing money is insufficient, because money is the whole business model. If you compete in the same market as the infringer, eBay changed surprisingly little. If you don’t, the path is steep.
Proving Irreparable Harm
Irreparable harm means injury a check cannot undo. The classic examples are permanent lost market share, price erosion caused by the infringing product driving prices down, and damage to brand reputation or a first-mover advantage.
The hardest piece is the causal nexus. The patent holder has to tie the harm to the patented feature. The Federal Circuit addressed this in Apple Inc. v. Samsung Electronics, holding that “sales lost to an infringing product cannot irreparably harm a patentee if consumers buy that product for reasons other than the patented feature.”3Justia Law. Apple Inc. v. Samsung Electronics Co., Ltd., No. 12-1507 (Fed. Cir. 2012) In multi-feature products like smartphones, which may contain thousands of patented components, that connection is a real challenge.
The Federal Circuit later softened the standard, clarifying that a patent holder does not need to prove the patented feature is the reason customers buy the accused product. Showing “some connection” between the feature and consumer demand is enough. Evidence that the feature is one of several factors in the purchasing decision, or that including it makes the product significantly more desirable, can suffice. A merely nominal contribution still fails.
Timing matters too. A patent holder that sat on infringement for years before seeking an injunction undercuts its own argument. Courts reasonably ask why, if the harm were truly irreparable, the plaintiff waited. Unexplained delay doesn’t automatically bar relief, but it makes the argument harder to sell.
Showing That Damages Alone Are Not Enough
This factor asks why a damage award is insufficient to make the patent holder whole. A jury can compensate past lost profits or set a reasonable royalty for past infringement, but that backward-looking remedy does not address future harm. If the infringer keeps selling, the patent holder faces a recurring problem that requires repeated litigation to fix.
Patent holders often argue that the right to exclude is the core of a patent, and forcing them to accept a royalty converts an exclusion right into a compulsory license. Courts are receptive when the patent holder actually exercises that right by making and selling products. When the patent holder does not practice the patent and exists primarily to collect licensing revenue, courts are far less sympathetic. A court-determined royalty is essentially the same thing a licensing entity was seeking all along.
Practical obstacles to collecting also count. If the infringer is financially unstable, operates overseas, or is likely to restructure before paying, a damage award on paper may be worthless. In those situations, courts find legal remedies inadequate because the patent holder may never actually receive the compensation.
Balance of Hardships
Even when the first two factors favor the patent holder, a court still weighs what shutting down the infringer’s product would cost. The question is proportionality.
Scale matters. If the patented feature is a small component of a complex product, ordering the entire product off the market can feel disproportionate. Courts consider whether the infringer could remove or redesign the infringing component without scrapping everything. In z4 Technologies v. Microsoft, the court weighed the enormous cost Microsoft would bear in re-engineering, testing, and repackaging Windows and Office, ultimately denying the injunction in part because Microsoft already had a non-infringing version in development.
Design-around feasibility carries a lot of weight. If the infringer has already developed or is actively working on a non-infringing alternative, the hardship of an injunction is temporary and manageable. Courts distinguish between legitimate design-around efforts and stalling tactics. A defendant that dragged its feet and then claims hardship gets less sympathy than one that invested in engineering alternatives early.
Relative size can shift the analysis too. A small startup whose entire revenue depends on one product faces potential collapse from an injunction, while a diversified corporation faces an inconvenience. That cuts both ways. A small patent holder against a well-resourced infringer can argue that without an injunction the larger company will simply absorb royalty payments as a cost of doing business and continue eroding the smaller company’s market position.
Public Interest
The final factor looks beyond the two parties. In most commercial disputes involving consumer products, it leans toward granting the injunction. The patent system exists to encourage innovation, and enforcing patent rights generally serves that goal.
The real action is in medical technology and public health. Courts have consistently denied injunctions when removing a product from the market would deprive patients of treatment options. In a case involving prosthetic vascular grafts, the court found that banning the infringing product would “deny many sick patients a full range of clinically effective and potentially life saving treatments.” In another involving extended-wear contact lenses, the court noted that millions of wearers would suffer “real adverse consequences.” Similar reasoning has protected coronary stents, IV catheters, contraceptive devices, and experimental heart valve technology.
The common thread is market concentration. When the infringing product is one of only a handful of options available to patients or clinicians, pulling it shrinks choices in ways that directly affect health outcomes. The same principle applies to critical infrastructure. Outside those narrow categories, the public interest factor rarely blocks an injunction on its own.
What Happens When an Injunction Is Denied
When a court finds infringement but concludes the eBay factors don’t support an injunction, the infringer doesn’t simply walk away. The patent statute requires courts to award “damages adequate to compensate for the infringement, but in no event less than a reasonable royalty.”4Office of the Law Revision Counsel. 35 U.S.C. 284 – Damages That includes an ongoing royalty for future use, effectively a court-imposed license.
The Federal Circuit has made clear that post-verdict royalties are “fundamentally different” from the rate used to calculate pre-verdict damages. Before trial, both validity and infringement are uncertain. After a verdict, that uncertainty is gone, and the infringer is now knowingly choosing to continue using technology a court has confirmed belongs to someone else. Post-verdict rates are often higher than pre-verdict rates.
Courts setting the ongoing rate consider changed market conditions since infringement began, how important the patented technology is to the product, whether a design-around is feasible, the relative market positions of the parties, and the commercial success of the technology. The process typically follows one of several paths: the court may give the parties a window to negotiate, hold an evidentiary hearing if negotiations fail, or in some cases have the jury determine a future royalty rate during the original trial.
For the patent holder, an ongoing royalty is a consolation prize. It provides revenue but strips away the power to exclude. For the infringer, it provides certainty at a rate reflecting its weaker bargaining position as an adjudicated infringer.
What the Injunction Actually Covers, and For How Long
A permanent patent injunction is not a blank check to block anything the infringer does in the future. Federal rules require that an injunction “state its terms specifically” and “describe in reasonable detail” the acts it prohibits.5Cornell Law Institute. Federal Rules of Civil Procedure Rule 65 – Injunctions and Restraining Orders The Federal Circuit has rejected injunctions that simply prohibit “future infringement of a patent” as overly broad. The order must identify the specific infringing products or processes, or products that are no more than trivially different.
That creates a clear enforcement boundary. If the infringer redesigns its product after the injunction issues, the patent holder cannot enforce the order against the redesigned product unless the changes are merely cosmetic. Genuine redesigns require a new infringement action, not a contempt proceeding.
Despite the label “permanent,” these injunctions don’t last forever. A patent injunction cannot outlive the patent itself. Once the patent expires, the right to exclude disappears and the injunction loses its legal basis. Courts have confirmed that injunctive relief is “not intended to prohibit future use of the invention” after expiration. A court may explicitly limit the injunction’s duration to the remaining patent term. Some post-expiration cleanup is permitted, like requiring destruction of infringing inventory produced during the patent’s life, but the former infringer is free to practice the now-expired patent going forward.
Enforcing the Order Through Contempt
An injunction is only as strong as the consequences for violating it. Patent injunctions are enforced through contempt, and the penalties can be severe. Under federal law, courts have the power to punish disobedience of their orders by fine, imprisonment, or both.6Office of the Law Revision Counsel. 18 U.S.C. 401 – Power of Court
Most enforcement proceeds as civil contempt in one of two forms. Compensatory sanctions reimburse the patent holder for actual losses caused by the violation. Coercive sanctions impose escalating fines or other penalties designed to pressure compliance, such as a daily fine that runs until the infringing product is pulled from the market. In egregious cases, courts can enhance damages up to treble the actual loss and order the infringer to hand over profits earned from the contemptuous conduct. Attorney fees for the contempt proceeding are frequently awarded too.
Criminal contempt for violating a patent injunction exists in theory but is extremely rare. The government must prove beyond a reasonable doubt that the violation was willful and deliberate, a standard borrowed from the general criminal contempt statute requiring a violation of “a clear and reasonably specific order” done intentionally.7United States Department of Justice. Criminal Resource Manual 753 – Elements of the Offense of Contempt The civil tools are usually strong enough that criminal proceedings almost never become necessary.
Courts can also modify the original injunction in response to violations, potentially expanding its scope. Each violation gives the court reason to tighten the restrictions further.
Preliminary Injunctions Before Trial
A patent holder facing ongoing harm does not have to wait for a trial verdict. Under the federal rules, courts can issue a preliminary injunction at the outset of litigation to halt allegedly infringing activity while the case proceeds.5Cornell Law Institute. Federal Rules of Civil Procedure Rule 65 – Injunctions and Restraining Orders The test looks similar to the eBay framework with one crucial addition: the patent holder must show a reasonable likelihood of success on the merits, meaning it will probably prove both that the patent is valid and that the defendant infringes.
The four preliminary-injunction factors are likelihood of success on the merits, irreparable harm before trial concludes, the balance of hardships from acting before a full trial, and the public interest.
There is a catch. The patent holder must post a security bond to cover the infringer’s losses if the injunction turns out to have been wrongfully granted.5Cornell Law Institute. Federal Rules of Civil Procedure Rule 65 – Injunctions and Restraining Orders For products with significant sales volume, the bond can be substantial, and it effectively caps the defendant’s later recovery if the preliminary injunction is reversed.
Preliminary injunctions are granted far less often than permanent ones. The likelihood-of-success requirement creates inherent tension: the patent holder must essentially preview its case before discovery is complete. Accused infringers routinely raise invalidity defenses at this stage specifically to undermine that factor, and if they present clear and convincing evidence of invalidity, a preliminary injunction cannot issue.
Appeals and Stays
Whether the trial court grants or denies a permanent injunction, the losing side can appeal to the Federal Circuit. Review is under an abuse-of-discretion standard, which is deliberately deferential. The Federal Circuit will overturn the trial court only if the judge made a clear error in weighing the factors or based the decision on a legal mistake or clearly wrong factual findings.
While the appeal is pending, the injunction remains in effect unless the court grants a stay. The stay standard mirrors the preliminary injunction test: strong likelihood of success on appeal, irreparable harm without the stay, no substantial harm to the other party, and a public interest that favors pausing. In practice, these often collapse into two questions: how likely is the appellant to win, and do the equities favor pausing enforcement.
For a patent holder who won an injunction, a stay is a frustrating delay while the infringer keeps selling. For an infringer facing a product ban, a stay can be essential to avoiding irreversible business damage from an order that might ultimately be reversed. Courts weigh the competing urgencies case by case, with no presumption in either direction.