Toolkit · Intellectual Property

The IP Clearance Toolkit: Securing Freedom to Operate Before You Launch

Type
Toolkit
Jurisdiction
National
Date
2025-03-13
Topic
Intellectual Property

This toolkit is a research-grade roadmap for clearing intellectual property risk before a product, brand, or codebase ships. It explains the counterintuitive core of clearance—that owning a patent, trademark, or copyright confers a right to exclude, not a right to practice—and walks the full clearance workflow across all four IP families. Stages cover patent freedom-to-operate searching and claim-by-claim infringement mapping, trademark availability and the common-law traps a register search misses, copyright and open-source license clearance, and the way a documented investigation plus a competent opinion of counsel blunts enhanced-damages exposure after Halo Electronics v. Pulse Electronics. Each stage pairs the governing statutes and cases with the search tools and cross-referenced UpLaw resources a practitioner actually needs. It is written for founders, product and brand teams, in-house counsel, and outside clearance counsel, and it treats clearance as calibrated risk management rather than a guarantee. Read the roadmap, then drill into the linked search checklists and deep-dive guides.

The IP Clearance Toolkit: Securing Freedom to Operate Before You Launch

Intellectual Property > Patent | Published 13 March 2025 - Updated 2 July 2025 | UpLaw - uplaw.us

A patent, a trademark registration, and a copyright each hand you a sword, not a shield. They let you stop others; none of them promises that what you are about to sell is legal to sell. Freedom-to-operate clearance is the discipline of finding the swords pointed at you—while a design review still costs less than a recall.

Keywords: freedom to operate, FTO, IP clearance, patent clearance, trademark clearance, copyright clearance, open source compliance, opinion of counsel, willful infringement, enhanced damages, Halo v. Pulse, design-around, claim construction, all-elements rule, software bill of materials, Tea Rose-Rectanus, 35 U.S.C. § 284, doctrine of equivalents

The permission paradox at the center of clearance

Here is the trap that catches sophisticated companies. A hardware team spends two years and seven figures inventing a new insulin pump, files a patent, gets it allowed, and hangs the certificate on the wall. Everyone assumes the patent means the pump is theirs to sell. It does not. A United States patent grants the right "to exclude others from making, using, offering for sale, or selling the invention." 35 U.S.C. § 154(a)(1). Read that sentence twice. It is a right to exclude, phrased entirely in the negative. Nowhere does the Patent Act grant an affirmative right to practice your own invention.

The consequence is not a technicality. Your pump can be covered by your own valid patent and still infringe a competitor's earlier patent on, say, the occlusion-detection algorithm buried in the firmware. Patents stack. The classic illustration is the improvement invention: Company A patents a chair; Company B invents and patents a chair with a reclining mechanism. B cannot make its reclining chair without practicing A's chair claims, and A cannot add a recliner without practicing B's. Each holds a blocking position over the other. Owning IP answers the question "can I stop someone from copying me?" It says nothing about the different, launch-critical question: can I ship this product, under this name, running this code, without practicing rights that belong to someone else?

That second question is freedom to operate (FTO), and clearance is the process of answering it before you are financially and reputationally committed. The economics are lopsided in a way that should make the decision easy. Catch a blocking right during design and the fix is an engineering ticket, a brand-name swap, or a dependency substitution—hours, not millions. Catch the same right after launch and the menu changes to injunction, product recall, corrective advertising, a rebrand across every SKU and every marketing surface, a damages award, and, where the infringement was willful, damages trebled under 35 U.S.C. § 284 plus the other side's attorney's fees under § 285. Clearance is the practice of front-loading cheap decisions so you never have to make the expensive ones.

This toolkit maps clearance across all four IP families—patent, trademark, copyright, and trade secret-adjacent open-source obligations—and hands you off at each stage to the operational checklist or the deep-dive analysis you will need. It is written for three audiences at once: founders and product or brand teams who need to understand what clearance buys; in-house counsel coordinating a launch across engineering, marketing, and finance; and outside counsel running the searches and writing the opinions. Read it as a navigation layer over UpLaw's clearance library, not as a substitute for the search itself.

A note on calibration before we start

Clearance is risk management, not a guarantee, and its cost should scale to the exposure. A weekend-project app with negligible revenue does not need—and cannot justify—the exhaustive FTO study a medical-device maker runs before a nine-figure launch into a crowded patent thicket. The craftsmanship in clearance is proportionality: matching the depth, breadth, and price of the investigation to what is actually at stake in the product, the market, and the jurisdictions that matter. Under-scope and you buy false comfort; over-scope and you burn budget and miss the launch window. Every stage below should be read through that lens.

Stage one: deciding to clear, and the knowledge problem

The threshold decision is not how to clear but whether to look at all—and that decision has a legal dimension people rarely see coming. There is an old instinct, sometimes called the ostrich strategy, that says: do not search for patents, because once you know about a blocking patent you can no longer plead ignorance, and knowledge is the raw material of a willful-infringement claim. On that view, the safest move is to keep your head in the sand.

That instinct is largely obsolete, and understanding why is the beginning of good clearance hygiene. Two developments buried it.

First, Congress closed the "no opinion" trap. Under 35 U.S.C. § 298, the failure to obtain the advice of counsel—and the failure to present that advice to the trier of fact—"may not be used to prove that the accused infringer willfully infringed" or induced infringement. Section 298 codified the America Invents Act's abrogation of the Federal Circuit's earlier suggestion, in Broadcom Corp. v. Qualcomm Inc., 543 F.3d 683, 698-99 (Fed. Cir. 2008), that a jury could hold the absence of an opinion against a defendant. You are no longer punished for not commissioning a formal opinion. That removes the perverse incentive to stay ignorant.

Second, willful blindness is not a hiding place. In the induced-infringement context the Supreme Court held in Global-Tech Appliances, Inc. v. SEB S.A., 563 U.S. 754 (2011), that a defendant who deliberately shields itself from confirming a high probability of infringement is treated as if it knew. The ostrich, in other words, gets no better treatment than the person who looked. Deliberate ignorance is a culpable mental state, not a safe one.

Put those together and the modern answer is the opposite of the ostrich: clear deliberately, document the analysis, and convert knowledge from a liability into a defense. A disciplined search paired with a reasoned non-infringement or invalidity opinion is not evidence of a guilty mind; it is affirmative proof of a good-faith one. Stage five returns to how that shield is built. For now the point is structural: build a clearance gate into your development process—a defined checkpoint before a design freezes, before a brand is announced, before a dependency is baked into the shipped binary—so that clearance happens when options are still cheap.

Worked example (hypothetical). Meridian Robotics is three weeks from unveiling a warehouse-picking robot when a director asks the question no one asked at the design review: "Has anyone checked whether this steps on someone's patents?" A rushed weekend search surfaces a competitor's patent whose independent claim reads squarely on Meridian's core gripper-and-vision loop. Because clearance was never gated into the design phase, the same discovery that would have cost a two-week firmware tweak eight months earlier now costs a delayed launch, an emergency opinion at rush rates, and a board that has lost confidence in the timeline. The patent did not change. The timing of the knowledge changed the price by two orders of magnitude.

Scope the project before you spend a dollar searching. Define the product, brand, or codebase with precision; the relevant markets and jurisdictions (FTO is territorial—a patent enforceable in Germany is irrelevant to a U.S.-only launch, and vice versa); the specific technical features or brand elements in play; the budget; and the launch timeline. Clearance also throws off strategic intelligence as a byproduct: a good patent search reveals competitors' filing patterns, white space where your own filings would sit unopposed, and licensing counterparties. For the founder-side view of how IP clearance folds into a broader readiness posture, see Getting Diligence-Ready Before You Raise; for the cross-regime primer on what each IP right actually protects, see Copyright, Trademark, Patent, and Trade Secret Compared.

Stage two: patent freedom-to-operate, or finding the live claims that block you

Patent FTO—also called a clearance or right-to-use search—hunts for one thing: unexpired, in-force patents whose claims your product would practice in your target markets. Keep that target in view, because it is narrower than people assume, and it is different from the search you may already have run.

FTO is not a patentability search

A patentability (or novelty) search asks whether you can obtain a patent. It sweeps in all prior art—expired patents, foreign references, journal articles, anything published—because anything in the public domain can defeat your novelty under 35 U.S.C. § 102 or render your claims obvious under § 103. An FTO search asks the inverse: whether someone else's live rights stand in your way. Expired patents are irrelevant to FTO (their subject matter is free for all to practice); only patents currently in force in your jurisdictions can block you. Confusing the two searches is a common and expensive error—a clean patentability opinion tells you nothing about whether you infringe. For the doctrinal machinery of patentability, see Obviousness Under Section 103 and the overview in How U.S. Patents Work.

The search workflow

  1. Deconstruct the product into features and functions. FTO reads on what the product does and how, not on its marketing name. Break the product into its technical elements—each subsystem, algorithm, material, and method step—because that decomposition is what you will map against claims.

  2. Search along multiple axes. No single database and no single query is complete. Search by Cooperative Patent Classification (CPC) codes for the technology; by keyword and synonym (accounting for the fact that patent drafters invent their own vocabulary to broaden scope); by assignee (your known competitors and the serial patentees in the field); and by forward and backward citation chaining from any patent you find, which surfaces the family and the neighbors.

  3. Use complementary tools. The workhorses are free and public. USPTO Patent Public Search (which replaced the legacy PatFT and AppFT systems) covers U.S. patents and published applications. Espacenet, the European Patent Office's database, offers broad international coverage with machine translation of foreign-language documents. Google Patents gives fast full-text search and clean family and citation views. WIPO PATENTSCOPE covers PCT international applications and many national collections. A serious FTO study consults all four and, for high-stakes launches, adds a commercial database and a professional searcher.

  4. Narrow to live patents in your markets. Filter out expired references, patents that lapsed for nonpayment of maintenance fees, and patents enforceable only in jurisdictions you are not entering. The U.S. utility patent term runs twenty years from the earliest non-provisional filing date, 35 U.S.C. § 154(a)(2), subject to patent-term adjustment and, in limited fields, extension under § 156; maintenance fees fall due at 3.5, 7.5, and 11.5 years, and a patent whose fees went unpaid has lapsed and cannot be asserted. Confirm in-force status for every candidate; a "blocking" patent that expired last year is not blocking anything.

Claim mapping is where FTO lives or dies

Infringement turns on the claims—the numbered sentences at the end of the patent—not the title, the abstract, or the pretty figures. The single most important rule in patent FTO is the all-elements rule: to literally infringe a claim, the accused product must practice every limitation of at least one claim. Omit even one element of an independent claim and you do not literally infringe that claim. See Patent Infringement, Claims, and Defenses and the element-by-element walkthrough in What Counts as Patent Infringement: An Anatomy of the Section 271 Analysis.

Read the independent claims first—they stand alone and are the broadest. Dependent claims incorporate an independent claim and add limitations, so they are necessarily narrower; if you clear the independent claim, you generally clear its dependents. Build a claim chart: one column for each claim limitation, an adjacent column for the corresponding feature (or absence) in your product. A limitation with no counterpart in your product is a gap that defeats literal infringement of that claim.

Two doctrines complicate the mapping and both must be on the chart:

Triage the hits into three buckets

Not every live patent that mentions your field is a threat. Sort every candidate into one of three categories and spend your expensive lawyering accordingly:

  • (a) Clear non-infringement. The product misses one or more limitations of every independent claim under any reasonable construction. Document the gap and move on.
  • (b) Genuine concern. The mapping is close, a claim term is contestable, or the doctrine of equivalents is live. These deserve opinion-level analysis (Stage five).
  • (c) Apparently valid and infringed. The product practices the claims and the patent looks solid. These demand a response—license, design-around, or a validity challenge—before launch.

For category (c), remember that "infringed" and "enforceable" are not the same. A patent is presumed valid, and an accused infringer must prove invalidity by clear and convincing evidence. 35 U.S.C. § 282(a); Microsoft Corp. v. i4i Ltd. Partnership, 564 U.S. 91 (2011). That heightened burden is the single biggest reason a non-infringement position is usually more comfortable than an invalidity position (more on that in Stage five). Still, invalidity is a real option: prior art you turned up in the search may anticipate under § 102 or render the claims obvious under § 103; software and business-method claims may be vulnerable under the Alice/Mayo eligibility framework (see Protecting Software and Business-Method Inventions After Alice); and inter partes review before the Patent Trial and Appeal Board, 35 U.S.C. §§ 311-319, offers a lower burden (preponderance) and a specialized forum for testing validity on patents and printed publications.

Worked example (hypothetical). Calla Health is clearing a continuous glucose monitor. The FTO search surfaces a competitor patent whose independent claim requires "a sensor comprising a wired transcutaneous electrode." Calla's device uses a fully wireless optical sensor—no electrode, no wire. Under the all-elements rule, the missing "electrode" limitation defeats literal infringement, and because the specification repeatedly distinguishes optical sensing as outside the invention, prosecution history and the specification make a doctrine-of-equivalents theory weak. The hit lands in bucket (a). A second patent, though, claims "a processor configured to transmit a glucose value to a paired mobile device at intervals of not more than five minutes." Calla transmits every ninety seconds. Every limitation reads on Calla's product; the hit is bucket (c), and it drives a licensing conversation and a parallel invalidity study before launch.

The four responses to a real blocking patent

When a category (c) patent survives triage, four moves remain, and they are not mutually exclusive:

  1. Take a license. Often the cleanest path, and clearance gives you the leverage of time and options. See Monetizing a Patent Through Licensing: From Valuation to Term Sheet and the deal-structure walkthrough in Building a Patent License Term Sheet. In standards-heavy fields, licensing may come with FRAND obligations—see Standard-Essential Patents and FRAND Licensing.
  2. Design around (Stage six).
  3. Challenge validity or enforceability—through prior art (§§ 102, 103), eligibility (§ 101), or, more rarely, inequitable conduct that renders the patent unenforceable (see Inequitable Conduct and the Duty of Candor).
  4. Accept the risk deliberately—a legitimate business decision when the exposure is small and quantified, but one that should be made with eyes open and, ideally, an opinion in the file.

For the operational search workflow end to end, work through the Freedom-to-Operate Patent Search checklist and the narrative companion, Clearing a Product for Launch: How to Conduct a Freedom-to-Operate Analysis. If litigation is already on the horizon, the Litigating Patent Infringement toolkit picks up where clearance ends.

Stage three: trademark clearance, or is the name free and defensible

Before you commit a brand name, logo, tagline, or product name to packaging, code signing, app-store listings, and a domain, clear it. Trademark clearance answers two distinct questions, and both matter:

  1. Is the mark available? Is it free of conflict with a senior user's or registrant's mark for related goods or services—that is, would your use create a likelihood of confusion?
  2. Is the mark protectable? Is it distinctive enough to register and enforce, or will it be refused or forever weak?

Protectability first: the distinctiveness spectrum

The second question is quick to screen and saves heartbreak. Marks fall along a spectrum from strongest to unprotectable: fanciful (KODAK), arbitrary (APPLE for computers), suggestive (COPPERTONE), descriptive (protectable only with acquired secondary meaning), and generic (never protectable). Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4 (2d Cir. 1976). A team that falls in love with a descriptive name ("CreamySmooth" for yogurt) is choosing a mark that is hard to register, easy to design around, and expensive to enforce. Screening for distinctiveness at the naming stage is the cheapest brand-strength decision you will ever make; the checklist Choosing a Mark the Law Will Protect walks the screen.

The search: from knockout to full

Clearance searching runs from cheap-and-narrow to thorough-and-broad:

  • Knockout search. A fast pass for identical and near-identical marks in the relevant classes, run on the USPTO's public database. Its job is to kill obviously doomed candidates before you invest in them.
  • Comprehensive (full) search. For a surviving finalist, a full search covers the federal register, the fifty state registers, common-law uses (business directories, trade publications, the open web), domain registrations, and social handles—usually through a professional search vendor whose reports are structured for legal analysis. Note the modern USPTO toolset: since the January 2025 platform transition, the cloud-based Trademark Search system replaced the legacy TESS interface, and applications route through the Trademark Center.

Analyze every hit under the multifactor likelihood-of-confusion standard, which each circuit implements with its own named factor list: the DuPont factors at the USPTO and Federal Circuit (In re E.I. du Pont de Nemours & Co., 476 F.2d 1357 (C.C.P.A. 1973)), the Polaroid factors in the Second Circuit (Polaroid Corp. v. Polarad Electronics Corp., 287 F.2d 492 (2d Cir. 1961)), and the Sleekcraft factors in the Ninth (AMF Inc. v. Sleekcraft Boats, 599 F.2d 341 (9th Cir. 1979)). The factors vary in wording but converge on the same intuition: similarity of the marks, relatedness of the goods, strength of the senior mark, evidence of actual confusion, and the sophistication of the buyers. For the analytical framework, see Likelihood of Confusion in Trademark Law; to run the factors methodically for a clearance file, use Working Up Likelihood of Confusion. The full clearance process is laid out in Trademark Clearance Searching: A Comprehensive Guide and, in operational form, Clearing a New Trademark.

Two traps a register-only search will not catch

Two recurring failures deserve their own warning, because a clean federal register can lull a team into a false sense of safety.

Trap one: common-law rights. Trademark rights in the United States arise from use, not registration, and a senior user who never filed can still block you within its geographic trading area. This is the Tea Rose-Rectanus doctrine, born of two Supreme Court decisions—Hanover Star Milling Co. v. Metcalf, 240 U.S. 403 (1916), and United Drug Co. v. Theodore Rectanus Co., 248 U.S. 90 (1918). A remote junior user who adopts in good faith may hold its own territory, but a senior common-law user can bar the newcomer where the senior user actually operates. A federal registration does confer nationwide constructive notice and priority, which can freeze a later common-law user's expansion (the Dawn Donut rule, Dawn Donut Co. v. Hart's Food Stores, Inc., 267 F.2d 358 (2d Cir. 1959)), but that protection runs forward from your filing—it does not erase a senior common-law user who was there first. And good faith is doing real work in this doctrine: a junior user who adopts with knowledge of the senior mark can forfeit the defense, as the Ninth Circuit held in Stone Creek, Inc. v. Omnia Italian Design, Inc., 875 F.3d 426 (9th Cir. 2017). See the doctrinal treatment in The Tea Rose-Rectanus Doctrine and Territorial Trademark Rights, the Stone Creek case study at How Knowledge of a Senior Mark Defeats the Good-Faith Defense, and the mapping guide The Geographic Reach of Common-Law Trademark Rights. The practical lesson is blunt: search beyond the register.

Trap two: related goods, not just the class. Confusion is judged by the commercial relationship between the goods and services, not by the administrative Nice class you happen to file in. A mark cleared in Class 25 (clothing) can still collide with a registrant in Class 18 (bags) if consumers would assume the two share a source. The class number determines filing fees and structures the search; it does not draw the boundary of your risk. Calibrate the search to real-world market overlap—the channels of trade, the buyers, the way the products sit next to each other on a shelf or in an app store—rather than to class lines.

Worked example (hypothetical). Larkspur Beverages clears "SUMMIT" for canned sparkling water on the federal register, finds nothing blocking in the beverage classes, and launches nationwide. Six weeks later it receives a cease-and-desist from a regional brewery that has sold "SUMMIT" craft beer across four mountain-state markets for eleven years and never registered. The federal-only knockout missed a senior common-law user with enforceable rights in its territory—and because beer and sparkling water travel the same retail channels to the same buyers, the "related goods" analysis makes confusion plausible even across nominal class lines. A full search that reached common-law sources would have flagged both problems before the cans were printed.

When clearance surfaces a conflict, the responses parallel the patent playbook: rebrand (a trademark design-around, Stage six), negotiate a coexistence or consent agreement, or, if the senior right is weak or abandoned, proceed with a documented risk assessment. If your launch is on a filing timeline, coordinate clearance with the priority mechanics in the From Brand Idea to Registered Mark toolkit.

Stage four: copyright and open-source clearance

Two distinct exposures live under this heading, and technology products usually underestimate the second.

Copyright clearance: verify the chain, not just the license

Copyright clearance covers everything you reproduce, adapt, distribute, or display—photographs, music, video, text, fonts, icons, UI assets, and third-party creative work of every kind. The naive version of clearance asks "do we have a license?" The competent version asks three harder questions:

  1. Does the license cover your actual use? Rights are sliced thin. A stock-photo license may permit web display but not merchandise; a font license may authorize desktop use but not embedding the glyphs in a distributed app or web font; a music sync license for a promo video may not cover a paid streaming ad. Confirm scope, not mere existence.
  2. Did the licensor actually own what it purported to grant? A vendor cannot license rights it never held. Chain-of-title defects propagate silently.
  3. Did rights in commissioned work actually transfer to you? This is the quiet killer. A freelancer's deliverable does not automatically belong to the commissioning party. Absent a qualifying work-made-for-hire relationship or a written assignment, the freelancer owns the copyright and you hold, at best, an implied license of uncertain scope. The work-made-for-hire doctrine turns on employment status under agency principles (Community for Creative Non-Violence v. Reid, 490 U.S. 730 (1989)) or on the work fitting one of the nine statutory commissioned categories and a signed writing, 17 U.S.C. § 101; independent-contractor code, design, and content usually satisfy neither without a backup assignment. Screen your contributor agreements now—see Determining Work Made for Hire and, for the invention-side analog that governs who owns your engineers' output, Employee Invention Assignment Agreements.

Copyright protects the exclusive rights to reproduce, adapt, distribute, perform, and display a work, 17 U.S.C. § 106, and infringement attaches to unauthorized exercise of any of them, § 501. "Found on the internet" is not a license. If enforcement or takedown work is in your future, the Copyright Enforcement and the DMCA toolkit maps the terrain.

Open-source clearance: the license obligations hiding in your dependency tree

For any software product this is usually the larger and more mismanaged risk. Modern applications are assembled, not written from scratch; a typical service pulls in hundreds or thousands of open-source components, transitively, each carrying license conditions that attach the moment you distribute. Two families matter:

  • Permissive licenses (MIT, BSD, Apache 2.0) impose light obligations—principally attribution and license-notice preservation, with Apache 2.0 adding a patent grant and notice requirements. Manageable, but not zero; failing to reproduce required notices is still a breach.
  • Copyleft licenses (GPL, LGPL, AGPL) are the ones that can reshape your business. Strong copyleft can require that, when you distribute a work that incorporates the licensed code, you make your own corresponding source available under the same terms. For a proprietary product, an inadvertent copyleft trigger—linking a GPL library into a shipped binary, or the AGPL's network-use provision reaching a hosted service—can convert a trade secret into a public disclosure obligation. The AGPL in particular closes the "SaaS loophole" by treating network interaction as a trigger.

These conditions are legally enforceable, and not merely as garden-variety contract breach. In Jacobsen v. Katzer, 535 F.3d 1373 (Fed. Cir. 2008), the Federal Circuit held that the terms of an open-source license can operate as conditions on the copyright grant, so violating them is copyright infringement—exposing the violator to injunctive relief and the copyright remedies, not just damages for breach. Later enforcement actions, including the Artifex Software, Inc. v. Hancom, Inc. litigation over the GNU GPL in the Northern District of California, confirmed that GPL obligations are enforceable in U.S. courts. The condition-versus-covenant line that Jacobsen drew is the same line that governs commercial software grants; the negotiation and drafting mechanics are worked through in Software License Agreements: Key Terms and How to Negotiate Them. (The scope of permissible reuse of software interfaces is its own contested question after Google LLC v. Oracle America, Inc., 593 U.S. 1 (2021), which found Google's reimplementation of the Java API declarations to be fair use on the specific facts—useful context, but not a license to copy code wholesale.)

The operational fix is a disciplined program: generate a software bill of materials (SBOM) that inventories every component and its license (including transitive dependencies), classify each license, flag incompatibilities and copyleft triggers, and remediate—swap, isolate, or relicense—before the code ships. Run the Open Source Compliance Playbook checklists, and read the risk map in Open Source Licensing Traps in Enterprise Software. For how these copyright and OSS exposures sit alongside patent and trade-secret protection in a single software stack, see How the Law Protects Software and, for app-specific mapping, Intellectual Property Strategy for Mobile Apps.

Worked example (hypothetical). Wren Analytics ships an on-premise data appliance. Its build pulls in a charting library that a well-meaning engineer swapped in during a late sprint; the library is licensed under the GPL. Because the appliance is distributed to customers and statically links the library, the copyleft condition attaches to the combined work—Wren's proprietary analytics engine now carries an obligation to offer corresponding source. Caught in a pre-ship SBOM review, the fix is a two-day swap to an MIT-licensed charting library. Caught by a customer's diligence team during an acquisition, it is a deal-blocking finding that forces a rushed rewrite under time pressure, with the acquirer's leverage at its peak.

Stage five: opinions of counsel and the willfulness shield

Clearance does double duty. Beyond avoiding infringement, a documented clearance investigation is the foundation of a defense against enhanced damages—and the amounts at stake are not marginal.

The enhanced-damages regime after Halo

Patent damages can be increased "up to three times the amount found or assessed" for willful infringement. 35 U.S.C. § 284. For years the Federal Circuit's rigid two-part test from In re Seagate Technology, LLC, 497 F.3d 1360 (Fed. Cir. 2007) (en banc), required a patentee to prove both an objectively high likelihood of infringement and subjective knowledge of that risk. The Supreme Court dismantled that framework in Halo Electronics, Inc. v. Pulse Electronics, Inc., 579 U.S. 93 (2016). Halo rejected Seagate's rigid objective prong, held that enhancement is committed to the district court's discretion, and refocused the inquiry on the infringer's subjective culpability judged at the time of the challenged conduct. Enhancement, the Court explained, is reserved for "egregious cases" of culpable behavior—the sort of "willful, wanton, malicious, bad-faith, deliberate" conduct "characteristic of a pirate." Halo, 579 U.S. at 103-04. Critically, Halo refused to let infringers "escape liability for enhanced damages" merely by conjuring a reasonable litigation defense years later; what matters is the state of mind when the infringement occurred.

The Federal Circuit has since clarified that willfulness under Halo centers on the accused infringer's subjective intent and knowledge of the patent, without a separate requirement to prove "wanton, malicious, or bad-faith" conduct at the liability stage—that heightened characterization goes to whether the court enhances, not to whether the infringement was willful. Eko Brands, LLC v. Adrian Rivera Maynez Enterprises, Inc., 946 F.3d 1367 (Fed. Cir. 2020); SRI International, Inc. v. Cisco Systems, Inc., 14 F.4th 1323 (Fed. Cir. 2021). The upshot for a launching company is that pre-suit knowledge of the patent, coupled with an absence of any good-faith basis for proceeding, is where willfulness exposure crystallizes.

What a competent opinion actually looks like

This is exactly where a documented clearance investigation plus a well-reasoned opinion of counsel earns its cost. Good-faith reliance on a competent opinion concluding non-infringement, invalidity, or unenforceability is strong evidence against willfulness. But courts scrutinize substance, not letterhead. A competent opinion:

  • comes from qualified counsel and analyzes the actual claims against the actual accused product, not a generic reassurance;
  • engages the real issues—claim construction, the all-elements analysis, the doctrine of equivalents, and, where relevant, validity;
  • reaches a reasoned conclusion, showing its work rather than asserting a result; and
  • predates or accompanies the launch. Because willfulness is judged by the infringer's state of mind at the time of the conduct, an opinion obtained only after suit is filed does little to negate pre-suit willfulness.

A conclusory "we see no problem" letter is worth close to nothing. And there is a strategic reason a non-infringement opinion is usually preferable to an invalidity opinion: invalidity must be proven by clear and convincing evidence against the statutory presumption of validity (§ 282(a); i4i), whereas non-infringement need only show that a limitation is missing. Where infringement risk is high but a solid non-infringement position exists, counsel may rest there; where both are genuinely in doubt, a mixed opinion addressing alternative claim constructions, or paired non-infringement and invalidity opinions, may be warranted.

The § 298 safe harbor and the privilege tradeoff

Two structural rules shape opinion strategy:

  • No affirmative duty, no adverse inference. There is no legal obligation to obtain an opinion, and under 35 U.S.C. § 298 the absence of one may not be used to prove willfulness or inducement. This is the statutory answer to the ostrich problem from Stage one: you are free to search and analyze without the trap of being punished for declining a formal opinion.
  • Reliance waives privilege. Asserting reliance on an opinion to defeat willfulness generally waives attorney-client privilege and work-product protection over that opinion and communications on the same subject matter—and courts read the same-subject-matter scope broadly. Many companies therefore separate opinion counsel from trial counsel to cabin the waiver, and they decide the privilege strategy before the opinion is written, not mid-litigation.

The trademark and broader parallel

The same good-faith logic runs through the other IP regimes. In trademark law, a documented clearance search and a written clearance opinion are the front-line proof of good faith that blunts arguments for willful infringement, enhanced recovery, and disgorgement of profits. Willfulness is no longer a strict precondition to a profits award after Romag Fasteners, Inc. v. Fossil, Inc., 590 U.S. 212 (2020)—the Court held that a plaintiff need not prove willfulness to recover an infringer's profits under 15 U.S.C. § 1117(a)—but the defendant's mental state remains a central equitable factor, which makes documented pre-adoption diligence more valuable, not less. The mechanics are worked through in Rebutting Willful Trademark Infringement: Clearance Searches, Opinions of Counsel, and USPTO Approval. For the full picture of what these exposures translate to in dollars across all four regimes, see What an IP Case Is Worth: A Guide to Intellectual Property Damages.

Worked example (hypothetical). Foxglove Devices learns during clearance of a competitor patent that arguably reads on its new controller. It commissions a written non-infringement opinion from outside patent counsel before launch; the opinion maps every claim limitation, identifies the missing "closed-loop feedback" element under the specification-supported construction, and concludes non-infringement. Foxglove keeps opinion counsel walled off from the trial team. Two years later the competitor sues and alleges willfulness. Foxglove's contemporaneous, reasoned opinion is powerful evidence that its pre-suit conduct was not "egregious," and the disciplined counsel structure limits the privilege waiver to the opinion's subject matter. Had Foxglove instead proceeded on a partner's verbal "looks fine to me," the same knowledge of the patent would read very differently to a jury.

Stage six: design-around, the payoff of clearing early

Design-around is not the clearance process failing—it is the clearance process working. The entire reason to clear early is to preserve cheap options, and a deliberate design-around is usually the cheapest option on the menu. When a blocking right surfaces, design-around is the engineering or branding response that makes the right stop mattering.

  • Patent design-around proceeds claim element by claim element. Eliminate or substantially change even one limitation of every independent claim and you defeat literal infringement. The essential caution, again, is the doctrine of equivalents: a change that swaps a limitation for an insubstantially different equivalent may not escape, so design-arounds should target function and means and result, and should exploit any prosecution-history estoppel that narrowed the original claims (Festo). A design-around that a competent opinion can bless is worth far more than one that "probably works."
  • Trademark design-around means choosing a genuinely distinct and distinctive name—moving up the Abercrombie spectrum toward suggestive, arbitrary, or fanciful, and away from the conflicting mark's zone. A rebrand caught pre-launch costs a naming exercise; caught post-launch it costs every printed surface, domain, and app listing.
  • Open-source design-around means replacing a copyleft-encumbered or incompatibly licensed component with a permissively licensed or proprietary equivalent, or isolating it across a process boundary so the copyleft condition does not reach your proprietary code.

In every case, document the design-around rationale. The record of a deliberate engineering or branding decision to avoid a known right is itself evidence of good faith, and it can anchor the non-infringement opinion that shields you against enhanced damages under Stage five. The narrative treatment of design-around within the FTO process lives in Clearing a Product for Launch.

Stage seven: clearance as a standing process, not a one-time gate

Clearance decays. New patents issue every Tuesday, new marks register continuously, dependency trees update with every build, and your own product drifts as features ship. A clearance done at launch is a snapshot, not a warranty, and treating it as permanent is how companies drift back into exposure. Build clearance into recurring operations:

  • Monitor. Watch competitor patent and trademark filings in your space; a watch service turns new registrations into early warnings rather than cease-and-desist surprises.
  • Re-clear at inflection points. Any material product change, new feature, new market entry, or rebrand should trigger a fresh, scoped clearance pass. A U.S.-only clearance says nothing about a European launch.
  • Keep the SBOM current. Regenerate and re-scan the software bill of materials on every release; an OSS clearance is only as good as the last build it examined.
  • Preserve the records. Retain search results, claim charts, opinions, and design-around memos. These are the good-faith evidence you will want if a dispute arises, and they are the diligence artifacts an acquirer or investor will demand.
  • Backstop with contracts and insurance. Revisit IP-infringement and media-liability coverage, and align indemnification and warranty terms in vendor and customer agreements so risk sits where it belongs. When you transact in IP or go through diligence, the IP Transactions Toolkit and the founder-facing Getting Diligence-Ready guide show how clearance records become deal assets.

Treat clearance as a living risk register tied to the product roadmap, owned by someone, and reviewed on a cadence—not a box checked once and forgotten.

The staged clearance checklist

A compact, repeatable version of the workflow above. Scale the depth of each step to the stakes.

Scope and decide (Stage one)

  • [ ] Define the product, brand, and codebase precisely, feature by feature.
  • [ ] Fix the relevant markets and jurisdictions; remember FTO is territorial.
  • [ ] Set the budget and the launch timeline; calibrate depth to exposure.
  • [ ] Install a clearance gate in the development process before design/brand/dependency freeze.

Patent FTO (Stage two)

  • [ ] Deconstruct the product into technical features and functions.
  • [ ] Search by CPC classification, keyword/synonym, assignee, and citation chaining.
  • [ ] Run USPTO Patent Public Search, Espacenet, Google Patents, and WIPO PATENTSCOPE.
  • [ ] Narrow to in-force patents in your markets (check term, maintenance-fee status, expiration).
  • [ ] Build claim charts; apply the all-elements rule to independent claims first.
  • [ ] Map under plausible claim constructions and flag doctrine-of-equivalents risk.
  • [ ] Triage hits into (a) clear, (b) opinion-level concern, (c) valid-and-infringed.
  • [ ] For category (c): license, design around, challenge validity (§§ 101/102/103, IPR), or accept documented risk.

Trademark clearance (Stage three)

  • [ ] Screen the candidate for distinctiveness on the Abercrombie spectrum.
  • [ ] Run a knockout search; advance survivors to a comprehensive/full search.
  • [ ] Reach beyond the federal register: state registers, common-law uses, domains, handles.
  • [ ] Analyze hits under the applicable likelihood-of-confusion factors (DuPont/Polaroid/Sleekcraft).
  • [ ] Check the "related goods" overlap, not just the Nice class.
  • [ ] Make the go/no-go/rebrand/coexistence call and document it.

Copyright and open-source clearance (Stage four)

  • [ ] Confirm each content license covers the actual use and that the licensor held the rights.
  • [ ] Verify work-made-for-hire or written assignment for all commissioned creative and code.
  • [ ] Generate an SBOM covering direct and transitive dependencies.
  • [ ] Classify every license; flag copyleft triggers and incompatibilities; remediate before shipping.

Opinions and willfulness (Stage five)

  • [ ] For category (b)/(c) patents, obtain a competent, written, pre-launch opinion.
  • [ ] Ensure the opinion analyzes actual claims against the actual product and reaches a reasoned conclusion.
  • [ ] Decide the privilege strategy up front; consider separating opinion and trial counsel.
  • [ ] Apply the same good-faith documentation to trademark and copyright clearance.

Design-around and ongoing risk (Stages six-seven)

  • [ ] Engineer or rebrand around blocking rights, element by element; watch equivalents.
  • [ ] Document the design-around rationale as good-faith evidence.
  • [ ] Stand up monitoring, re-clearance triggers, SBOM maintenance, record retention, and insurance/indemnity review.

Clearance will never reduce risk to zero; no responsible practitioner promises otherwise. What it does is convert unknown, uncapped, launch-day exposure into known, priced, and managed risk—made while the fixes are still cheap and the options are still open. A patent, a registration, and a copyright arm you against copyists. Clearance is how you make sure no one is arming themselves against you.

This toolkit is general legal information from the UpLaw digital law library, not legal advice, and it does not create an attorney-client relationship. Verify current search tools, fees, and authorities at the official sources, and consult qualified counsel for any specific launch.

Related UpLaw Resources

Related Guides

Related Checklists

Related Toolkits

Related Articles