Trade Secret Damages and Remedies: Unjust Enrichment, Reasonable Royalties, Injunctions, and Exemplary Awards

Summary. Trade secret cases are often won on liability and then shrunk, reversed, or retried on remedies, so the remedy theory deserves as much care as the proof of misappropriation. This article explains every remedy the Defend Trade Secrets Act and the Uniform Trade Secrets Act make available: injunctions, damages for actual loss, unjust enrichment including avoided development costs, reasonable royalties, exemplary damages of up to twice the compensatory award, attorney's fees, and interest. It walks through the leading decisions, including Syntel v. TriZetto, Epic Systems v. Tata Consultancy, Motorola Solutions v. Hytera, Texas Advanced Optoelectronic Solutions v. Renesas, the Virginia Appian v. Pegasystems reversal, and the classic head start cases, and it flags the real splits on avoided costs, burden of proof, judge versus jury, and the meaning of malice. It also covers injunction limits on employee mobility, apportionment and double counting, expert proof, criminal restitution, preemption, and the defenses that cut awards down. It closes with strategy for both sides, costs, common mistakes, two worked examples with real numbers, an FAQ, and practical takeaways.

In October 2020 a federal jury in Manhattan found that Syntel had misappropriated TriZetto's trade secrets and awarded $284.9 million in compensatory damages, plus twice that in punitive damages. Three and a half years later the liability verdict stood, a remitted punitive award stood, and TriZetto had won its attorney's fees, but its compensatory damages were zero. Nobody had decided that Syntel was innocent. TriZetto had measured its harm with a ruler the Second Circuit would not accept, and in October 2024 the district court granted a new trial just to put a compensatory number back on the board.

That sequence is the whole subject of this article in miniature. Trade secret plaintiffs spend years proving that information was secret and that the defendant took it, and then discover that the remedy (the injunction, the damages model, the multiplier, the fee award) is where the case is actually won, shrunk, or lost on appeal. Defendants who cannot win on liability often win on remedies. What follows is a working map of the remedies available under the federal Defend Trade Secrets Act (DTSA) and the state Uniform Trade Secrets Act (UTSA) statutes, the cases that tell you how courts really apply them, and the arguments each side should be ready to make.

The short answer

A court that finds misappropriation of a trade secret under 18 U.S.C. § 1836(b)(3) may award four kinds of relief, and the UTSA, adopted in some form by every state except New York, offers nearly the same menu:

1. An injunction against actual or threatened misappropriation, including orders requiring affirmative steps such as return and destruction, and, in exceptional circumstances, a "royalty injunction" that lets the defendant keep using the secret if it pays. An injunction may not stop a person from taking a job, and any conditions on employment must rest on evidence of threatened misappropriation, not merely on what the person knows.
2. Compensatory damages, measured by the plaintiff's actual loss plus any unjust enrichment (the defendant's gain) "not addressed in computing damages for actual loss," or, in lieu of those, a reasonable royalty for the unauthorized disclosure or use.
3. Exemplary damages of up to two times the compensatory award if the misappropriation was willful and malicious.
4. Attorney's fees to the prevailing party for willful and malicious misappropriation, for claims made in bad faith, and for bad faith motions to terminate injunctions.

Six practical points govern how those remedies play out:

• Every monetary measure has a clock. Damages and injunctions are generally limited to the "head start" period, the time it would have taken the defendant to get the information lawfully by reverse engineering or independent development.
• Avoided development costs are recoverable but contested. The Second Circuit refused them where the owner kept the full value of its secret and an injunction stopped further use; the Seventh and Third Circuits have affirmed them without proof of a matching loss.
• Double counting is the most common reversible error. Lost profits and the defendant's profits on the same sale cannot both be recovered.
• Apportionment matters. A defendant's profits must be tied to the secret, not to everything the defendant did, and courts disagree about who bears that burden.
• Who decides depends on the measure. Lost profits and royalties go to the jury; a growing line of authority treats disgorgement of the defendant's profits as equitable relief for the judge.
• Injunctions are not automatic. The four-factor test of eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006), applies, and most circuits refuse to presume irreparable harm.

The legal framework: the DTSA, the UTSA, and the states that changed the rules

The DTSA remedies provision, clause by clause

The federal statute is short enough to quote nearly in full, and the precise words matter in nearly every remedies fight. Under 18 U.S.C. § 1836(b)(3), the court may:

• (A)(i) grant an injunction "to prevent any actual or threatened misappropriation" on reasonable terms, provided the order does not "(I) prevent a person from entering into an employment relationship," with conditions on employment "based on evidence of threatened misappropriation and not merely on the information the person knows," or "(II) otherwise conflict with an applicable State law prohibiting restraints on the practice of a lawful profession, trade, or business";
• (A)(ii) require "affirmative actions to be taken to protect the trade secret";
• (A)(iii) "in exceptional circumstances that render an injunction inequitable," condition future use "upon payment of a reasonable royalty for no longer than the period of time for which such use could have been prohibited";
• (B)(i) award "(I) damages for actual loss caused by the misappropriation of the trade secret; and (II) damages for any unjust enrichment caused by the misappropriation of the trade secret that is not addressed in computing damages for actual loss," or (B)(ii) "in lieu of damages measured by any other methods," a reasonable royalty;
• (C) award exemplary damages "in an amount not more than 2 times the amount of the damages awarded under subparagraph (B)" if the secret "is willfully and maliciously misappropriated"; and
• (D) award reasonable attorney's fees to the prevailing party if a claim "is made in bad faith, which may be established by circumstantial evidence," if a motion to terminate an injunction "is made or opposed in bad faith," or if the misappropriation was willful and malicious.

Four neighboring provisions shape the remedy picture. The limitations period is three years from actual or constructive discovery, and "a continuing misappropriation constitutes a single claim," so the clock does not restart with each new use (18 U.S.C. § 1836(d)). The DTSA does not preempt state law (18 U.S.C. § 1838), which is why nearly every complaint pleads both. An employer that fails to give employees the whistleblower immunity notice required by 18 U.S.C. § 1833(b) cannot recover exemplary damages or attorney's fees against an employee who did not receive it. And the definitions in 18 U.S.C. § 1839 exclude "reverse engineering, independent derivation, or any other lawful means of acquisition" from improper means, which is the doctrinal root of the head start limit.

The Uniform Trade Secrets Act, sections 2 through 4

The UTSA, as amended in 1985, supplies the template the DTSA copied. Section 2(a) allows actual or threatened misappropriation to be enjoined and directs that an injunction "shall be terminated when the trade secret has ceased to exist," while permitting it to continue "for an additional reasonable period of time in order to eliminate commercial advantage that otherwise would be derived from the misappropriation." Section 2(b) authorizes royalty injunctions in "exceptional circumstances," which "include, but are not limited to, a material and prejudicial change of position prior to acquiring knowledge or reason to know of misappropriation." Section 3(a) provides for actual loss plus unjust enrichment not counted in actual loss, or a reasonable royalty "in lieu of damages measured by any other methods," and section 3(b) caps exemplary damages at twice the section 3(a) award. Section 4 authorizes fees on the same three triggers the DTSA later adopted. Section 7 displaces conflicting state tort and restitution claims for misappropriation while preserving contract claims and claims not based on misappropriation.

State variations that change the numbers

The model act is uniform; the enacted versions are not. The differences below routinely move verdicts by eight or nine figures, so check the governing statute before you build a damages model.

Two treatise references are worth keeping on the desk: 4 Roger M. Milgrim, Milgrim on Trade Secrets § 15.02 (monetary relief) and the Restatement (Third) of Unfair Competition §§ 44 and 45 (1995), which courts quote constantly. Section 45 directs a "comparative appraisal of all the factors of the case," including the certainty of the proof of loss or gain, "the nature and extent of the appropriation," "the relative adequacy to the plaintiff of an injunction and of other remedies," the defendant's intent and good faith reliance, the plaintiff's delay, and any misconduct by the plaintiff. Comment f to section 45 also puts on the defendant the burden of proving deductions from its sales, a rule that has turned into one of the sharpest splits in the field.

The leading cases, explained

Eight decisions do most of the work in modern trade secret remedies arguments. Read them for the numbers as much as the holdings; each one shows a specific way a damages model succeeds or fails.

Syntel v. TriZetto: avoided costs need a matching harm

In Syntel Sterling Best Shores Mauritius Ltd. v. TriZetto Group, Inc., 68 F.4th 792 (2d Cir. 2023), Syntel, a former TriZetto subcontractor, kept using TriZetto's Facets-related trade secrets to perform services for a health insurer after it terminated the parties' services agreement. TriZetto's expert offered three numbers: about $285 million in avoided development costs (TriZetto's own research and development spend on the misappropriated material), a royalty set at half of that, and $8.5 million in TriZetto's lost profits. The jury awarded the $285 million under the DTSA. The Second Circuit affirmed liability but vacated the damages judgment, holding that avoided costs were unavailable "in this specific case." The court asked whether Syntel's misappropriation "injure[d] TriZetto beyond its actual loss of $8.5 million in lost profits" and answered no: Syntel earned only about $823,899 in profits from the misuse, TriZetto kept the full use and value of its secrets, Facets was worth more than ever, and a permanent injunction ended Syntel's use. Awarding the owner's entire development cost on those facts would be "more punitive than compensatory." The court took care to say that avoided costs remain available where misappropriation destroys or diminishes a secret, where the defendant uses it to build a competing product, or where the defendant cannot be stopped from profiting.

Epic v. Tata and PPG v. Jiangsu Tie Mao: the other side of the split

The Seventh Circuit sees avoided costs more generously. In Epic Systems Corp. v. Tata Consultancy Services Ltd., 980 F.3d 1117 (7th Cir. 2020), applying Wisconsin's UTSA, Tata's consultants downloaded thousands of Epic documents and built a "comparative analysis" spreadsheet that Tata used in an attempt to enter the U.S. electronic health records market and to identify gaps in its own Med Mantra product. The jury awarded $140 million for the benefit Tata derived from the comparative analysis, $100 million for "other uses," and $700 million in punitive damages. The court affirmed the $140 million as the value of a "head start," measured by avoided research and development costs. The method is instructive: Epic's expert started from Epic's roughly $306 million cost to develop the information that made its way into the spreadsheet, subtracted coding costs that conferred no benefit on Tata, adjusted for technology decay to about $200 million, and the jury then applied the defense expert's own 30 percent labor-cost discount for Tata's Indian engineers to land exactly on $140 million. The $100 million "other uses" award, untethered to any specific use, stayed vacated. The Third Circuit went further in PPG Industries, Inc. v. Jiangsu Tie Mao Glass Co., 47 F.4th 156 (3d Cir. 2022), affirming an $8.8 million avoided-cost award under Pennsylvania's UTSA even though the plaintiff showed no actual loss and an injunction barred future use. Syntel expressly declined to follow both decisions. That is a genuine circuit split, and it matters most in cases where the plaintiff's own losses are small.

Motorola v. Hytera: the larger of two paths, and a DTSA multiplier that survived

Motorola Solutions, Inc. v. Hytera Communications Corp., 108 F.4th 458 (7th Cir. 2024), involved what the court called "a large and blatant theft": Hytera hired three Motorola engineers in Malaysia who downloaded more than 10,000 technical documents and source code, and Hytera then sold radios "functionally indistinguishable" from Motorola's worldwide. The jury awarded $764.6 million. The district court treated the amount of unjust enrichment as an equitable question for the court, treated the jury's figures as advisory, and found that the jury had double counted Hytera's profits and its avoided research and development costs. It subtracted $73.6 million of avoided costs from the $209.4 million DTSA award, leaving $135.8 million in unjust profits, and reset punitive damages at twice that, $271.6 million. It denied a permanent injunction and instead ordered an ongoing royalty of 100 percent of Hytera's profits on the infringing products. The Seventh Circuit affirmed the DTSA compensatory and punitive awards and made three holdings that matter beyond the case. First, the DTSA reaches foreign sales when "an act in furtherance of the offense was committed in the United States" under 18 U.S.C. § 1837. Second, a DTSA plaintiff is entitled to the larger of two paths: the defendant's total profits from the theft as unjust enrichment, which the defendant may reduce by proving its own contributions, or the plaintiff's actual losses plus any unjust enrichment not counted in them (here $86.2 million in lost profits plus $73.6 million in avoided costs, $159.8 million), and because the second path exceeded the award, any failure to apportion was harmless. Third, the DTSA's two-times cap on exemplary damages comports with due process. The court also told the district court to take a fresh look at a permanent injunction in light of Hytera's failure to pay the royalty.

TAOS v. Renesas: the head start limits money too, and disgorgement goes to the judge

In Texas Advanced Optoelectronic Solutions, Inc. v. Renesas Electronics America, Inc., 895 F.3d 1304 (Fed. Cir. 2018), applying Texas law, the jury awarded disgorgement of Intersil's profits on sales from 2006 through 2014 for misappropriation of a light-sensor design. The Federal Circuit vacated the award for two independent reasons. The design had become accessible by proper means (Intersil reverse-engineered the product no later than January 2006), and secrecy protection ended when the head start period after that lawful access ran out, yet more than 90 percent of the award came from sales years later. And there was no Seventh Amendment right to have a jury decide disgorgement of the defendant's profits, an equitable remedy, at least where the disgorgement claim was not a proxy for the plaintiff's own losses. On remand any "sales-based monetary relief" required findings on when the secret became properly accessible and how long the head start lasted.

Steves & Sons v. JELD-WEN: overclaiming and the meaning of malice

Steves & Sons, Inc. v. JELD-WEN, Inc., 988 F.3d 690 (4th Cir. 2021), is mostly remembered as an antitrust divestiture case, but its trade secret counterclaim teaches two lessons. JELD-WEN asserted 67 trade secrets; the jury found only eight protectable and misappropriated, found the misappropriation not willful and malicious, and awarded $1.2 million. And the court declined to disturb an instruction that "malicious" under the DTSA requires an "intent to cause injury or harm," noting that courts interpreting the UTSA split between that standard and "conscious disregard" of the owner's rights.

Appian v. Pegasystems: two billion dollars and a burden-shifting instruction

The largest trade secret verdict in Virginia history, $2,036,860,045 against Pegasystems for using a contractor to obtain access to a competitor's software platform, was reversed in Pegasystems Inc. v. Appian Corp., 81 Va. App. 433 (2024). The trial court had told the jury, following comment f to Restatement section 45, that Appian need prove only Pega's sales, after which Pega bore the burden of proving which sales were not attributable to the trade secrets. Appian's counsel argued accordingly: "We just got to show all the money that flowed in, the 6 billion plus." The Court of Appeals held that the instruction relieved Appian of its burden to prove that misappropriation proximately caused the unjust enrichment, and that the Virginia statute's text did not support burden-shifting; it also found evidentiary errors and ordered a new trial. The Supreme Court of Virginia affirmed in Appian Corp. v. Pegasystems Inc., Record No. 240736 (Va. Jan. 8, 2026), holding that nothing in the Virginia statute displaces the common-law rule that the plaintiff carries the burden of proving its damages, so no part of that burden shifts to the defendant, while also holding the liability evidence sufficient to reach a jury. The case goes back for a new trial. Its lesson travels well beyond Virginia: a plaintiff allowed to start from total revenue can turn a narrow misappropriation into a bet-the-company verdict, and a defendant that loses the burden argument may spend the trial disproving profits it never made from the secret.

University Computing v. Lykes-Youngstown: the flexible measure

The modern law of trade secret damages still runs through University Computing Co. v. Lykes-Youngstown Corp., 504 F.2d 518 (5th Cir. 1974). The defendants misappropriated a computerized inventory system intending to market it and failed. The Fifth Circuit held that the defendant must have put the secret to some commercial use, but that failure to profit is no defense: "the risk of defendants' venture, using the misappropriated secret, should not be placed on the injured plaintiff." When neither the plaintiff's loss nor the defendant's profit can be measured, the factfinder may set a reasonable royalty based on what the parties would have agreed to at the time of misappropriation, considering the resulting changes in competitive posture, prices paid by past licensees, the secret's total value to the owner including development costs, the nature and extent of the defendant's intended use, and other factors such as available alternatives. Later courts call this the "flexible and imaginative" approach.

Winston Research and K-2 Ski: the head start injunction

Winston Research Corp. v. Minnesota Mining & Manufacturing Co., 350 F.2d 134 (9th Cir. 1965), rejected both extremes then on offer: the Shellmar rule of a perpetual injunction against a faithless confidant even after public disclosure, from Shellmar Products Co. v. Allen-Qualley Co., 87 F.2d 104 (7th Cir. 1936), and the Conmar rule of no injunction once the secret becomes public, from Conmar Products Corp. v. Universal Slide Fastener Co., 172 F.2d 150 (2d Cir. 1949). The court affirmed a two-year injunction measured by "the approximate period it would require a legitimate Mincom competitor to develop a successful machine after public disclosure," which "denied the employees any advantage from their faithlessness" while imposing "the minimum restraint." K-2 Ski Co. v. Head Ski Co., 506 F.2d 471 (9th Cir. 1974), refined the measure to the time it would have taken the defendant, "either by reverse engineering or by independent development," to develop the product legitimately, and remanded for the trial court to consider credit for 27 months the defendant had already spent under a preliminary injunction.

Actual loss: lost profits, price erosion, and the destroyed secret

Actual loss is the owner's injury, and it is a legal remedy tried to a jury. The usual components are the owner's lost profits on sales diverted to the defendant, price erosion when the owner had to cut prices to compete with a product built on its own secret, increased costs such as marketing spend to win back customers, and, where the secret has been destroyed by disclosure, the lost value of the secret itself. The Sixth Circuit put the default rule plainly in an unpublished decision often quoted on this point: "When the misappropriated trade secret is used to field competing products, the best measure of damages is the plaintiff's lost profits or the defendant's illicit gains." Avery Dennison Corp. v. Four Pillars Enterprise Co., 45 F. App'x 479 (6th Cir. 2002).

Lost profits need proof of causation (the sale would have been the owner's but for the misappropriation), capacity, and margin. Patent law's four-factor test for lost profits from Panduit Corp. v. Stahlin Bros. Fibre Works, Inc., 575 F.2d 1152 (6th Cir. 1978) (demand, absence of acceptable noninfringing substitutes, capacity, and the amount of profit) is a useful discipline even though trade secret law does not require it. Patent cases also supply the authority for price erosion and increased marketing costs as recoverable losses, Lam, Inc. v. Johns-Manville Corp., 718 F.2d 1056 (Fed. Cir. 1983), and for the role of available but unused alternatives in testing causation, Grain Processing Corp. v. American Maize-Products Co., 185 F.3d 1341 (Fed. Cir. 1999).

Two trade secret decisions show the method. In Sokol Crystal Products, Inc. v. DSC Communications Corp., 15 F.3d 1427 (7th Cir. 1994), the expert computed a per-unit margin of about $23.50 (an $11.50 cost against a $35 market price), projected unit sales over eight years with a decline as technology moved on, discounted at 6.5 percent, and arrived at $5.1 million; the jury awarded $2.49 million, and the Seventh Circuit affirmed, explaining that head start limits apply to lost profits because once the defendant "would have discovered" the secret lawfully, later losses are not caused by the misappropriation. And in Wellogix, Inc. v. Accenture, L.L.P., 716 F.3d 867 (5th Cir. 2013), the Fifth Circuit affirmed a $26.2 million compensatory award built on a valuation of the plaintiff's business before and after the misappropriation, along with punitive damages the district court had remitted from $68.2 million to $18.2 million.

When the secret is destroyed (published on the internet, disclosed in a patent filing, or spread so widely it is no longer secret), the owner's loss is the value of the secret, measured the way a willing buyer would value it. The Fifth Circuit's formulation, from Precision Plating & Metal Finishing, Inc. v. Martin-Marietta Corp., 435 F.2d 1262 (5th Cir. 1970), is the value "a reasonably prudent investor" would have paid for the secret. Syntel's footnote 34 identifies exactly this scenario as the strongest case for awarding the owner's development costs.

Unjust enrichment: profits, avoided costs, and the double counting trap

Unjust enrichment measures the defendant's gain rather than the owner's loss. It comes in two main forms: the defendant's profits from sales made possible by the secret, and the defendant's savings, usually the research and development costs it avoided by taking a shortcut. The Restatement (Third) of Unfair Competition § 45 cmt. c recognizes both, and the Restatement (Third) of Restitution and Unjust Enrichment § 1 cmt. d adds the useful observation that "a saved expenditure . . . is no less beneficial to the recipient than a direct transfer."

Profits, and the apportionment fight

A defendant's profits are recoverable only to the extent they are attributable to the secret. Patent law has required apportionment since Garretson v. Clark, 111 U.S. 120 (1884); copyright law adopted it in Sheldon v. Metro-Goldwyn Pictures Corp., 309 U.S. 390 (1940); and Motorola applied the copyright causation framework of 17 U.S.C. § 504(b) to DTSA unjust enrichment, describing a path on which "once the plaintiff proves the defendant's total profits from the theft, the defendant has an opportunity to seek apportionment by proving how its own efforts contributed to those profits." That allocation tracks Restatement § 45 cmt. f and the old commingling rule of Westinghouse Electric & Manufacturing Co. v. Wagner Electric & Manufacturing Co., 225 U.S. 604 (1912), which puts the cost of an inseparable mixture on the wrongdoer.

The Virginia Court of Appeals in Pegasystems v. Appian rejected that allocation under Virginia's statute, holding that the plaintiff must prove that the misappropriation proximately caused the specific sales it claims. Defendants elsewhere will cite Pegasystems and the Supreme Court's insistence in Liu v. SEC, 591 U.S. 71 (2020), that equitable disgorgement be limited to net profits from the wrongdoing after legitimate expenses are deducted. Plaintiffs will cite Motorola, the Restatement, and Bigelow v. RKO Radio Pictures, Inc., 327 U.S. 251 (1946), and Story Parchment Co. v. Paterson Parchment Paper Co., 282 U.S. 555 (1931), for the principle that the wrongdoer bears the risk of the uncertainty its own wrong created. Even a court that shifts the burden of proving deductions will not let a plaintiff stop at total revenue for a product in which the secret played a minor role.

The Ninth Circuit's handling of the Bratz litigation shows how far apportionment reaches into equitable relief. In Mattel, Inc. v. MGA Entertainment, Inc., 616 F.3d 904 (9th Cir. 2010), the district court imposed a constructive trust transferring MGA's entire Bratz trademark portfolio to Mattel on the theory that the doll line began with two names a Mattel employee conceived. The Ninth Circuit vacated, holding that it was "not equitable to transfer this billion dollar brand" when its value was "overwhelmingly the result of MGA's legitimate efforts," and quoting Dobbs for the principle that restitution must recapture the wrongful profit "without capturing the fruits of the defendant's own labors or legitimate efforts." Dan B. Dobbs, Law of Remedies § 6.6(3) (2d ed. 1993).

Avoided development costs

Avoided costs are the classic measure when the defendant used the secret to accelerate its own development but made little or no profit. GlobeRanger Corp. v. Software AG United States of America, Inc., 836 F.3d 477 (5th Cir. 2016), affirmed a $15 million award against a defendant that earned only $860,000 on the project in question, holding that the plaintiff's own development costs "can be a proxy for the costs that the defendant saved" and that "the wrongdoer should not benefit from hindsight perspective that its gamble of misappropriating the trade secret turned out not to be so profitable." The jury there awarded about $5 million less than the expert's $19.7 million figure, and rejected the defendant's claim that it could have built the technology for $140,000. That gap between the expert's number and the verdict is typical, and it is why both sides need a credible alternative number in front of the jury.

The limits come from Syntel: avoided costs must reflect a benefit gained "at the expense of" the owner, and the comparative appraisal of Restatement § 45 matters. Expect defendants to argue that the owner kept the secret, that an injunction stops further use, that the defendant never built a product, and that the owner's development cost overstates what a competitor would have spent (Tata's 30 percent labor discount in Epic is the model). Expect plaintiffs to argue diminished value, a competing product, and benefits the injunction cannot claw back, such as a market entry already made. In New York state law claims, avoided costs are off the table under E.J. Brooks.

Double counting: "not addressed in computing damages for actual loss"

The statutory phrase is a prohibition on double recovery. The cleanest illustration is Salsbury Laboratories, Inc. v. Merieux Laboratories, Inc., 908 F.2d 706 (11th Cir. 1990). The trial court awarded Salsbury $1 million as the defendants' savings in research, development, and marketing (discounted from about $3 million that Salsbury had spent) and also $52,000 of profits on the competing vaccine. The Eleventh Circuit struck the $52,000: had the defendant spent what the shortcut saved it, it would have made no profit at all, so awarding both counted the same benefit twice. Motorola applied the same logic to Hytera's profits and avoided costs, and the Seventh Circuit's earlier copyright decision in Taylor v. Meirick, 712 F.2d 1112 (7th Cir. 1983), quoted there, explains the owner's side of the equation: a plaintiff who recovers its own lost profits can add only the part of the defendant's profits it would not have earned itself. Build the damages model so that every dollar is assigned to one bucket, and make the verdict form do the same.

Reasonable royalties: pricing a license nobody would have granted

A royalty is the fallback measure and sometimes the best one. Under the DTSA and the UTSA it is available "in lieu of" other measures; in California, under Ajaxo, a court may award it when neither actual loss nor unjust enrichment is "provable," which includes the case in which a jury finds the defendant made no profit. Ajaxo Inc. v. E*Trade Financial Corp., 187 Cal. App. 4th 1295 (2010). In Virginia it is available only if the plaintiff cannot prove a greater amount another way.

Courts build the royalty from a hypothetical negotiation at the time misappropriation began, using factors drawn from University Computing and from the patent factors of Georgia-Pacific Corp. v. United States Plywood Corp., 318 F. Supp. 1116 (S.D.N.Y. 1970). The Second Circuit described the exercise in Vermont Microsystems, Inc. v. Autodesk, Inc., 88 F.3d 142 (2d Cir. 1996), as a "suppositious meeting between the parties" to determine "a fair licensing price," and after remand affirmed a royalty because the evidence of the defendant's profits was too imprecise, Vermont Microsystems, Inc. v. Autodesk, Inc., 138 F.3d 449 (2d Cir. 1998). Real-world anchors carry the day:

• In Sikes v. McGraw-Edison Co., 665 F.2d 731 (5th Cir. 1982), a $900,000 verdict on 1,536,226 string trimmers worked out to 59 cents per unit, against the inventor's own 50-cent proposal that the defendant had called "about right" and expert opinions ranging from ten cents to over a dollar. The Fifth Circuit affirmed.
• In Bohnsack v. Varco, L.P., 668 F.3d 262 (5th Cir. 2012), deal terms the parties had negotiated over years of testing supported a $600,000 award as the value to "a reasonably prudent investor."
• In Learning Curve Toys, Inc. v. PlayWood Toys, Inc., 342 F.3d 714 (7th Cir. 2003), the Seventh Circuit reinstated a jury verdict awarding the inventor of a clicking toy train track a percentage royalty and remanded for a jury trial on exemplary damages.

The Texas Supreme Court's warning in Southwestern Energy Production Co. v. Berry-Helfand, 491 S.W.3d 699 (Tex. 2016), belongs in every royalty brief. The expert there applied a flat 3 percent overriding royalty drawn from an earlier deal that actually used a sliding scale. A license fee of $2 per widget, the court explained, is strong evidence of a $2 royalty for the same widgets, but an average of $1, $2, and $3 fees for small, medium, and large widgets says little about a defendant that made only small ones. "[R]elying on imagination is not justified when objective evidence is available." Patent law offers the same lessons with more appellate polish: comparable licenses must be adjusted for differences in what was licensed, Lucent Technologies, Inc. v. Gateway, Inc., 580 F.3d 1301 (Fed. Cir. 2009), and Ericsson, Inc. v. D-Link Systems, Inc., 773 F.3d 1201 (Fed. Cir. 2014); rules of thumb such as the 25 percent rule are not evidence, Uniloc USA, Inc. v. Microsoft Corp., 632 F.3d 1292 (Fed. Cir. 2011); and the royalty base must be apportioned to the protected feature, VirnetX, Inc. v. Cisco Systems, Inc., 767 F.3d 1308 (Fed. Cir. 2014), and LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51 (Fed. Cir. 2012). For the full patent treatment, see Patent Damages: Reasonable Royalties, Lost Profits, Apportionment, and Enhancement.

Head start: the clock that runs on every remedy

Trade secret law protects against improper acquisition, not against the information itself, so every remedy is bounded by the time the defendant could have obtained the information honestly. That bounded period is the head start. Section 2(a) of the UTSA builds it into injunctions; the Restatement (Third) of Unfair Competition § 44 cmt. f limits injunctions to the time needed to eliminate the commercial advantage; Sokol and TAOS apply it to money. Computing it takes four inputs:

1. The lawful availability date. When did the information become public, or when could the defendant have obtained it by reverse engineering a product on the market? A published patent application ends secrecy, as Attia v. Google LLC, 983 F.3d 420 (9th Cir. 2020), held for information "completely parallel" to what the plaintiff's own applications disclosed. A product release may start the reverse engineering clock, as it did in TAOS.
2. The legitimate development time. How long would a competent competitor need, starting from the lawful information, to reach the same point? Use the defendant's actual development history (in Winston the defendant took fourteen months with the stolen information, which the court considered a fair measure), the plaintiff's own development time, and expert testimony on reverse engineering effort.
3. Credits and extensions. Subtract time already spent under a preliminary injunction that served the same purpose (K-2 Ski). Consider extending for delays the defendant caused, such as poaching the plaintiff's key engineers, which the district court in Winston used to lengthen the injunction (the Ninth Circuit did not decide whether that was proper because the issue was not raised).
4. The defendant who never would have gotten there. If the evidence shows the defendant could not have developed the technology at all without the theft, there is no head start cap. The Sokol jury "may well have concluded" exactly that.

Courts do not demand mathematical precision. In Lamb-Weston, Inc. v. McCain Foods, Ltd., 941 F.2d 970 (9th Cir. 1991), the Ninth Circuit affirmed an eight-month, worldwide preliminary injunction against a competitor that had obtained a confidential patent application for a curly-fry cutting blade, holding that the trial court's statement that the period corresponded to the defendant's head start satisfied Fed. R. Civ. P. 52(a), while encouraging more specific findings.

Injunctions: eBay, irreparable harm, and the employment proviso

The traditional test applies

The DTSA says a court "may" grant an injunction, and the Supreme Court reads that word as preserving equitable discretion rather than mandating relief. eBay rejected categorical rules for patent injunctions and required the plaintiff to show irreparable injury, inadequacy of legal remedies, a favorable balance of hardships, and no disservice to the public interest. Winter v. Natural Resources Defense Council, Inc., 555 U.S. 7 (2008), requires a likelihood (not a possibility) of irreparable harm for preliminary relief; Monsanto Co. v. Geertson Seed Farms, 561 U.S. 139 (2010), calls an injunction "a drastic and extraordinary remedy"; and Starbucks Corp. v. McKinney, 602 U.S. 339 (2024), reaffirmed that courts must apply the traditional four factors absent a clear statutory command otherwise. The older cases make the same point about statutory injunctions generally: Hecht Co. v. Bowles, 321 U.S. 321 (1944); Weinberger v. Romero-Barcelo, 456 U.S. 305 (1982); Amoco Production Co. v. Village of Gambell, 480 U.S. 531 (1987).

Whether harm is presumed is the practical question. The Second Circuit's answer in Faiveley Transport Malmo AB v. Wabtec Corp., 559 F.3d 110 (2d Cir. 2009), is no presumption where a misappropriator "seeks only to use those secrets" in pursuit of profit, "without further dissemination or irreparable impairment of value," because "an award of damages will often provide a complete remedy for such an injury"; a rebuttable presumption "might be warranted" where the defendant is likely to disseminate the secret to a wider audience. Salinger v. Colting, 607 F.3d 68 (2d Cir. 2010), applied eBay to preliminary injunctions in copyright cases, reasoning that its principles are not limited to patents. Plaintiffs therefore win injunctions by proving concrete risks money cannot fix: dissemination to third parties, destruction of the secret's value, loss of customer relationships that cannot be quantified, and a defendant who cannot or will not pay a judgment. See Preliminary Injunctions and Temporary Restraining Orders: Emergency Relief in Federal Court.

What courts actually enjoin, and what they will not

Good injunctions are specific. Fed. R. Civ. P. 65(d) requires every order to "state its terms specifically" and describe the restrained acts "in reasonable detail." Winston struck provisions enjoining use of "knowledge of what not to do" because their only practical enforcement would bar former employees from working in the field at all. The Third Circuit's decision in Mallet & Co. v. Lacayo, 16 F.4th 364 (3d Cir. 2021), vacated a preliminary injunction that barred a competitor from making or selling any competing release agents and barred an employee who had no noncompete from working for any competitor. The court held that a party seeking an injunction must identify its trade secrets with specificity, that it "would take a truly extraordinary showing" to justify "an order ejecting a competitor from the marketplace altogether," and that the bond under Fed. R. Civ. P. 65(c) must be tied to the scope of the injunction through a case-specific analysis of the harm a wrongful injunction would cause, which the district court had not done when it set a $500,000 bond against the defendants' request for more than $21.5 million. The orders that hold up on appeal usually include a use and disclosure ban limited to identified secrets, return and destruction with forensic verification (an "affirmative action" under § 1836(b)(3)(A)(ii)), a time limit tied to the head start, and, where justified by evidence, narrow role restrictions.

The employment proviso and inevitable disclosure

Under § 1836(b)(3)(A)(i)(I), a DTSA injunction may not "prevent a person from entering into an employment relationship," and any conditions on employment must be "based on evidence of threatened misappropriation and not merely on the information the person knows." That language was a deliberate response to the inevitable disclosure doctrine of PepsiCo, Inc. v. Redmond, 54 F.3d 1262 (7th Cir. 1995), in which the Seventh Circuit affirmed an Illinois-law injunction barring a former PepsiCo executive from taking up his position at Quaker for roughly six months, and permanently from using PepsiCo secrets, because he could not help but rely on PepsiCo's strategic plans and had shown a "lack of forthrightness." State law still varies widely. Pennsylvania courts may enjoin new employment on a showing of substantial threat, as the Third Circuit held in Bimbo Bakeries USA, Inc. v. Botticella, 613 F.3d 102 (3d Cir. 2010); California rejects inevitable disclosure, Whyte v. Schlage Lock Co., 101 Cal. App. 4th 1443 (2002), and voids most noncompetes; Texas's 2017 amendments bar injunctions against using "general knowledge, skill, and experience." Under § 1836(b)(3)(A)(i)(II), a DTSA injunction may not conflict with those state restraint-of-trade rules. For the state-by-state picture, see Non-Compete Agreements Under Siege: FTC Rulemaking and State Law Developments.

Royalty injunctions and ongoing royalties

Both statutes allow a court, in exceptional circumstances that make a prohibitory injunction inequitable, to let the defendant continue using the secret on payment of a royalty, but only "for no longer than the period of time for which such use could have been prohibited." The classic exceptional circumstance is the good faith defendant who invested heavily before learning the information was misappropriated. That idea is old: in Conmar, Judge Learned Hand denied relief against a new employer that had committed $40,000 to the venture before receiving notice. Motorola shows the modern version, and its risk: the district court substituted an ongoing royalty of 100 percent of profits for an injunction, Hytera dragged its feet in paying, and the Seventh Circuit sent the injunction question back. A plaintiff facing a royalty-instead-of-injunction proposal should demand security and a self-executing default provision; a defendant offering one should be prepared to pay on time.

Enforcement and appeal

Injunctions bind the parties, their officers, agents, employees, and attorneys, and "other persons who are in active concert or participation" with them who receive actual notice. Fed. R. Civ. P. 65(d)(2). In ClearOne Communications, Inc. v. Bowers, 651 F.3d 1200 (10th Cir. 2011), the Tenth Circuit upheld civil contempt against a nonparty who helped move enjoined assets and technology into a sham successor company. Orders granting or refusing injunctions are immediately appealable under 28 U.S.C. § 1292(a)(1), and stays pending appeal turn on the factors in Nken v. Holder, 556 U.S. 418 (2009), under Fed. R. App. P. 8. One limit to remember: under Grupo Mexicano de Desarrollo, S.A. v. Alliance Bond Fund, Inc., 527 U.S. 308 (1999), a federal court generally cannot freeze a defendant's assets before judgment to secure a claim for money damages, so asset restraints in trade secret cases must rest on an equitable claim to specific property, such as identifiable proceeds of the misappropriation. For contempt practice, see Contempt and Enforcing Injunctions: Civil, Criminal, and What an Order Actually Compels.

Exemplary damages, attorney's fees, and interest

"Willful and malicious," and who decides

Both statutes cap exemplary damages at twice the compensatory award, so the maximum monetary exposure is three times compensatory damages before fees. The DTSA does not define "willful and malicious," and courts split on the meaning of malice. One line requires an intent to cause injury or harm, the standard the Fourth Circuit left undisturbed in Steves and the one supported by 1 Milgrim on Trade Secrets § 1.01. The other line asks whether the defendant acted intentionally or with "conscious disregard of the rights of another," the Illinois standard applied in Mangren Research & Development Corp. v. National Chemical Co., 87 F.3d 937 (7th Cir. 1996), which affirmed exemplary damages of exactly twice a $252,684.69 compensatory award, and in Learning Curve. Texas now codifies the conscious-disregard formulation by statute. Evidence that moves factfinders is concrete: concealment, deleted files, false testimony, recruiting employees specifically to bring information, and continued use after a demand letter.

The DTSA says "the court may award" exemplary damages, while Texas assigns them to the fact finder and the Illinois case law in Learning Curve sent them to a jury; check the governing statute and preserve any jury demand under Fed. R. Civ. P. 38. The whistleblower notice requirement can take exemplary damages and fees off the table against an employee who never received notice, so audit the employment agreements before promising a client a multiplier.

Due process limits

Statutory caps do most of the work, but state common-law punitive damages (New York, and the Wisconsin claim in Epic) are not capped by the UTSA multiplier, and constitutional review then matters. The guideposts come from BMW of North America, Inc. v. Gore, 517 U.S. 559 (1996): reprehensibility, the ratio to compensatory damages, and comparable civil penalties. State Farm Mutual Automobile Insurance Co. v. Campbell, 538 U.S. 408 (2003), added that few awards exceeding a single-digit ratio will satisfy due process and that when compensatory damages are substantial, a ratio of about one to one may be the outer limit. Appellate courts review the constitutional question de novo under Cooper Industries, Inc. v. Leatherman Tool Group, Inc., 532 U.S. 424 (2001). Earlier decisions, Pacific Mutual Life Insurance Co. v. Haslip, 499 U.S. 1 (1991), and TXO Production Corp. v. Alliance Resources Corp., 509 U.S. 443 (1993), and the maritime one-to-one ceiling of Exxon Shipping Co. v. Baker, 554 U.S. 471 (2008), round out the framework.

The trade secret applications are instructive. In Epic, the jury's $700 million punitive award was reduced to $280 million under Wisconsin's statutory cap, and the Seventh Circuit cut it to $140 million, a one-to-one ratio, because the $140 million compensatory award was itself substantial. In Motorola, the same court held that the DTSA's two-to-one cap "functions like a host of other federal statutes authorizing double or treble damages" and affirmed $271.6 million. Hallmark Cards, Inc. v. Monitor Clipper Partners, LLC, 758 F.3d 1051 (8th Cir. 2014), affirmed $10 million in punitive damages on a $21.3 million compensatory award for a private equity firm's use of Hallmark's market research, emphasizing repeated concealment. And in Wellogix the trial court remitted a $68.2 million punitive award to the $18.2 million the plaintiff had asked for, which the plaintiff accepted.

Attorney's fees

Fees follow three triggers: willful and malicious misappropriation, a claim made in bad faith, and a bad faith motion to terminate (or opposition to terminating) an injunction. The fee provision cuts both ways, and it is the defendant's best deterrent against suits filed to slow down a departing employee or a new competitor. California courts applying the parallel CUTSA provision ask whether the claim was objectively specious and pursued in subjective bad faith, FLIR Systems, Inc. v. Parrish, 174 Cal. App. 4th 1270 (2009); federal courts look for similar evidence under the DTSA's "circumstantial evidence" language. "Prevailing party" is a question of federal law: a plaintiff's voluntary dismissal without prejudice does not make the defendant a prevailing party, Dunster Live, LLC v. LoneStar Logos Management Co., 908 F.3d 948 (5th Cir. 2018), but a defendant need not win on the merits if the plaintiff's claim is rebuffed, CRST Van Expedited, Inc. v. EEOC, 578 U.S. 419 (2016). The amount is a lodestar adjusted for the degree of success, Hensley v. Eckerhart, 461 U.S. 424 (1983), and Perdue v. Kenny A. ex rel. Winn, 559 U.S. 542 (2010). Fee motions are due 14 days after judgment under Fed. R. Civ. P. 54(d)(2) unless a statute or order provides otherwise. The Syntel district court granted TriZetto's fee motion on remand on the strength of the jury's willful and malicious finding even while compensatory damages stood at zero. Without a statute or contract, there are no fees: the Ninth Circuit reversed a fee award in K-2 Ski, a pre-UTSA case, under the American rule. Patent and trademark practice supply useful analogies for what makes a case stand out, Octane Fitness, LLC v. ICON Health & Fitness, Inc., 572 U.S. 545 (2014), and for enhanced awards reserved for egregious misconduct, Halo Electronics, Inc. v. Pulse Electronics, Inc., 579 U.S. 93 (2016). For the mechanics, see Preparing and Opposing a Fee Petition: A Practical Guide.

Prejudgment and post-judgment interest

The DTSA says nothing about prejudgment interest, so federal courts exercise discretion, generally awarding it on compensatory damages to make the plaintiff whole for the lost use of money, by analogy to cases such as General Motors Corp. v. Devex Corp., 461 U.S. 648 (1983) (patent damages), and City of Milwaukee v. Cement Division, National Gypsum Co., 515 U.S. 189 (1995). State law usually governs interest on state claims, and the amounts are not trivial: the Texas judgment later reversed in Title Source, Inc. v. HouseCanary, Inc., No. 04-19-00044-CV (Tex. App. San Antonio Aug. 26, 2020), included almost $29 million in prejudgment interest on top of $235.4 million in compensatory and $470.8 million in punitive damages. Courts rarely award prejudgment interest on exemplary damages, which are not compensatory. Post-judgment interest runs at the federal statutory rate under 28 U.S.C. § 1961.

Procedure and timing: who decides what, and in what order

Remedies are won or lost on a calendar that starts with the complaint. The sequence below is the federal version; state courts vary the details but not the logic.

Plead every measure, then disclose a computation

Plead each remedy you may want: preliminary and permanent injunctions, actual loss, unjust enrichment (profits and avoided costs), a reasonable royalty in the alternative, exemplary damages with the facts that make the misappropriation willful and malicious, fees, and interest. Fed. R. Civ. P. 8(a)(3) requires a demand for the relief sought, and Fed. R. Civ. P. 9(g) requires items of special damage to be "specifically stated," which some courts apply to lost profits.

The more dangerous document is the initial disclosure. Rule 26(a)(1)(A)(iii) requires "a computation of each category of damages claimed" and the documents behind it. Trade secret plaintiffs often cannot compute unjust enrichment until they see the defendant's books, and courts accept a disclosed method that is supplemented under Rule 26(e) as the data arrives. What courts do not accept is silence followed by a new theory on the eve of trial. Fed. R. Civ. P. 37(c)(1) bars undisclosed information unless the failure was substantially justified or harmless. The Second Circuit affirmed preclusion of lost profits evidence first surfaced in a pretrial order in Design Strategy, Inc. v. Davis, 469 F.3d 284 (2d Cir. 2006), while making clear that preclusion is discretionary rather than automatic. The Ninth Circuit called the sanction "self-executing" in Hoffman v. Construction Protective Services, Inc., 541 F.3d 1175 (9th Cir. 2008). And in MicroStrategy, Inc. v. Business Objects, S.A., 429 F.3d 1344 (Fed. Cir. 2005), a trade secret plaintiff lost its damages expert for unreliable methodology and then lost its fallback theories because it had never added them to its interrogatory answers, which left it with partial summary judgment against it on damages.

Judge or jury

Damages for actual loss and reasonable royalties are legal remedies that go to the jury on a demand under Fed. R. Civ. P. 38. When a plaintiff seeks damages rather than equitable relief, "the claim is essentially legal in nature," as the court put it in LinkCo, Inc. v. Fujitsu Ltd., 232 F. Supp. 2d 182 (S.D.N.Y. 2002). Injunctions belong to the judge, who must make findings under Fed. R. Civ. P. 52(a).

Disgorgement of the defendant's profits sits in between. TAOS found no jury right to disgorgement under Texas law; the Eleventh Circuit reached the same result for Lanham Act profits in Hard Candy, LLC v. Anastasia Beverly Hills, Inc., 921 F.3d 1343 (11th Cir. 2019); and Liu describes disgorgement of net profits as a traditional equitable remedy. Pulling the other way, Beacon Theatres, Inc. v. Westover, 359 U.S. 500 (1959), and Dairy Queen, Inc. v. Wood, 369 U.S. 469 (1962), require that legal issues common to legal and equitable claims be tried to the jury first, and Dairy Queen treats a demand labeled an "accounting" as legal when it is really a claim for money. The historical test of Chauffeurs, Teamsters & Helpers Local No. 391 v. Terry, 494 U.S. 558 (1990), Tull v. United States, 481 U.S. 412 (1987), and Feltner v. Columbia Pictures Television, Inc., 523 U.S. 340 (1998), compares the remedy with what courts of law offered in 1791, and Great-West Life & Annuity Insurance Co. v. Knudson, 534 U.S. 204 (2002), adds that restitution is equitable only when it reaches particular funds or property traceable in the defendant's hands. Avoided costs are the hardest to classify, and in practice they have often gone to the jury, as in Epic and GlobeRanger.

The practical rule: settle the question at the Rule 16 conference, measure by measure, and if the court will decide disgorgement itself, ask it to use an advisory jury under Fed. R. Civ. P. 39(c), as the Motorola court effectively did, so that one trial produces usable findings whichever way the appellate court classifies the remedy.

Bifurcation, verdict forms, and instructions

Fed. R. Civ. P. 42(b) lets the court try liability and damages separately, or hold exemplary damages and fees for a later phase. The Federal Judicial Center's Trade Secret Case Management Judicial Guide § 9.2.3 (2023) suggests separating bad faith fee questions so that privileged communications and trial counsel's conduct stay out of the jury's case, and § 10.7 warns that the verdict form in a trade secret case "is at least as important and fraught with the potential for error as the instructions."

Use special verdicts or written questions under Fed. R. Civ. P. 49: secret by secret, was it a trade secret and was it misappropriated; measure by measure, what amount; and on separate lines, the head start period and any amount already counted elsewhere. Two cases show why. In O2 Micro International Ltd. v. Monolithic Power Systems, Inc., 399 F. Supp. 2d 1064 (N.D. Cal. 2005), the jury found five of eleven secrets misappropriated and awarded $12 million in unjust enrichment, but the expert had valued only the full set of secrets. The court granted judgment as a matter of law on that award because the jury had no reasonable basis to apportion, then awarded a $900,000 paid-up royalty under California's statute and $1.8 million in exemplary damages. In CardiAQ Valve Technologies, Inc. v. Neovasc Inc., 708 F. App'x 654 (Fed. Cir. 2017), by contrast, the plaintiff's technical witness testified that the key secrets were of roughly equal value and that an eighteen-month head start was realistic (the plaintiff had needed about twenty months), so the damages expert's $90 million royalty survived a split verdict. The jury awarded $70 million, and the Federal Circuit affirmed in all respects. A single-number model can work, but only if the record explains why each secret alone supports it.

Object to instructions on the record at the opportunity Fed. R. Civ. P. 51 requires the court to provide, before the instructions and closing arguments are delivered, and move under Rule 50(a) before the case goes to the jury if you want to renew the motion after the verdict. Neovasc lost one of its damages arguments in CardiAQ partly because it never moved to set aside the verdict as inconsistent.

After the verdict: remittitur, election, and the equitable phase

A court that finds an award excessive may order a remittitur, but the Seventh Amendment requires that the plaintiff be offered a new trial instead, Hetzel v. Prince William County, 523 U.S. 208 (1998). In diversity cases on state claims, the state's standard for excessiveness governs, Gasperini v. Center for Humanities, Inc., 518 U.S. 415 (1996). And a plaintiff whose damages rest on one expert should remember Weisgram v. Marley Co., 528 U.S. 440 (2000): if the appellate court holds the expert's testimony inadmissible and the remaining evidence cannot support the verdict, it may direct judgment for the defendant without a new trial. For the full post-trial toolkit, see Post-Trial Motions in Federal Court: Judgment as a Matter of Law, New Trial, Remittitur, and Relief From Judgment.

After the verdict the plaintiff elects among overlapping measures, taking the largest non-duplicative combination; the FJC guide § 7.3.4 describes that choice as the plaintiff's, subject to the bar on double recovery. The court then takes up the equitable questions, usually after post-trial motions are decided: the permanent injunction, any royalty in lieu of an injunction, disgorgement, exemplary damages, and fees. The injunction must fit the money. The comment to UTSA section 3 says an injunction "ordinarily will preclude a monetary award for a period in which the injunction is effective," and in CardiAQ the court refused an eighteen-month suspension of the defendant's development program because the royalty already compensated the same eighteen months. The Third Circuit in PPG drew the same line from the other side, upholding avoided-cost damages for past use alongside a forward-looking injunction because the two covered different periods.

Evidence and experts: building a number that survives Rule 702

The three experts most cases need

A trade secret damages case usually needs three witnesses with different jobs. The damages expert (an economist or forensic accountant) computes lost profits, unjust enrichment, avoided costs, and the royalty. The technical expert establishes which secrets the defendant used, in which products or processes, and how long legitimate development or reverse engineering would have taken, which is the head start. The forensic examiner shows what was taken, when, and where it went, which fixes the start of the misappropriation and often the date of the hypothetical negotiation.

The damages expert depends on the other two. Fed. R. Evid. 703 lets an expert rely on facts that experts in the field would reasonably rely on, and in CardiAQ the damages expert properly built on a technical witness's testimony about the relative value of each secret. But an expert who simply repeats another expert's untested conclusions invites reversal: in Mike's Train House, Inc. v. Lionel, L.L.C., 472 F.3d 398 (6th Cir. 2006), the Sixth Circuit ordered a new trial in a trade secret case because the district court admitted an expert's unreliable methodology without findings and let him vouch for his own conclusions by pointing to other experts who were not in the courtroom.

What Rule 702 now demands

Fed. R. Evid. 702, as amended in December 2023, makes the proponent show that it is more likely than not that the opinion rests on sufficient facts, uses reliable methods, and "reflects a reliable application of the principles and methods to the facts of the case." Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), and Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999), extend the gatekeeping duty to economists and accountants, and General Electric Co. v. Joiner, 522 U.S. 136 (1997), allows a court to exclude an opinion when "there is simply too great an analytical gap between the data and the opinion proffered." The FJC judicial guide § 8.3.5 observes that in trade secret cases speculation shows up most often in causation and damages. For the amended rule in depth, see Expert Witnesses After the 2023 Amendment to Rule 702: Daubert, Disclosure, and Cross-Examination, and for taking apart a damages expert under oath, Expert Depositions: Taking Apart the Opinion, Protecting Your Own Expert, and Building the Daubert Record.

The recurring ways damages opinions fail

• No apportionment. In Storage Technology Corp. v. Cisco Systems, Inc., 395 F.3d 921 (8th Cir. 2005), the plaintiff's expert treated the entire $450 million Cisco paid for a startup as the value of the employees and secrets taken from the plaintiff, although the startup had other assets and other key employees, and he could not say what the technology was. The Eighth Circuit affirmed summary judgment for Cisco.
• Ignoring other causes. MicroStrategy affirmed exclusion of reports that did not link "a single loss to a specific misconduct" and ignored significant alternative explanations, and TNS Media Research, LLC v. TRA Global, Inc., 977 F. Supp. 2d 281 (S.D.N.Y. 2013), rejected a claim that the defendants' conduct destroyed a company's value because one event followed another, which is the post hoc fallacy, not a method.
• Valuing the whole set when only some secrets are found. O2 Micro, discussed above, is the cautionary tale; the defense should press the per-secret allocation in deposition so that a partial verdict leaves the plaintiff's number unsupported.
• No reliable value at all. In Carbo Ceramics, Inc. v. Keefe, 166 F. App'x 714 (5th Cir. 2006), the Fifth Circuit affirmed summary judgment for the defendants because there was "no sound and reliable evidence from which to derive a dollar value for the alleged trade secrets."
• Argument dressed as expertise. In an earlier ruling in the LinkCo litigation, the court excluded a plaintiff's expert whose report offered "arguments and conclusory statements about questions of fact masquerading behind a veneer of technical language," and it granted both sides' motions to exclude the opposing damages experts, with leave to serve revised reports. LinkCo, Inc. v. Fujitsu Ltd., No. 00 Civ. 7242 (SAS) (S.D.N.Y. July 15, 2002).

What tends to survive

Juries that land between the parties' numbers usually keep their awards. The Tenth Circuit in Russo v. Ballard Medical Products, 550 F.3d 1004 (10th Cir. 2008), affirmed an unjust enrichment award that fell "somewhere in between the extremes suggested by the evidence," and recognized the plaintiff's right to "the full incremental value added" by the secret. GlobeRanger ($15 million against an expert's $19.7 million) and CardiAQ ($70 million against $90 million) follow the same pattern. Development cost evidence anchors royalties and avoided costs well: in MacDermid Printing Solutions LLC v. Cortron Corp., 833 F.3d 172 (2d Cir. 2016), testimony that the plaintiff spent about $3.8 million to develop the disclosed technical information supported a $3,790,939 hypothetical-negotiation award, and in Bourns, Inc. v. Raychem Corp., 331 F.3d 704 (9th Cir. 2003), evidence that the secrets cost $3 million a year to develop and saved the defendant at least three years supported a $9 million unjust enrichment award. A royalty does not require the defendant's commercial use under a statute like Utah's: in StorageCraft Technology Corp. v. Kirby, 744 F.3d 1183 (10th Cir. 2014), the Tenth Circuit affirmed a $2.92 million royalty against a former director who gave stolen source code to a rival, while stressing that the scope of the license the defendant "assumed for himself" drives the price.

Two evidentiary rulings recur. The court in LinkCo's 2002 opinion excluded the defendant's evidence that it never made a profit on the product, under Fed. R. Evid. 403, because a royalty prices the bargain at the time of misappropriation and later losses would "defeat the purpose" of that measure; expect courts to vary on that point. And the financial data behind every model usually comes in through business records under Fed. R. Evid. 803(6) and summaries under Fed. R. Evid. 1006; producing the underlying data and making the summary's author available avoids a foundation fight in front of the jury.

Criminal restitution, forfeiture, and the parallel prosecution

The Economic Espionage Act runs alongside the civil remedies, and its money provisions borrow the civil vocabulary. Theft of trade secrets under 18 U.S.C. § 1832 carries up to ten years in prison, and an organization may be fined up to the greater of $5 million or three times "the value of the stolen trade secret to the organization, including expenses for research and design and other costs of reproducing the trade secret that the organization has thereby avoided." Economic espionage for the benefit of a foreign government under 18 U.S.C. § 1831 raises the ceilings to fifteen years and $5 million for individuals and to the greater of $10 million or three times that value for organizations. Avoided costs, in other words, are written into the criminal fine.

Restitution is mandatory. 18 U.S.C. § 1834 incorporates 18 U.S.C. § 2323, which directs the court to order restitution under 18 U.S.C. § 3663A and 18 U.S.C. § 3664 and authorizes forfeiture of proceeds and of property used in the offense. The government must prove the victim's loss by a preponderance of the evidence under § 3664(e), restitution is limited to loss caused by the offense of conviction under Hughey v. United States, 495 U.S. 411 (1990), and Lagos v. United States, 584 U.S. 577 (2018), limits reimbursement of investigation and proceeding expenses to government investigations and criminal proceedings. Lagos calls into question earlier awards that reimbursed a company's own internal investigation, such as the $759,649.55 of forensic and legal fees claimed in United States v. Pu, 814 F.3d 818 (7th Cir. 2016) (vacated there for want of a complete accounting), and part of the award reviewed in United States v. Nosal, 844 F.3d 1024 (9th Cir. 2016).

Sentencing turns on "loss," and the Guidelines let the court estimate it in a trade secret case by "the cost of developing that information or the reduction in the value of that information that resulted from the offense." U.S.S.G. § 2B1.1 cmt. n.3(C)(ii). The courts of appeals insist that the estimate be reasoned. In United States v. Howley, 707 F.3d 575 (6th Cir. 2013), the district court found no loss at all despite the government's alternatives of $305,000 (a supply contract), $520,000 (the cost to recreate a tire-machine design), and $20 million (annual tire sales); the Sixth Circuit vacated the sentences because a conviction premised on a valuable secret required some reasoned estimate. In Pu, the Seventh Circuit vacated a $12 million intended-loss finding that simply adopted the victims' development costs, without addressing the defendant's argument that he took the code's outputs rather than the code or explaining how he intended to cause that loss. And in United States v. Liew, 856 F.3d 585 (9th Cir. 2017), where DuPont could quantify neither lost profits nor development cost, the district court used the defendants' gain ($22.5 million net, $28 million gross) as the measure; the Ninth Circuit vacated the sentence on other grounds without reaching that choice.

Three statutory rules link the criminal case to the civil one. Under § 3664(l), a conviction estops the defendant from denying the essential allegations of the offense in a later civil suit by the victim. Under § 3664(j)(2), restitution paid is reduced by any later compensatory damages for the same loss, so there is no double recovery. And 18 U.S.C. § 1835 requires orders preserving the confidentiality of the secret in the prosecution. A defendant facing both cases may invoke the Fifth Amendment in civil discovery, but under Baxter v. Palmigiano, 425 U.S. 308 (1976), a civil factfinder may draw an adverse inference from that silence. A criminal referral buys search warrants and mandatory restitution; it costs the owner control of the timetable and of the loss theory, and a sentencing court's loss figure is often lower than the civil damages model. For a victim's rights in that process, see Crime Victims' Rights: Notice, Participation, Restitution, and Compensation.

Preemption and the claims that ride along

Preemption decides which remedies outside the trade secret statute remain available, and that can matter more than the trade secret count itself. Section 7 of the UTSA displaces "conflicting tort, restitutionary, and other law" providing civil remedies for misappropriation, but it preserves contract remedies, other civil remedies "not based upon misappropriation of a trade secret," and criminal remedies. The DTSA preempts nothing, under § 1838.

The split is over confidential information that does not qualify as a trade secret. The Seventh Circuit's "dominant view" in Hecny Transportation, Inc. v. Chu, 430 F.3d 402 (7th Cir. 2005), is that claims are foreclosed "only when they rest on the conduct that is said to misappropriate trade secrets," so theft, fraud, and duty-of-loyalty claims survive; as Judge Easterbrook put it, it is "unimaginable that someone who steals property, business opportunities, and the labor of the firm's staff would get a free pass just because none of what he filched is a trade secret." The Fifth Circuit followed similar reasoning under Louisiana law in Brand Services, L.L.C. v. Irex Corp., 909 F.3d 151 (5th Cir. 2018), letting a conversion claim proceed for non-secret confidential information, and Wisconsin's high court took the same line in Burbank Grease Services, LLC v. Sokolowski, 2006 WI 103, 294 Wis. 2d 274. California goes the other way: Silvaco Data Systems v. Intel Corp., 184 Cal. App. 4th 210 (2010), disapproved on other grounds by Kwikset Corp. v. Superior Court, 51 Cal. 4th 310 (2011), held that CUTSA supersedes claims based on the same nucleus of facts even where the information is not a trade secret. Pleaded duplicates fail everywhere: the Federal Circuit applied Virginia's statute to bar a business conspiracy claim built on misappropriation in MicroStrategy, and Kansas's statute to bar an indistinguishable fraud claim in C&F Packing Co. v. IBP, Inc., 224 F.3d 1296 (Fed. Cir. 2000).

The remedial stakes are concrete. A surviving common-law claim may carry punitive damages without the statutory 2x cap, a conversion measure, or, in New York, the faithless servant rule, under which a disloyal employee forfeits compensation from the date disloyalty began, Phansalkar v. Andersen Weinroth & Co., 344 F.3d 184 (2d Cir. 2003). Contract claims add fee-shifting clauses, liquidated damages (subject to the penalty doctrine), and clauses reciting that a breach causes irreparable harm. Courts treat those recitals as evidence, not commands: the Second Circuit relied on one in Ticor Title Insurance Co. v. Cohen, 173 F.3d 63 (2d Cir. 1999), while CardiAQ held that a contractual acknowledgment did not oblige the court to enjoin.

Federal preemption rarely bites. Trade secret claims survive copyright preemption under 17 U.S.C. § 301 because breach of confidence is an extra element, Computer Associates International, Inc. v. Altai, Inc., 982 F.2d 693 (2d Cir. 1992), and patent law does not preempt state trade secret law, Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470 (1974). One federal remedy deserves a place in any plan against an importer: the International Trade Commission may exclude goods made with misappropriated secrets under 19 U.S.C. § 1337 even when the misappropriation happened abroad, TianRui Group Co. v. International Trade Commission, 661 F.3d 1322 (Fed. Cir. 2011). See the ITC Section 337 Toolkit: Complaints, Domestic Industry Proof, and Remedy Briefing.

Defenses that cut awards down

Most remedies defenses are not affirmative defenses in the technical sense; they are attacks on the plaintiff's proof of causation, value, and time. The FJC judicial guide § 10.7 makes the point for independent development: the burden of proving misappropriation stays with the plaintiff, even though the defendant will naturally offer the evidence. The defenses that move numbers most often:

1. Independent development and reverse engineering. Neither is improper means under 18 U.S.C. § 1839(6)(B), and trade secret law has never protected against "discovery by fair and honest means," Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470 (1974); see also Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141 (1989). Clean-room records, lab notebooks, and commit histories are the evidence. Even partial independent work shortens the head start.
2. The secret went public or was readily ascertainable. Publication ends protection prospectively: Attia (published patent applications), Group One, Ltd. v. Hallmark Cards, Inc., 254 F.3d 1041 (Fed. Cir. 2001) (damages limited to the pre-publication head start, which the plaintiff could not prove), and TAOS (reverse engineering by January 2006). The limit runs both ways, though: taking a compilation wholesale can be improper even if each piece was public, as the Eleventh Circuit held of a scraped insurance-quote database in Compulife Software Inc. v. Newman, 959 F.3d 1288 (11th Cir. 2020), and a combination of public elements can itself be a secret, Minnesota Mining & Manufacturing Co. v. Pribyl, 259 F.3d 587 (7th Cir. 2001).
3. Causation and apportionment. Syntel (no injury beyond $8.5 million in lost profits), the Virginia Appian decisions (the plaintiff's burden), Storage Technology (an acquisition price is not a secret's value), and Mattel (no constructive trust over a brand built mostly by the defendant) are the core citations.
4. The head start already ran or was already served. Compute the lawful-availability date and the development time, and credit time already spent under a preliminary injunction (K-2 Ski).
5. Limitations and laches. The DTSA's three-year period runs from actual or constructive discovery, and a continuing misappropriation is one claim; delay also weighs against equitable relief under Restatement (Third) of Unfair Competition § 45.
6. Money is adequate. Without a presumption of irreparable harm (Faiveley), a defendant that can pay and is not disseminating the secret can defeat or narrow an injunction, especially when the damages award already compensates the same period, as in CardiAQ.
7. Good faith change of position. UTSA section 3(a) withholds damages "to the extent that a material and prejudicial change of position prior to acquiring knowledge or reason to know of misappropriation renders a monetary recovery inequitable," and section 2(b) offers the same defendant a royalty instead of an injunction.
8. No whistleblower notice. Under 18 U.S.C. § 1833(b)(3)(C), an employer that did not give the required notice cannot recover exemplary damages or attorney's fees against that employee.
9. Caps, ratios, and extraterritorial limits. Virginia caps exemplary damages at $350,000; due process limits uncapped common-law awards (Epic); and foreign sales stay out of the DTSA base unless the offender is a U.S. person or "an act in furtherance of the offense was committed in the United States" under § 1837, the test Motorola applied through RJR Nabisco, Inc. v. European Community, 579 U.S. 325 (2016).

Defendants should remember that equity cuts both ways. In Wyeth v. Natural Biologics, Inc., 395 F.3d 897 (8th Cir. 2005), a defendant that destroyed evidence and gave false testimony lost its own equitable defenses to unclean hands and received a permanent injunction, not a head start injunction, reaching its development work on the competing product, because the district court found it "cannot be trusted" and no one had legitimately replicated the process in decades.

Strategy for each side

For the trade secret owner

• Decide the remedy theory at intake. Ask what the client lost, what the defendant gained or saved, and what a license would have cost, and write down which measure you expect to lead with. The answer shapes the preservation letter, the complaint, and the first discovery requests.
• Move fast if you want an injunction. Delay is the defendant's best irreparable-harm argument, and the preliminary injunction hearing is where the head start period is first argued. See Preliminary Injunctions and Temporary Restraining Orders: Emergency Relief in Federal Court.
• Build the identification and the damages model together. Number the secrets, and make the expert value each one or explain why each alone carries the whole value; the identification fight is covered in Identifying Trade Secrets in Litigation: Reasonable Particularity, Sequencing Discovery, and Protecting the Secret While You Sue.
• Present alternatives. Put lost profits, unjust enrichment, avoided costs, and a royalty in front of the factfinder with a non-overlap table, and elect after the verdict.
• Prove willfulness with documents. Concealment, deletion, recruitment aimed at information, and use after a demand letter carry multipliers and fees.
• Watch collectibility. A judgment against a foreign or thinly capitalized defendant is a piece of paper; seek security, consider the ITC, and price the case accordingly.

For the accused employee or competitor

• Shrink the target. Force a precise identification and test each secret; every secret that falls takes its share of the damages model with it.
• Build the head start record early. Collect your own development history and the public record (patent publications, product releases, conference papers) that show when lawful access began.
• Offer a credible alternative number. A defense expert who says "zero" is easy to ignore; one who computes a realistic royalty, a discounted development cost (Tata's 30 percent labor adjustment in Epic), or a short head start gives the jury somewhere to land. In GlobeRanger, the defense claim that it could have built the technology for $140,000 did not.
• Fight double counting and the burden. Ask for instructions on causation, apportionment, and non-duplication, and for a verdict form that separates each measure and period.
• Keep disgorgement with the judge where the law allows, and preserve the jury-trial objection either way.
• Use the fee statute. A claim pressed in bad faith exposes the plaintiff to fees; document why the claim is weak, in writing, early.

Costs and practical realities

Trade secret remedies litigation is slow and expensive, and the remedies phase is where much of the money goes. The figures below are rough planning ranges drawn from practice, not survey data, and large cases routinely exceed them.

Time is the other cost. Epic sued Tata in 2014 and the Seventh Circuit ruled in 2020; Motorola sued Hytera in 2017 and the Seventh Circuit ruled in 2024; the Syntel litigation began in 2015 and was headed for a second damages trial nearly a decade later. Remedies drive that length, because every appellate reversal on damages sends the case back for more.

Three realities shape settlement. First, the multiplier and fee exposure make willfulness evidence the biggest swing factor in valuation; a case with a deleted-files narrative settles at a different number than a case about a confused former employee. Second, collectibility matters more than the verdict: Motorola involved a foreign defendant that slow-walked a court-ordered royalty, and in many departing-employee cases the individual cannot pay and the new employer is the only real target. Third, insurance rarely covers misappropriation, so check coverage early, but do not count on it. For settlement mechanics, see Settlement Agreements, Releases, and Rule 68: Ending a Case on Terms That Hold, and for collection, Collecting a Judgment: Discovery in Aid of Execution, Liens, Levies, and Garnishment.

Common mistakes

• Claiming the owner's full development cost when the owner lost little. After Syntel, that theory needs a harm beyond the owner's lost profits in the Second Circuit, and it is unavailable on New York common-law claims.
• One number for many secrets. If the jury may find only some secrets, the model must say what each is worth (O2 Micro), or explain why each alone supports the whole (CardiAQ).
• Starting from total revenue. Storage Technology and the Virginia Appian decisions show what happens when the plaintiff skips apportionment.
• Ignoring the clock. Sales after the head start ends are not caused by the misappropriation (TAOS, Sokol).
• Letting the verdict form merge measures. Double counting is the most common reversible error (Salsbury, Motorola); separate lines fix it.
• Late or thin damages disclosures. Rule 37(c)(1) preclusion is a real risk (Design Strategy, MicroStrategy).
• Overclaiming. Asserting 67 secrets and proving 8 (Steves) costs credibility, and a claim made in bad faith risks a fee award.
• For defendants: no alternative number, and no preservation. A zero-damages position with no fallback, a missed Rule 50(a) motion, or an instruction objection never made leaves the appellate court nothing to fix. And destroying evidence can turn a head start case into a permanent injunction (Wyeth).
• Forgetting the whistleblower notice. Old employment agreements without the § 1833(b) notice quietly forfeit exemplary damages and fees against employees.

A worked example: a two-year head start on a manufacturing process

Suppose Kestrel Coatings spent five years and $6 million developing a curing process for industrial powder coatings that saves $1.20 per unit in energy and labor. In March 2023 Kestrel's process engineer leaves for Halcyon Finishes, taking the oven profiles and parameter spreadsheets, and Halcyon converts its line in June 2023. Halcyon makes 8 million units a year with the process; 1.5 million of those units a year are sales it takes from two former Kestrel customers. Kestrel's incremental margin on lost units is $0.90, and Kestrel cuts its price by $0.05 a unit on its remaining 10 million units a year to hold its other customers.

The head start is the main fight. Kestrel relies on Halcyon's own shelved internal project plan, which budgeted 24 months and $3.5 million to develop a comparable process, so the head start runs to June 2025. Halcyon points to a September 2024 academic paper disclosing the key parameter window and an expert who says implementing it would have taken about three months, so lawful availability came in December 2024 and the head start is 18 months.

The savings line is unjust enrichment measured by the old "standard of comparison," which compares the defendant's cost of reaching a result with and without the appropriated process, International Industries, Inc. v. Warren Petroleum Corp., 248 F.2d 696 (3d Cir. 1957). It excludes Halcyon's savings on the 1.5 million diverted units because Kestrel's lost profits already cover those sales, which is what "not addressed in computing damages for actual loss" requires. Kestrel's expert also computed Halcyon's avoided $3.5 million development budget. Whether that can be added depends on the counterfactual: if Halcyon would have spent the money and made the same sales anyway, savings and profits overlap, which is the Salsbury reasoning; if, as here, Halcyon would have had no process at all during the head start, the budget is a separate benefit. Because the point is contested and adds less than 20 percent to Kestrel's figure, Kestrel puts it on its own verdict line so that an appellate court can strike it without disturbing the rest.

Two more consequences follow. Under either side's head start, no use injunction issues at a late 2025 trial, because both proposed periods have run by then and an injunction may not put Halcyon behind where honest development would have left it; Kestrel still gets an order requiring return and certified destruction of its files. And the exemplary claim rests on evidence such as the engineer's downloads on his last night and a Halcyon executive's email asking him to "bring the oven profiles," which is the kind of record that supports a willful and malicious finding.

A worked example: a software company's avoided development costs

Suppose Larkspur Analytics spent four years and $24 million building a claims-scoring engine for health insurers. Brightwater Systems, a subcontractor with access to Larkspur's design documents, downloads the feature specifications and tuning parameters and builds a competing module in eleven months. Brightwater sells the module to two insurers that were never Larkspur prospects, earning $5.2 million in revenue and $0.8 million in profit over two years. Larkspur loses no customers, and an injunction will stop further use after trial. This is the Syntel problem with one difference: Brightwater built a competing product.

Larkspur's expert builds an avoided-cost number the way Epic's expert did:

1. Start with Larkspur's $24 million development cost.
2. Remove $6 million for components Brightwater did not use (the user interface and regulatory certification work): $18 million.
3. Discount 20 percent for technology decay, because part of the early work was obsolete by the time of the taking: $14.4 million.
4. Apply a 30 percent labor-rate discount for Brightwater's cheaper engineering team, the adjustment the Epic jury adopted: $10.08 million.
5. Subtract the $2.1 million Brightwater actually spent: about $8 million.

A cross-check by time supports it. The defense's own technical expert says a competent team starting from public literature would need about 30 months, so Brightwater gained roughly 19 months. Twenty engineers for 30 months at a loaded cost of $220,000 a year is $11 million, less the $2.1 million actually spent, or $8.9 million. Two methods that land between $8 million and $8.9 million are much harder to attack than one.

The forum decides how much of that survives. In the Seventh and Third Circuits, Epic and PPG support the award even though Larkspur lost nothing. In the Second Circuit, Syntel requires more than a disembodied cost. Its list of circumstances that can justify avoided costs begins with a defendant that used the secret to build a competing product, which fits Brightwater, but Larkspur should still prove how the competing module diminished the secret's value to it, because that proof was what TriZetto lacked. On a New York common-law claim, E.J. Brooks takes avoided costs off the table, and Larkspur falls back on a royalty: in a hypothetical negotiation at the time of access, MacDermid supports an expert's view that Larkspur would not have licensed for less than a meaningful share of its development cost. Brightwater's best responses are the ones that cut the base rather than deny it: if the jury finds only some of the twelve asserted secrets, the model must allocate cost by secret; the decay and labor discounts should be larger; and the head start should be shorter. If the jury finds the taking willful and malicious, the DTSA allows up to twice the compensatory award on top.

Frequently asked questions

Can I recover both my lost profits and the defendant's profits? Yes, but only the defendant's profits on sales you would not have made yourself. Profits on sales you lost are already in your lost profits, and the statutes allow unjust enrichment only to the extent it is "not addressed in computing damages for actual loss."

What if the defendant never made a dime from the secret? Failure to profit is no defense, as University Computing put it, because the misappropriator should bear the risk of its own venture. Look to avoided development costs, where the forum allows them, or a reasonable royalty, which does not depend on the defendant's profits (StorageCraft; and California's statute expressly makes a royalty available when neither loss nor enrichment is provable).

If I win, do I automatically get an injunction? No. eBay's four factors apply, most courts refuse to presume irreparable harm, a DTSA injunction cannot stop someone from taking a job, and any injunction should last only as long as needed to erase the head start.

How long does a head start injunction last? As long as the evidence says a legitimate competitor would have needed: two years in Winston, eight months on a preliminary basis in Lamb-Weston, and longer only when the defendant could never have gotten there lawfully or cannot be trusted (Wyeth).

Who decides exemplary damages and fees? Under the DTSA, the court, usually after the jury finds the misappropriation willful and malicious. Some states assign exemplary damages to the factfinder; Texas requires clear and convincing evidence.

Are the defendant's profits its whole revenue? No. Disgorgement reaches net profits attributable to the secret. Whether the defendant must prove the deductions and the non-attributable sales depends on the forum: Motorola and Restatement § 45 comment f shift that burden, while Virginia's courts in Appian refused to.

Can a company collect both criminal restitution and a civil judgment? Not for the same loss. Under 18 U.S.C. § 3664(j)(2), restitution is reduced by later compensatory damages for the same loss, and the conviction can estop the defendant in the civil case.

Can a defendant recover its fees if the plaintiff loses? Yes, if the claim was made in bad faith and the defendant is a prevailing party. A voluntary dismissal without prejudice does not make the defendant a prevailing party (Dunster Live).

Does a royalty require proof that I would ever have licensed the secret? No. The hypothetical negotiation assumes a willing licensor and licensee at the time the misappropriation began, a "suppositious meeting" in the words of Vermont Microsystems, and the owner's development cost can set a floor (MacDermid).

Practical takeaways

• Choose the remedy theory at intake, and plead and disclose every measure you may need.
• Value each identified secret, or explain why each alone carries the whole value.
• Compute the head start from evidence: lawful-availability dates, real development histories, and reverse-engineering effort.
• Keep lost profits, unjust enrichment, avoided costs, and royalties on separate verdict lines, and elect after the verdict.
• Know your circuit on avoided costs (Syntel versus Epic and PPG) and your state on royalties, burdens, and exemplary caps.
• Settle who decides disgorgement before trial, and use an advisory jury when the answer is uncertain.
• Seek injunctions fast and narrow, tied to identified secrets and to the head start, with a bond the evidence supports.
• Treat willfulness evidence as the biggest swing factor in exposure, and audit the § 1833(b) notice before promising a multiplier.
• For the step-by-step version, see Proving or Defending Against Trade Secret Damages: A Practical Guide; for task lists, the Trade Secret Damages and Remedies Checklist: A Practical Checklist; and for model language and calculation templates, the Trade Secret Remedies Toolkit: Damages Models, Head Start Analyses, and Injunction Briefing.

This document is general information, not legal advice for any particular situation; the law changes, and the facts matter.

Related Documents

Related Articles

• Identifying Trade Secrets in Litigation: Reasonable Particularity, Sequencing Discovery, and Protecting the Secret While You Sue: the numbered identification that every per-secret damages model depends on.
• Expert Depositions: Taking Apart the Opinion, Protecting Your Own Expert, and Building the Daubert Record: deposing and defending the damages and head start experts.
• Post-Trial Motions in Federal Court: Judgment as a Matter of Law, New Trial, Remittitur, and Relief From Judgment: what happens to a damages verdict after the jury leaves.
• Patent Damages: Reasonable Royalties, Lost Profits, Apportionment, and Enhancement: the royalty and apportionment law that trade secret courts borrow.
• Proving Damages in Civil Litigation: Certainty, Causation, and the Numbers That Survive: reasonable certainty and causation across claim types.
• Copyright Damages and Remedies: Statutory Damages, Profits, Injunctions, and Fees: the profits-apportionment framework Motorola adapted.
• Protection of Trade Secrets: the substantive law of what qualifies and how it is lost.
• Non-Compete Agreements Under Siege: FTC Rulemaking and State Law Developments: the state restraint-of-trade rules that limit DTSA injunctions.

Related Guides

• Proving or Defending Against Trade Secret Damages: A Practical Guide: the step-by-step companion to this article for both sides.
• Trade Secret Misappropriation Litigation Under the Defend Trade Secrets Act: A Practical Guide: the full life cycle of a DTSA case.
• Managing Discovery in a Trade Secret Case: A Practical Guide: getting the financial and development records a damages model needs.
• Building a Damages Case: A Practical Guide from Pleading to Verdict: the general damages workflow, stage by stage.
• Preparing and Opposing a Fee Petition: A Practical Guide: fee motions after a willful and malicious or bad faith finding.

Related Checklists

• Trade Secret Damages and Remedies Checklist: A Practical Checklist: every remedies task in checklist form.
• Trade Secret Identification and Discovery Checklist: A Practical Checklist: identification and discovery steps that feed the damages case.
• Damages Proof and Expert Model Checklist: A Practical Checklist: testing any damages model before the other side does.
• Preliminary Injunction and TRO Application Checklist: A Practical Checklist: emergency relief papers, bonds, and hearings.

Related Toolkits

• Trade Secret Remedies Toolkit: Damages Models, Head Start Analyses, and Injunction Briefing: templates for the remedies phase.
• Trade Secret Litigation Toolkit: Identification Statements, Two-Tier Protective Orders, and Forensic Protocols: identification statements and protective orders for the liability case.
• Patent Damages Toolkit: Royalty Models, Expert Reports, and Daubert Motions: royalty models and Daubert briefing that translate to trade secret cases.
• Civil Damages Toolkit: Contract, Tort, Statutory, and Punitive Recovery: punitive damages and statutory multipliers across claims.
• Expert Witness Toolkit: retaining, preparing, and disclosing experts.