In August 2012, a jury in San Jose handed Apple a verdict against Samsung for more than a billion dollars. The number flashed across every business page in the world, and for a brief moment the public imagined that suing over a rounded-corner rectangle was a reliable path to riches. What the headlines rarely mentioned is what happened next: the trial judge knocked roughly $450 million off the award and ordered a partial retrial; the case bounced to the Supreme Court on a single question about how to apply one sentence of the Patent Act; it came back down for yet another trial; and the whole saga finally ended in a quiet, confidential settlement in 2018--six years and three damages trials after that first electrifying number. The billion-dollar verdict was real. It was also, in a sense, a mirage.

That gap--between the eye-popping headline and the messy, contested, often-reduced reality--is the single most important thing to understand about damages in intellectual property litigation. IP awards are the most skewed in all of civil law. A handful of cases produce nine- and ten-figure judgments that dominate the conversation, while the overwhelming majority of IP disputes resolve for sums that would not cover a senior partner's monthly billings. Both facts are true at once, and any honest account of "IP damage statistics" has to hold them together.

This guide is that honest account. It explains, in plain language a judge, a lawyer, and a business owner can all follow, how American courts actually calculate what an act of infringement is worth. We will move through the four great IP regimes in turn--patents, copyrights, trademarks, and trade secrets--because each has its own statute, its own vocabulary, and its own logic for translating a wrong into a dollar figure. Along the way we will meet the reasonable royalty and its imaginary negotiation, the fifteen Georgia-Pacific factors, the willfulness question that can triple a verdict, copyright's strange and powerful statutory-damages menu, the trademark plaintiff's right to take the infringer's profits, and the trade-secret regime's three-headed measure of recovery. We will also do something the original version of this article could not: we will tell you the truth about the numbers, including why the "average IP verdict" is one of the least useful statistics in law.

A note before we begin. Nothing here is a prediction of what any particular case is worth. Damages depend on facts--sales volumes, profit margins, the strength of the right, the conduct of the defendant, and the persuasiveness of an expert witness sitting in front of a jury. Treat what follows as a map of the terrain, not a price list.

Why "Average" Is a Trap: How to Read IP Damage Statistics Honestly

Start with the headline you should be suspicious of: "The average IP damages award is X." You will see numbers like that quoted confidently. Ignore them, or at least handle them with tongs.

The problem is statistical, and it is fundamental. Damage awards in IP cases follow a distribution so lopsided that the mean--the ordinary "average" you learned in school--is almost meaningless. Imagine a room of one hundred IP plaintiffs. Ninety-something of them recovered, settled for, or won default judgments of a few thousand to a few hundred thousand dollars. A couple recovered a few million. And one walked out with a verdict of $300 million. Compute the "average" and you get a figure in the millions--a number that describes none of the actual people in the room. The single giant verdict drags the mean far above the experience of nearly everyone else.

This is why serious analysts speak in terms of medians and distributions, not averages. The median--the midpoint, where half the awards are larger and half smaller--tells you what a typical case looks like. And the typical IP case is modest. Aggregated dockets across two decades of federal district-court IP and technology cases show a distribution heavily weighted toward small recoveries: a large share of awards fall below $50,000, a meaningful middle band sits in the tens to low hundreds of thousands, and only a small single-digit percentage of dockets cross into the multi-million-dollar territory that makes the news. Median compensatory awards in the broad run of IP dockets tend to cluster in the low tens of thousands of dollars, while attorney-fee awards--when granted at all--often run in the single-digit thousands. (We cite these as directional patterns drawn from docket-level data, not as precise national statistics; the underlying datasets vary in coverage, classification, and the cases they capture, and we will not pretend to a precision the sources do not support.)

Three structural facts drive the skew, and they are worth internalizing.

First, most cases never reach a damages verdict at all. The great majority of filed IP cases settle, are dismissed, or end in default judgments against absent defendants (think of the thousands of counterfeit-goods suits filed against anonymous online sellers, where a court enters a statutory-damages judgment that may never be collected). The cases that go all the way to a jury verdict on damages are a tiny, unrepresentative slice--and they are precisely the slice that generates the headlines. Any statistic built only on tried verdicts oversamples the giants.

Second, the four IP regimes behave completely differently, so lumping them together produces a meaningless blend. A patent reasonable-royalty case and a counterfeit-handbag statutory-damages case have almost nothing in common except the courthouse. Patent trials, though rare, produce the largest median and mean awards by a wide margin. Trademark and copyright cases span an enormous range, from nominal default judgments to substantial profit disgorgements. Mixing them is like averaging the price of a house and a sandwich.

Third, the headline number is rarely the final number. Post-trial motions, remittitur (a judge's reduction of a "shocking" award), appeals, and settlements routinely cut large verdicts dramatically or erase them. The Apple-Samsung story is the canonical example, but it is the rule, not the exception, at the top of the distribution. Counting the verdict and ignoring what came after overstates real recoveries.

So when this guide gives figures, it gives them as ranges, medians, and clearly labeled illustrations. It will not invent a national "average IP verdict," because that number would be both unknowable with precision and actively misleading. With that honesty established, let us look at how the dollars are actually built, regime by regime.

Patent Damages: The Floor, the Hypothetical Negotiation, and the Multiplier

Patent law sets the tone for the whole field, both because patent verdicts are the biggest and because the patent damages statute is the most fully developed. That statute is 35 U.S.C. § 284, and its first sentence is one of the most consequential lines in American commercial law:

"Upon finding for the claimant the court shall award the claimant damages adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer."

Read that "in no event less than" clause carefully, because it does a lot of work. It establishes a floor. Even a patentee who cannot prove that it lost a single sale--because, say, it does not even make a product, or because the market is crowded--is still entitled to a reasonable royalty. Section 284 guarantees that a proven infringement is never free. The statute also lets the court award interest and costs, and--crucially--allows the court to "increase the damages up to three times the amount found or assessed." Hold that treble-damages power; we will return to it.

For a deeper treatment of how patent cases move from pleading through post-trial motions, see our companion piece, a comprehensive guide to patent infringement litigation, from summary judgment denial to post-trial. And for the threshold question of what actually counts as infringement in the first place, see what constitutes patent infringement.

Patent compensatory damages come in two basic flavors, and a patentee may pursue either or both for different sales.

Lost Profits: The Sales You Would Have Made

Lost profits answer a counterfactual question: but for the infringement, how much money would the patentee have made? This is the patentee's preferred theory, because it captures the full margin on sales the patentee says were stolen, not merely a royalty slice.

The classic test for proving lost profits comes from Panduit Corp. v. Stahlin Bros. Fibre Works, 575 F.2d 1152 (6th Cir. 1978), which has become the Federal Circuit's working framework. To recover lost profits under Panduit, a patentee generally must show four things: (1) demand for the patented product; (2) the absence of acceptable, non-infringing substitutes--because if customers could have bought a perfectly good legal alternative, the patentee cannot claim it would have captured those sales; (3) the patentee's manufacturing and marketing capacity to have made the additional sales; and (4) the amount of profit it would have earned. The second factor is where most lost-profits theories live or die. In a crowded market full of alternatives, a patentee is often pushed back to a royalty.

Worked example. Suppose Acme Surgical holds a patent on a uniquely safe surgical stapler and is the only firm selling a stapler with that safety feature. Beta Devices copies it and sells 10,000 infringing units. If Acme can show that surgeons demanded that safety feature, that no acceptable non-infringing substitute existed, that Acme had the factory capacity to make those 10,000 units, and that Acme's margin is $400 per unit, Acme can claim $4 million in lost profits. Now change one fact: suppose three other companies sold lawful staplers with comparable safety features. Suddenly Acme cannot show it would have captured those particular sales, and its recovery collapses back to a reasonable royalty on Beta's 10,000 units.

The Reasonable Royalty and the Hypothetical Negotiation

When lost profits are unavailable or only partly available, the patentee falls back to the statutory floor: a reasonable royalty. The dominant method for computing one is the hypothetical negotiation--a delightful legal fiction in which we pretend that the patentee and the infringer sat down at a conference table just before infringement began and bargained, in good faith, over a license. The "willing licensor, willing licensee" both assume the patent is valid and infringed. The royalty they would have agreed to becomes the damages.

The canonical list of inputs to that imaginary negotiation comes from a single influential district-court opinion, 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 court laid out fifteen factors--the famous Georgia-Pacific factors--that bear on a reasonable royalty. They include things like: royalties the patentee has actually received for licensing the patent; rates the licensee pays for comparable patents; the nature and scope of the license (exclusive or not, territory, field of use); the commercial relationship between the parties (are they competitors?); the profitability of the patented product and its commercial success; the utility and advantages of the patent over old modes; the portion of profit attributable to the invention as opposed to non-patented features; expert opinion; and--factor fifteen, the catch-all--the outcome of the hypothetical negotiation itself. Damages experts march through these factors like a checklist, and cross-examination often consists of attacking the expert's treatment of two or three of them.

The hypothetical-negotiation framework sounds tidy. In practice it is where patent cases become a battle of dueling economists, and where some of the most important damages law of the last two decades has been made--almost all of it pushing royalties down and demanding more rigor.

Apportionment: You Get Paid for the Invention, Not the Whole Machine

The single most important development in modern patent damages is the hardening of the apportionment requirement. The principle is old--it traces to Garretson v. Clark, 111 U.S. 120 (1884), where the Supreme Court said a patentee "must in every case give evidence tending to separate or apportion" the defendant's profits between the patented improvement and the unpatented features. But the Federal Circuit revitalized it in a series of opinions that reshaped how royalties are calculated.

The core idea: when a patent covers one small component of a large, multi-feature product, the damages must reflect the value of that component, not the value of the whole product. A patent on an improved smartphone antenna does not entitle the patentee to a royalty on the entire phone. The royalty base must be tied to the "smallest salable patent-practicing unit," and the patentee may use the larger product's revenue as the base only in narrow circumstances--chiefly when the patented feature drives consumer demand for the whole product, the demanding entire market value rule (EMVR). Key cases include LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51 (Fed. Cir. 2012), and VirnetX, Inc. v. Cisco Systems, Inc., 767 F.3d 1308 (Fed. Cir. 2014), which together made clear that pointing a jury at a huge revenue number and applying a modest-sounding percentage is a recipe for reversal. Courts are vigilant about the "anchoring" risk--that a jury shown a billion-dollar base will award a big number regardless of the rate.

Worked example. Imagine Gadgetron holds a patent on a power-saving circuit used in a $1,200 laptop. The patented circuit is one of hundreds of components and adds, say, $3 of value to the device. A damages theory that takes 1% of the laptop's $1,200 price ($12 per unit) is almost certainly improper; the patentee must instead anchor the royalty to the value the patented circuit actually contributes. Get this wrong and the verdict gets thrown out, no matter how sympathetic the patentee.

This apportionment rigor is the main reason eye-popping patent verdicts so often shrink on appeal. It is also why the damages expert is the most important witness in a modern patent trial--more on that below.

Enhanced Damages and Willfulness After Halo

Now back to that treble-damages power in § 284. The court "may increase the damages up to three times." For decades, the Federal Circuit policed that discretion with a rigid two-part test from In re Seagate Technology, LLC, 497 F.3d 1360 (Fed. Cir. 2007), which required a patentee to prove "objective recklessness" by clear and convincing evidence--a standard so demanding that enhanced damages became rare.

The Supreme Court swept Seagate away in Halo Electronics, Inc. v. Pulse Electronics, Inc., 579 U.S. 93 (2016). Writing for a unanimous Court, Chief Justice Roberts held that § 284 commits the enhancement decision to the discretion of the district court, guided by nearly two centuries of practice reserving enhanced damages for egregious cases of culpable misconduct--"the wanton and malicious pirate," in the Court's vivid phrase. Halo lowered the burden of proof to a preponderance of the evidence and eliminated the rigid objective-recklessness gate. The upshot: willful infringement--knowing, deliberate copying, or continued infringement in the teeth of a known patent--can now lead a court to enhance damages up to threefold more readily than before. Halo did not make enhancement automatic; courts still reserve it for genuinely bad conduct, and the Read Corp. v. Portec factors guide how much to enhance. But Halo meaningfully raised the stakes for a company that infringes with its eyes open.

This is also why opinions of counsel and clearance still matter. A defendant who obtained a competent, good-faith opinion that it did not infringe, or that the patent was invalid, has a far better story to tell on willfulness. Our discussion of how clearance and good faith blunt willful-infringement exposure in the trademark context applies in spirit here as well; see the shield of good faith: how trademark clearance searches, attorney opinions, and USPTO approval can protect against willful infringement claims.

Fees, Interest, and the Marking Trap

Two more pieces complete the patent picture. Under 35 U.S.C. § 285, a court may award attorney's fees to the prevailing party "in exceptional cases." The Supreme Court loosened the standard in Octane Fitness, LLC v. ICON Health & Fitness, Inc., 572 U.S. 545 (2014), defining an exceptional case as simply one that "stands out from others" with respect to the strength of a party's litigating position or the unreasonable manner in which the case was litigated--judged by a preponderance and reviewed deferentially under the companion case Highmark Inc. v. Allcare Health Management System, Inc., 572 U.S. 559 (2014). This cuts both ways: it lets a defendant recover fees against a weak "troll" suit, and it lets a patentee recover fees against a defendant who litigated abusively.

Finally, a trap that quietly destroys damages: the patent marking requirement of 35 U.S.C. § 287. A patentee who sells products embodying the patent but fails to mark them (with the patent number or a virtual-marking web address) generally cannot recover damages for the period before the infringer received actual notice--often the filing of the complaint. We cover this in detail in understanding patent marking requirements: a practical guide for practitioners and entrepreneurs. A failure to mark can wipe out years of otherwise-recoverable royalties.

A Note on Design Patents: The § 289 Bombshell

Design patents--which protect the ornamental appearance of an article rather than its function--carry a uniquely powerful damages remedy under 35 U.S.C. § 289: the infringer's total profit on the article of manufacture, with no apportionment in the usual sense. This was the engine of the Apple-Samsung verdict. The Supreme Court's decision in Samsung Electronics Co. v. Apple Inc., 580 U.S. 53 (2016), held that the "article of manufacture" to which the total-profit remedy applies need not be the entire end product sold to consumers; it can be just a component. That ruling sent the case back for a fresh determination of which "article" mattered--a phone, or a few of its design elements--before the parties finally settled. We unpack design-patent remedies in patent litigation: design patents and contrast design and utility protection in design patents v. utility patents: key differences.

Copyright Damages: Actual Loss, Infringer's Profits, or the Statutory Menu

Copyright damages run on a different and, in some ways, more plaintiff-friendly statute: 17 U.S.C. § 504. A successful copyright plaintiff elects between two paths, and the choice is often dispositive.

Path One: Actual Damages Plus the Infringer's Profits

Under § 504(b), the copyright owner may recover its actual damages (the market harm--lost sales, lost license fees, diminished value of the work) plus any profits of the infringer that are attributable to the infringement and not already counted in actual damages. The statute helpfully shifts the burden on profits: the plaintiff need only prove the infringer's gross revenue, and the burden then falls on the infringer to prove its deductible expenses and the portion of profit attributable to factors other than the copyrighted work. This is a meaningful advantage--the plaintiff points at the defendant's revenue, and the defendant has to do the work of carving it down.

But carving it down is exactly what defendants do, and courts require the apportionment of profits to be reasonable, not speculative. If a publisher made $5 million selling a textbook that infringed three pages of the plaintiff's work, the plaintiff is not getting $5 million; the defendant will (rightly) argue that the overwhelming bulk of the profit came from the other 800 pages, the authorship, the editing, the brand, and the sales force.

Path Two: Statutory Damages--Copyright's Trump Card

Here is where copyright becomes distinctive. Under § 504(c), the plaintiff may, at any time before final judgment, elect statutory damages instead of actual damages and profits. Statutory damages require no proof of actual harm at all--a powerful tool when the harm is hard to quantify (which it often is) or when the infringer's profits are murky. The range:

  • $750 to $30,000 per work infringed, "as the court considers just."
  • Up to $150,000 per work where the infringement was willful.
  • As low as $200 per work where the infringer proves it was an innocent infringer who had no reason to believe its conduct infringed.

Two features make statutory damages formidable. First, the award is per work infringed, not per act of infringement, and a defendant who copies many works can face the multiplier stacking up fast: 50 infringed photographs at even a mid-range figure becomes a very large number. Second, willfulness pushes the ceiling to $150,000 per work, and "willful" in copyright includes reckless disregard of the copyright owner's rights--a standard easier to meet than one might think.

There is a crucial gatekeeper, however: under 17 U.S.C. § 412, statutory damages and attorney's fees are available only if the work was registered with the Copyright Office before the infringement began (or, for published works, within three months of first publication). This is the single best reason to register promptly. An owner who waits until after the infringement to register is limited to actual damages and profits--often far less, and far harder to prove. We walk through the mechanics in how to register a copyright with the U.S. Copyright Office and the broader framework in copyright registration: a comprehensive guide. And recall that under Fourth Estate Public Benefit Corp. v. Wall-Street.com, LLC, 586 U.S. 296 (2019), a copyright owner generally cannot even file suit until registration has issued.

Worked example. Pixel Studio, a photographer, registers a portfolio of 40 images. A fashion blog scrapes and republishes all 40 without a license, ignoring two cease-and-desist letters. Because the works were registered before the infringement and the conduct looks willful, Pixel can elect statutory damages and ask for up to $150,000 per image--a theoretical ceiling of $6 million for 40 works--plus attorney's fees under § 505. The real award will land well below the ceiling (courts calibrate within the range), but the leverage that menu gives Pixel in settlement talks is enormous, and it explains why copyright "trolling"--mass suits over scraped photos--is a viable, if unlovely, business model. The consequences of unauthorized use are explored further in what are the consequences of pirating intellectual property.

Attorney's Fees and the Online Wrinkle

Section 505 gives courts discretion to award full costs and a reasonable attorney's fee to the prevailing party. The Supreme Court explained the standard in Kirtsaeng v. John Wiley & Sons, Inc., 579 U.S. 197 (2016), directing courts to give "substantial weight" to the objective reasonableness of the losing party's position while considering all the circumstances. As with statutory damages, the § 412 registration timing gate applies.

For online and platform infringement, the damages calculus interacts with the DMCA's safe harbors, which can immunize service providers who follow the notice-and-takedown rules--meaning the dollars may run only against the uploader, not the platform. See Digital Millennium Copyright Act safe harbors for online service providers and how to file a DMCA takedown notice and respond to one.

Trademark Damages: Profits, Actual Damages, and the Romag Earthquake

Trademark remedies live in Section 35 of the Lanham Act, 15 U.S.C. § 1117. A prevailing trademark owner may, "subject to the principles of equity," recover three things: (1) the defendant's profits, (2) any damages sustained by the plaintiff, and (3) the costs of the action. As in copyright, the statute makes proving the defendant's profits easier on the front end: the plaintiff proves the defendant's sales, and the defendant must prove its costs and deductions.

Trademark damages have always been a bit awkward, because the core trademark harm--consumer confusion about source--is real but slippery to monetize. A plaintiff rarely loses one sale for every infringing sale (consumers confused into buying the knockoff might not have bought the genuine article at all). So courts lean heavily on disgorgement of the infringer's profits as both a compensatory proxy and a deterrent--taking away the unjust enrichment the infringer earned by trading on the plaintiff's goodwill.

Romag v. Fossil: Willfulness Is No Longer a Hard Gate to Profits

For years, the federal circuits split bitterly over a threshold question: must a trademark plaintiff prove the infringement was willful before it can recover the defendant's profits? Many circuits said yes--no willfulness, no disgorgement. The Supreme Court resolved the split in Romag Fasteners, Inc. v. Fossil, Inc., 590 U.S. 212 (2020).

The Court, in a crisp opinion by Justice Gorsuch, held that willfulness is not an absolute precondition to a profits award under § 1117(a) for ordinary trademark infringement. The statutory text demands willfulness for one specific category (dilution claims under § 1125(c)) but conspicuously does not impose it as a gateway for § 1125(a) infringement. That said--and this is the part people forget--Romag did not make willfulness irrelevant. The Court emphasized that a defendant's mental state remains "a highly important consideration" in deciding whether an award of profits is equitable. So after Romag, willfulness is no longer a yes/no switch that bars disgorgement, but it remains a heavy thumb on the scale. An innocent infringer may still avoid disgorgement; a deliberate one is far more exposed. We discuss the case-law backdrop in navigating the maze of trademark confusion: key considerations for brand owners and the remedies landscape in trademark overview: infringement and related rights.

Enhanced Awards, Counterfeiting, and Statutory Damages

The Lanham Act gives courts power to adjust monetary awards in the interest of justice. Under § 1117(a), a court may enter judgment for up to three times actual damages if the compensatory amount is inadequate, and may increase or decrease a profits award the court finds excessive or inadequate--though the statute cautions that such adjustments must be compensation and not a penalty.

Counterfeiting--the use of a spurious mark identical to a registered mark--triggers a far harsher regime. Under § 1117(b), use of a counterfeit mark generally requires the court to award treble damages or profits, plus reasonable attorney's fees, absent extenuating circumstances. And under § 1117(c), a plaintiff in a counterfeiting case may elect statutory damages--ranging from $1,000 to $200,000 per counterfeit mark per type of goods, and up to $2,000,000 per mark for willful counterfeiting. This statutory-damages option is the workhorse of the thousands of online counterfeit-goods suits filed each year against anonymous overseas sellers, where actual damages are impossible to prove and the defendants never appear. It is also a major reason raw IP "verdict" datasets are so distorted: a single suit may name dozens of counterfeit storefronts and generate enormous paper judgments that bear little relation to collectible dollars.

Section 1117 also authorizes attorney's fees in "exceptional cases"--the same statutory phrase the Supreme Court interpreted for patents in Octane Fitness, and which most circuits now apply to Lanham Act fee awards as well.

For brand owners trying to police infringement and counterfeiting at scale, the practical playbook is in brand protection online: a strategic guide for businesses, and the false-advertising cousin of trademark damages is covered in false advertising and Lanham Act Section 43(a): competitor claims in technology marketing.

Trade Secret Damages: The Three-Headed Remedy and Exemplary Multipliers

Trade secrets are the youngest member of the federal IP family. For most of American history, trade-secret law was purely a creature of state law (the Uniform Trade Secrets Act, adopted in some form by nearly every state). That changed in 2016 with the Defend Trade Secrets Act (DTSA), 18 U.S.C. § 1836, which created a federal civil cause of action without displacing the state-law UTSA claims. The two regimes largely parallel each other on damages, and plaintiffs frequently plead both.

Trade-secret damages are distinctive because the statute offers a three-part menu, and the plaintiff can often stack the parts so long as it avoids double-counting. Under DTSA § 1836(b)(3)(B), a court may award:

  1. Actual loss caused by the misappropriation (the plaintiff's lost profits, lost value, or the cost of research and development that the defendant got for free); and
  2. Unjust enrichment caused by the misappropriation, to the extent not already captured in actual loss (the gains the defendant reaped from using the secret--head-start profits, avoided development costs); or
  3. In lieu of damages measured by actual loss and unjust enrichment, a reasonable royalty for the unauthorized use--the same hypothetical-negotiation logic borrowed from patent law, useful when actual loss and enrichment are hard to pin down.

On top of that, the DTSA authorizes exemplary (punitive) damages of up to two times the compensatory award where the misappropriation was willful and malicious, and attorney's fees to the prevailing party in cases of willful and malicious misappropriation (or bad-faith claims/defenses). So a willful trade-secret theft can expose a defendant to compensatory damages plus double that amount in exemplary damages, plus fees. That is why some of the largest IP verdicts of recent years--against companies that hired away a competitor's engineers and walked off with the playbook--have been trade-secret cases rather than patent cases.

Worked example. Suppose Nimbus Robotics spends $8 million over four years developing a proprietary motion-control algorithm and guards it carefully. A departing lead engineer copies the source and brings it to rival Vertex Automation, which uses it to leapfrog two years of R&D and launch a competing product. Nimbus can claim its actual loss (lost sales and license value), plus Vertex's unjust enrichment (the development costs Vertex avoided and the head-start profits it earned), so long as those measures do not overlap. If a jury finds the theft willful and malicious, the court may add exemplary damages up to twice the compensatory figure and award Nimbus its attorney's fees. The combination is what makes trade-secret litigation so dangerous for the defendant--and so attractive to the plaintiff.

The flip side is that trade-secret protection is fragile: it exists only so long as the information is, in fact, secret and reasonably protected. A plaintiff who did not take "reasonable measures" to keep the information confidential may have no trade secret to vindicate at all. The damages discussion therefore presupposes a well-run protection program. We cover building one in building a trade secret protection program from scratch and the doctrinal basics in protection of trade secrets. The growing risk of secret loss through cloud and remote-work channels is addressed in trade secrets in the age of remote work and cloud computing.

Why IP Awards Swing So Wildly--and the Outsized Role of the Damages Expert

Pull back from the four regimes and a pattern emerges. More than almost any other field, IP damages are constructed, not merely measured. There is no invoice for "infringement." Instead, a damages number is built from a chain of assumptions: What is the right royalty base? What rate would the parties have agreed to? How much of the product's value is attributable to the patented feature, the copyrighted element, the brand? How many sales were really diverted? Each link in that chain is contestable, and small changes in assumptions produce enormous changes in the result. A royalty rate of 2% versus 5% on a large base is the difference between a modest verdict and a fortune.

That is why the damages expert--almost always an economist or a forensic accountant--is frequently the most important witness in an IP trial. The expert builds the damages model, ties it to the Georgia-Pacific factors or the apportionment of profits or the unjust-enrichment calculation, and defends it under withering cross-examination. And because the model rests on assumptions, it is vulnerable to a Daubert challenge--the gatekeeping inquiry from Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), now embodied in Federal Rule of Evidence 702. A 2023 amendment to Rule 702 sharpened the court's gatekeeping duty, making clear that the proponent must show by a preponderance that the expert's opinion reliably applies the methodology to the facts. Courts increasingly exclude damages experts whose royalty theories rest on non-comparable licenses, arbitrary "rules of thumb" (the Federal Circuit famously rejected the old "25% rule" in Uniloc USA, Inc. v. Microsoft Corp., 632 F.3d 1292 (Fed. Cir. 2011)), or improper use of the entire market value. When a damages expert is struck on the eve of trial, the plaintiff's case can collapse to the statutory floor or to nothing.

Several other forces amplify the volatility:

The right's strength. A broad, valid, plainly infringed patent on a feature that drives demand is worth far more than a narrow patent of doubtful validity. Defendants routinely counterattack on validity--and an invalidity finding (or an inter partes review that cancels the claims) reduces damages to zero, no matter how large the would-be award. Damages and validity are inseparable in practice.

Willfulness. As we have seen, willfulness can treble patent damages, push copyright statutory damages to $150,000 per work, tilt the equities toward trademark profit disgorgement, and unlock double exemplary damages and fees under the DTSA. Whether the defendant infringed knowingly is often worth more than the underlying compensatory number.

The forum and the jury. Some districts have reputations--deserved or not--as patentee-friendly or defendant-friendly. Docket data show meaningfully different median awards across districts, reflecting case mix, local rules, and jury pools. A case tried in one district may be worth a multiple of the same case tried in another.

Injunctions, which do not show up in the damages column at all. After eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006), injunctions in patent cases are no longer automatic; a plaintiff must satisfy the traditional four-factor equitable test. But where an injunction does issue--common in trademark and trade-secret cases--it can be worth far more than any damages award, because it shuts the infringer out of the market. As the original version of this article rightly noted, an injunction is sometimes the real prize, and it never appears in a "damages statistic." When you read that an IP plaintiff "only" recovered a modest sum, ask whether it also walked away with an injunction that reshaped a market.

Settlement, the great invisible resolver. Because trial is expensive and uncertain, the vast majority of meritorious IP disputes settle--often confidentially, for sums that never enter any public dataset. The verdicts that do get reported are a biased sample: cases where one side badly misjudged its position, or where the stakes were too large to compromise. The "statistics," in other words, describe the cases that failed to settle, not the field as a whole. For a structured way to think about case value before you ever get near a verdict, see evaluating and assessing a civil case, and for the opening move in many disputes, writing a demand letter basics.

Reading the Numbers: A Field Guide to IP Damage Data

So what can we responsibly say about the numbers? Here is a sober field guide.

The distribution is bimodal and skewed. A very large share of IP dockets resolve for small sums--default judgments, modest settlements, and statutory-damages awards in counterfeiting and copyright cases that may run from a few thousand to a few tens of thousands of dollars. A thin upper tail of contested patent, trade-secret, and high-stakes trademark cases produces the multi-million- and occasionally billion-dollar figures that define the public image of the field. Aggregated district-court data across roughly two decades show a clear weighting toward the lower bands, with median monetary recoveries in the broad IP-and-technology category sitting in the low tens of thousands of dollars and attorney-fee awards, when granted, often in the single-digit thousands. We present these as directional medians, not precise national constants; classification choices (what counts as "IP," whether defaults are included, how settlements are coded) move the numbers, and no single dataset captures the field cleanly.

Patents sit at the top. Among IP categories, patent cases produce by far the largest median and mean awards--driven by the reasonable-royalty floor, the entire-market-value temptation, design-patent total-profit recoveries, and treble enhancement. They are also the most likely to be reduced post-trial, precisely because the big numbers attract the most rigorous appellate scrutiny on apportionment and the EMVR.

Copyright and trademark span the full range. At the bottom, mass default and statutory-damages judgments; at the top, substantial profit disgorgements and willful-infringement awards. Copyright's per-work statutory multiplier and trademark's per-mark counterfeiting statutory damages create the potential for very large paper judgments that may or may not be collectible.

Trade-secret verdicts have grown. The DTSA's federal forum, the three-part damages menu, and the up-to-double exemplary multiplier have made trade-secret cases a leading source of the very largest IP verdicts in recent years--especially employee-mobility and competitor-poaching disputes.

And one more piece of advice for anyone tempted to extrapolate from a verdict to a case value: don't. The verdict you read about is the product of a specific record, a specific expert, a specific jury, and a specific district--and it is usually not the last word, because remittitur, post-trial motions, appeal, and settlement so often follow. Use the framework in this guide to understand how the number was built, not the number itself as a benchmark.

Key Takeaways

  • IP damages live in four different statutory worlds. Patents (35 U.S.C. § 284: lost profits or reasonable royalty, with a hard royalty floor and up-to-treble enhancement), copyrights (17 U.S.C. § 504: actual damages plus profits, or statutory damages of $750–$150,000 per work), trademarks (15 U.S.C. § 1117: the infringer's profits, damages, and costs, with treble/statutory weapons against counterfeiting), and trade secrets (the DTSA, 18 U.S.C. § 1836: actual loss plus unjust enrichment, or a reasonable royalty, plus up-to-double exemplary damages). Know which world you are in before you talk dollars.
  • The "average IP verdict" is a statistical mirage. The distribution is so skewed by a few giant awards that medians and distributions, not averages, tell the truth--and the typical IP recovery is far more modest than the headlines.
  • Willfulness is the great multiplier. It can treble patent damages, push copyright statutory damages to $150,000 per work, tip the equities toward trademark disgorgement after Romag, and unlock double exemplary damages and fees under the DTSA.
  • Apportionment is the great limiter. In patents, you are paid for the value of the invention, not the whole machine; in copyright and trademark profit cases, you recover only the profit attributable to the infringement. This is why headline verdicts so often shrink on appeal.
  • The damages expert is usually the decisive witness, and a successful Daubert/Rule 702 challenge can collapse a case to the statutory floor or to nothing.
  • Register copyrights early. Under 17 U.S.C. § 412, statutory damages and attorney's fees are off the table unless the work was registered before the infringement began (or within three months of publication).
  • The biggest prize may not be money at all. An injunction that ejects an infringer from the market can be worth more than any damages award--and it never appears in a "damage statistic."

Frequently Asked Questions

What is the difference between lost profits and a reasonable royalty in a patent case? Lost profits compensate the patentee for sales it would have made but for the infringement, capturing the full margin on those sales; it requires proof (typically under the Panduit factors) that the patentee would actually have captured the diverted sales. A reasonable royalty is the statutory floor under 35 U.S.C. § 284--the license fee a willing licensor and willing licensee would have agreed to in a hypothetical negotiation, evaluated through the fifteen Georgia-Pacific factors. A patentee who cannot prove lost profits is still guaranteed at least a reasonable royalty.

Can I get statutory damages for copyright infringement if I never registered my work? Generally no. Under 17 U.S.C. § 412, statutory damages and attorney's fees are available only if the work was registered before the infringement began (or, for a published work, within three months of first publication). An owner who registers only after learning of the infringement is limited to actual damages and the infringer's profits, which are often smaller and harder to prove. This is the single best reason to register promptly.

Did Romag v. Fossil mean a trademark plaintiff no longer has to prove willfulness? Not exactly. Romag Fasteners, Inc. v. Fossil, Inc., 590 U.S. 212 (2020), held that willfulness is not an absolute precondition to recovering the infringer's profits under § 1117(a) for ordinary infringement. But the Court stressed that the infringer's mental state remains "a highly important consideration" in the equitable decision whether to award profits. So willfulness is no longer a yes/no gate, but it remains a heavy factor--an innocent infringer may still avoid disgorgement, and a deliberate one is much more exposed.

How much can trade-secret damages reach under the DTSA? The Defend Trade Secrets Act (18 U.S.C. § 1836) lets a plaintiff recover actual loss plus the defendant's unjust enrichment (avoiding double-counting), or alternatively a reasonable royalty. Where the misappropriation was willful and malicious, the court may add exemplary damages of up to twice the compensatory award, plus attorney's fees. That up-to-triple total exposure is why willful trade-secret theft is among the most dangerous IP claims a defendant can face.

Why do huge IP verdicts so often get reduced? Two main reasons. First, apportionment and the entire market value rule require damages to track the value of the patented feature (or the infringing portion of profits), not the whole product, and appellate courts scrutinize big numbers hard on this point. Second, post-trial motions, remittitur, appeals, and settlements routinely shrink top-tier awards--the Apple v. Samsung saga, with multiple damages trials and a trip to the Supreme Court, is the classic example. The headline verdict is rarely the final recovery.

Are punitive damages available in IP cases? It depends on the regime. Patent law has no "punitive damages" as such, but § 284 allows enhancement up to three times the award for egregious, willful infringement (Halo Electronics v. Pulse Electronics, 579 U.S. 93 (2016)). Copyright uses elevated statutory damages (up to $150,000 per work) for willful infringement rather than punitive damages. The Lanham Act allows treble damages and statutory damages for counterfeiting. The DTSA authorizes exemplary damages up to double the compensatory amount for willful and malicious misappropriation. Each regime achieves a punitive effect through its own mechanism.

What does a "damages expert" actually do, and can the other side keep them out? A damages expert--usually an economist or forensic accountant--builds the financial model that translates infringement into dollars: the royalty base and rate, the apportionment, the lost-profit or unjust-enrichment calculation. The opposing party can challenge the expert under Federal Rule of Evidence 702 and Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993). If the court finds the methodology unreliable--non-comparable licenses, arbitrary rules of thumb, improper entire-market-value use--it can exclude the expert, which can gut the damages case.

Related Articles

This article is general information, not legal advice. IP damages turn on the specific facts, the governing statute, the forum, and developing case law; consult qualified counsel before relying on anything here for a particular matter.