Somewhere right now, an engineer is sketching a gadget on a napkin, a chemist is staring at an unexpected reaction, and a weekend tinkerer is convinced that the thing in the garage is going to change everything. All three share the same flutter of excitement and the same nagging question: how do I keep someone bigger and faster from simply taking this?

The answer the United States has offered since 1790 is the patent. It is one of the oldest and strangest deals in American law — a bargain struck directly in the Constitution, in which the government promises an inventor a temporary legal monopoly in exchange for the inventor telling the whole world exactly how the invention works. You give up your secret; you get the right to exclude. It is a trade of disclosure for exclusivity, and understanding that trade is the key to understanding everything else about patents.

This article is the trailhead. It is the broad, friendly overview of the entire U.S. patent system — the kind of single map you wish someone had handed you before you went looking for a lawyer. We will cover what a patent is and the constitutional reason it exists; how patents differ from copyrights and trademarks (people confuse these constantly, and the confusion is expensive); the three types of patents; the handful of statutory gates every invention must pass through; what can and cannot be patented after the Supreme Court's eligibility decisions; the journey of an application from idea to issued patent; the difference between a provisional and a nonprovisional filing; how to extend protection abroad; what it costs to keep a patent alive; how the public can challenge a patent even after it issues; and, finally, what rights a patent actually gives you and how you enforce them.

Because this is the overview, we will keep each leg of the trip readable rather than exhaustive, and we will point you toward the deeper guides as we go: Patent Basics — A Plain-English Guide for the gentlest possible introduction, Utility Patent Basics for the workhorse patent in detail, Patent FAQs for quick consumer-level answers, and the deep USPTO-process hub at Patent FAQs — Answers to Common USPTO Patent Questions. Let's begin where the system itself begins: with a sentence written into the Constitution before there were any patents to argue about.

The Constitutional Bargain: Why Patents Exist at All

Most areas of law grow up haphazardly — a statute here, a court decision there. Patent law is different. It rests on a single, deliberate clause that the Framers placed in the Constitution itself. Article I, Section 8, Clause 8 gives Congress the power "[t]o promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries." U.S. Const. art. I, § 8, cl. 8.

Read that clause slowly, because every word is doing work. The purpose is to "promote the progress" — patents are not a reward for being clever in the abstract; they are a tool to get more useful innovation into the world. The means is "securing . . . the exclusive Right" — a legal right to exclude others. The crucial limiter is "for limited Times" — the monopoly is temporary by constitutional design. And the split between "Authors" with their "Writings" (copyright) and "Inventors" with their "Discoveries" (patents) is baked right into the text. The Framers, several of whom were inventors themselves, understood that people invest more in invention when they can capture the value of what they create, and that society benefits when the recipe is eventually published for all to use.

Congress acted on that power immediately. The first Patent Act passed in 1790, and Thomas Jefferson — Secretary of State and a serial gadgeteer — personally reviewed early applications. The law was overhauled in 1836 (creating the Patent Office and the modern examination system), comprehensively revised in the Patent Act of 1952 (which is the backbone of today's law, codified in Title 35 of the United States Code), and most recently transformed by the Leahy-Smith America Invents Act of 2011 (the "AIA"), Pub. L. No. 112-29, 125 Stat. 284. The AIA is the single most important change in modern patent law, and we will keep returning to it, because it switched the United States from a "first to invent" country to a "first inventor to file" country — a change that affects how you should think about racing to the patent office.

The takeaway from the constitutional bargain is this: a patent is not a natural right you "own" the way you own a couch. The Supreme Court has described a patent as a "public franchise" — a grant of a specific right by the government to serve the public interest — even while the Patent Act treats issued patents as personal property that can be bought, sold, mortgaged, and licensed. See Oil States Energy Servs., LLC v. Greene's Energy Grp., LLC, 138 S. Ct. 1365, 1373–75 (2018); 35 U.S.C. § 261. That dual nature is not a contradiction; it is the deal. You disclose; the public lets you exclude others for a while; and when the clock runs out, your invention belongs to everyone. That framing explains nearly every rule that follows.

Patents, Copyrights, and Trademarks: Sorting the Confusion

Before going deeper, let's clear up the most common mix-up in all of intellectual property, because clients arrive with it constantly. Patents, copyrights, and trademarks are three different tools that protect three different things. Using the wrong one is like trying to fix a leaky pipe with a stapler.

A patent protects functional inventions — how something works or what it does. New machines, processes, chemical compositions, and ornamental product designs fall here. A patent is granted by the U.S. Patent and Trademark Office (USPTO), an agency of the Department of Commerce, and it is the only one of the three that you must affirmatively apply for and have examined before you get any rights at all.

A copyright protects original works of authorship — the expression in books, songs, films, photographs, paintings, and software code. Copyright arises automatically the moment you fix an original work in a tangible medium; you do not need to register it to own it (though registration brings big advantages and is a prerequisite to suing for infringement of a U.S. work). Crucially, copyright protects the form of expression, not the underlying idea or function — a principle codified in the idea/expression dichotomy of 17 U.S.C. § 102(b). The classic example, drawn straight from the USPTO's own materials: you can copyright a written description of a machine, but that copyright only stops others from copying your words — it does nothing to stop them from building and selling the machine itself. For the machine, you need a patent. If your work is a song, a manuscript, or website content, see our companion guides at Copyright Registration — A Comprehensive Guide and Copyright vs. Trademark — What Is the Difference?.

A trademark protects brand identifiers — words, names, logos, slogans, and other symbols that identify the source of goods or services and distinguish them from competitors. A trademark does not stop anyone from making the same product; it stops them from selling their product under a confusingly similar brand. Trademark rights arise from use in commerce and can be strengthened by federal registration under the Lanham Act. For more, see Trademark Basics and the lifecycle walk-through in The Trademark Process.

A simple worked example ties it together. Imagine our hypothetical company, Acme Hydration Corp., invents a water bottle with a genuinely novel internal filter mechanism, packages it in a distinctive teal bottle with a swooping logo, and writes a clever instruction manual. The filter mechanism is patent territory (it is functional and new). The swooping logo and the brand name are trademark territory (they tell consumers "this is an Acme bottle"). The distinctive teal bottle shape might even qualify for trade dress protection if it is nonfunctional and has acquired distinctiveness, an overlap explored in Design Patents vs. Trade Dress Protection for Product Configurations. And the instruction manual and marketing photographs are copyright territory (original expression). One product, four overlapping layers of protection, three different offices, three different bodies of law. Lawyers earn their keep partly by knowing which tool fits which problem — and often the answer is "more than one at once." For a comprehensive treatment of how these layers stack on a single product, see Protecting Your Mobile App — A Comprehensive IP Strategy Guide, which works through exactly this kind of overlap, and Legal Protection of Software.

The Three Types of Patents

People often say "patent" as if it were one thing. In the United States it is really three, and the differences matter enormously for cost, term, and what you can protect.

Utility Patents — The Workhorse

The utility patent is what most people mean when they say "patent." It may be granted to anyone who "invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof." 35 U.S.C. § 101. Those four categories — process, machine, manufacture, and composition of matter — were drafted to be broad. A "process" is a method or series of steps (think a manufacturing technique or a method of treating a disease). A "machine" is a device with moving or interacting parts. A "manufacture" (or "article of manufacture") is essentially any made thing that isn't a machine or a composition. A "composition of matter" covers chemical compounds and mixtures. The Supreme Court has memorably described the reach of these categories as including "anything under the sun that is made by man." Diamond v. Chakrabarty, 447 U.S. 303, 309 (1980) (quoting the legislative history of the 1952 Act).

Utility patents are the most powerful and the most expensive of the three, and they run for 20 years from the earliest effective U.S. filing date of the application (subject to maintenance fees, which we cover below). 35 U.S.C. § 154(a)(2). The overwhelming majority of patents granted each year are utility patents — so much so that practitioners use "patent," unqualified, to mean a utility patent. Because they are the default, they get their own dedicated guide; for the anatomy of a utility patent and the requirements in depth, see Utility Patent Basics.

Design Patents — Protecting How It Looks

A design patent protects the ornamental appearance of an article of manufacture — its shape, surface ornamentation, or visual design — rather than how it works. It "may be granted to anyone who invents a new, original, and ornamental design for an article of manufacture." 35 U.S.C. § 171. The contours of a smartphone, the pattern on the sole of a sneaker, the silhouette of a chair, the graphical layout of an icon on a screen — these are the stuff of design patents. The catch is in the word "ornamental": if a feature exists only because the function demands it (it has to be that shape to work), it is not protectable by a design patent. That functionality bar is its own developing body of law, explored in The Evolution of Design Patent Functionality.

Design patents are cheaper and faster to obtain than utility patents, contain (typically) a single claim defined almost entirely by drawings, and currently last 15 years from the date of grant (for applications filed on or after May 13, 2015; older design patents ran 14 years from grant), with no maintenance fees. 35 U.S.C. § 173. The law in this area has been shifting recently — most notably the Federal Circuit's en banc decision in LKQ Corp. v. GM Global Technology Operations LLC, 102 F.4th 1280 (Fed. Cir. 2024) (en banc), which overhauled how obviousness is analyzed for design patents by discarding the rigid Rosen-Durling test. We cover that development in The LKQ Decision — A Seismic Shift in Design Patent Obviousness Analysis, and the increasingly common practice of protecting one product with both a design and a utility patent in The Intersection of Design and Utility Patents.

Plant Patents — Yes, Really

The least-known member of the family is the plant patent, which "may be granted to anyone who invents or discovers and asexually reproduces any distinct and new variety of plant." 35 U.S.C. § 161. "Asexually reproduces" is the key phrase — it means propagating the plant by means other than seeds (grafting, budding, cuttings), so that the new plant is genetically identical to the parent. A breeder who develops a new variety of rose, apple tree, or ornamental shrub and reproduces it by cutting can seek a plant patent. (Tuber-propagated plants and plants found in an uncultivated state are excluded by statute.) Plant patents last 20 years from filing. They are a niche but genuinely important corner of agricultural and horticultural innovation, and they sit alongside two separate schemes: the Plant Variety Protection Act, administered by the Department of Agriculture, for seed-reproduced varieties, and ordinary utility patents, which the Supreme Court confirmed may also cover plants in J.E.M. Ag Supply, Inc. v. Pioneer Hi-Bred International, Inc., 534 U.S. 124 (2001).

One historical note worth keeping in your pocket: the Atomic Energy Act of 1954 forbids patents on inventions "useful solely in the utilization of special nuclear material or atomic energy in an atomic weapon." 42 U.S.C. § 2181(a). So the one thing you genuinely cannot patent for policy reasons is a better nuclear bomb. Everything else is at least in the conversation.

The Gates of Patentability: Sections 101, 102, 103, and 112

Here is the part that trips up nearly every first-time inventor. Having a great idea is not enough. To earn a patent, an invention must pass through a series of statutory gates, each one named for the section of Title 35 that defines it. Practitioners rattle these off as "one-oh-one, one-oh-two, one-oh-three, one-twelve," and once you know what each gate tests, the whole system snaps into focus. Think of it as a four-station obstacle course; your invention has to clear all four.

Gate One — Section 101: Eligible Subject Matter (and Usefulness)

Section 101 does two jobs. First, it asks whether the invention falls into one of the four eligible categories (process, machine, manufacture, composition of matter). Second, it imposes the requirement that the invention be useful — it must have a real-world utility and actually work. A perpetual-motion machine fails not because it isn't clever but because it doesn't operate; an invention that does nothing useful, or whose only asserted use is speculative, is not patentable. The Supreme Court has explained that the utility must be "specific and substantial" and not merely a hunt for a use to be found later. Brenner v. Manson, 383 U.S. 519, 534–35 (1966).

But Section 101 hides a famous trap. Even when an invention fits a statutory category, courts have long held that three things are not eligible for patenting no matter how they are dressed up: laws of nature, natural phenomena, and abstract ideas. Diamond v. Diehr, 450 U.S. 175, 185 (1981). You cannot patent gravity, you cannot patent a naturally occurring mineral or gene as it exists in nature, and you cannot patent a pure mathematical formula or an abstract concept. The rationale is that these are the "basic tools of scientific and technological work" that belong to everyone, and that monopolizing them would impede rather than promote progress. Mayo Collaborative Servs. v. Prometheus Labs., Inc., 566 U.S. 66, 71 (2012).

This is where modern patent law gets genuinely difficult, especially for software and business methods. In Alice Corp. v. CLS Bank International, 573 U.S. 208 (2014), the Supreme Court adopted a two-step test (building on Mayo) for sorting eligible inventions from ineligible abstract ideas. Step one: is the claim "directed to" a patent-ineligible concept (a law of nature, natural phenomenon, or abstract idea)? Step two: if so, does the claim contain an "inventive concept" — claim elements, individually and as an ordered combination, that add "significantly more" than the abstract idea itself, enough to "transform" it into a patent-eligible application? Simply taking an old abstract idea (in Alice, the long-familiar practice of intermediated settlement) and saying "do it on a generic computer" does not pass. Alice, 573 U.S. at 217–18, 221–26.

The Alice/Mayo framework has reshaped huge swaths of patent practice. It is why two skilled patent attorneys can look at the same software invention and disagree, in good faith, about whether it is patentable — and why claim drafting (the precise wording of what you claim) has never mattered more. Later Federal Circuit decisions such as Enfish, LLC v. Microsoft Corp., 822 F.3d 1327 (Fed. Cir. 2016) (claims directed to a concrete improvement in computer functionality survived step one) and DDR Holdings, LLC v. Hotels.com, L.P., 773 F.3d 1245 (Fed. Cir. 2014) (claims "rooted in computer technology" to solve an Internet-specific problem were eligible) chart a path through the doctrine: frame the invention as a technical improvement, not as automation of a human task. We devote an entire article to navigating it: Patent Eligibility After Alice — Strategies for Protecting Software and Business Method Innovations. For the related (and rapidly evolving) question of whether an invention generated with the help of artificial intelligence can be patented and who counts as the inventor, see AI-Generated Inventions — Who Owns What the Machine Creates and Artificial Intelligence and Inventorship — Global Perspectives.

Gate Two — Section 102: Novelty

The invention must be new. Section 102, as rewritten by the AIA, says in essence that you cannot patent something that was already "patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention." 35 U.S.C. § 102(a)(1). Everything in that bucket — every prior patent, journal article, product on the market, public demonstration, conference talk, YouTube video, or website that shows the invention — is called prior art. If a single piece of prior art discloses every element of your claimed invention, arranged as in the claim, your invention is "anticipated" and fails the novelty gate. Anticipation is a single-reference inquiry; combining two references is the province of obviousness, the next gate.

The America Invents Act made two structural changes here that every inventor should internalize, and they apply to any patent or application with at least one claim having an effective filing date on or after March 16, 2013. First, as noted, the U.S. is now first-inventor-to-file: as between two people who independently invent the same thing, the one who files at the patent office first generally wins, regardless of who scribbled it on a napkin first. The old "interference" proceedings that used to sort out who invented first are gone, replaced by narrow "derivation" proceedings aimed only at catching an applicant who stole the invention from the true inventor. Second, the relevant cutoff is the "effective filing date," which can reach back to an earlier provisional application or foreign priority application that adequately described the invention. 35 U.S.C. § 102(d).

There is one important mercy in the statute: the one-year grace period. A disclosure made by the inventor (or someone who got the information from the inventor) one year or less before the effective filing date does not count as disqualifying prior art against that inventor — and an inventor's own earlier public disclosure can even neutralize an intervening third-party disclosure of the same subject matter. 35 U.S.C. § 102(b)(1). In plain terms, if you publish or publicly demonstrate your own invention, you have up to twelve months to file a U.S. application before your own disclosure sinks you. But beware: most foreign countries have no such grace period and apply "absolute novelty," meaning any public disclosure anywhere before filing can destroy your foreign rights entirely. The safe rule, which good patent counsel will drill into you: file before you disclose. Loose lips at a trade show, an enthusiastic pre-launch demo, a Kickstarter page, or a published academic paper can quietly torpedo patent rights around the world. The on-sale and public-use bars are unforgiving — even a confidential commercial sale can trigger the on-sale bar, as the Supreme Court confirmed in Helsinn Healthcare S.A. v. Teva Pharmaceuticals USA, Inc., 586 U.S. 123 (2019).

Gate Three — Section 103: Non-Obviousness

This is the gate that does the heaviest lifting, and the one most applications struggle with. Even if no single piece of prior art shows your exact invention, you still cannot get a patent if the differences between your invention and the prior art "are such that the claimed invention as a whole would have been obvious . . . to a person having ordinary skill in the art" at the time. 35 U.S.C. § 103. That hypothetical "person having ordinary skill in the art" (lawyers abbreviate it "PHOSITA") is the patent law equivalent of the "reasonable person" — an imaginary, competent practitioner in the field who is presumed to know all the relevant prior art.

The framework for analyzing obviousness comes from Graham v. John Deere Co., 383 U.S. 1, 17–18 (1966), which directs courts and examiners to assess (1) the scope and content of the prior art, (2) the differences between the prior art and the claims, (3) the level of ordinary skill in the field, and (4) "secondary considerations" — objective evidence like commercial success, long-felt but unmet need, the failure of others, copying by competitors, and industry praise — that can point toward non-obviousness. Those secondary considerations are not an afterthought; they are often the inventor's best weapon, because they are real-world proof that the invention was not the trivial step a hindsight reading of the prior art might suggest. The Supreme Court later recalibrated the test in KSR International Co. v. Teleflex Inc., 550 U.S. 398 (2007), holding that obviousness analysis must be flexible and that combining familiar elements according to known methods is likely obvious when it does no more than yield predictable results. KSR rejected a rigid requirement of an explicit "teaching, suggestion, or motivation" to combine references and made it somewhat easier for examiners (and accused infringers) to reject or invalidate claims as obvious.

The USPTO's own homely examples capture the idea: merely substituting one known color for another, or scaling something up or down in size, is usually obvious and therefore not patentable. The trick of the trade is showing that your combination produces something unexpected — a result a skilled person wouldn't have predicted. Because obviousness is where so many applications live or die, we have a full guide on beating these rejections during prosecution: Overcoming Obviousness Rejections — A Comprehensive Guide to Section 103 Analysis.

Gate Four — Section 112: Disclosure (The Other Half of the Bargain)

Remember the constitutional bargain — disclosure in exchange for exclusivity? Section 112 is where you hold up your end. It requires that the application contain a written description of the invention (proving the inventor actually possessed what is claimed), that the disclosure be enabling (it must teach a person skilled in the art how to make and use the full scope of the invention without "undue experimentation"), and that it disclose the best mode the inventor knew for carrying out the invention. 35 U.S.C. § 112(a). The Supreme Court underscored how demanding enablement can be in Amgen Inc. v. Sanofi, 598 U.S. 594 (2023), invalidating broad antibody claims because the specification did not enable the entire genus the inventors had claimed — a sharp reminder that you cannot claim more than you have taught. Section 112 also requires that the claims — the numbered sentences at the end that legally define the boundaries of the invention — "particularly point[] out and distinctly claim[]" the subject matter. § 112(b). Claims that fail to inform skilled artisans, with reasonable certainty, of the scope of the invention are "indefinite" and invalid. Nautilus, Inc. v. Biosig Instruments, Inc., 572 U.S. 898, 901 (2014). (One AIA footnote worth knowing: while best mode is still required by the statute, the AIA eliminated failure to disclose the best mode as a basis for invalidating or rendering an issued patent unenforceable. 35 U.S.C. § 282(b)(3)(A).)

It helps to think of claims as the property lines of your invention. Just as a deed describes the metes and bounds of a parcel of land, the claims describe the metes and bounds of what the patent covers. Everything inside the claim lines is yours to exclude others from; everything outside is free for the taking. This is why claim drafting is the most specialized art in all of patent practice, and why two patents on similar inventions can have radically different value depending on how broadly or narrowly their claims are written.

To summarize the four gates: 101 asks is it the right kind of thing, and is it useful?; 102 asks is it new?; 103 asks is it more than an obvious tweak?; and 112 asks did you describe it well enough to teach the world and to mark your boundaries? Clear all four and you have an invention worth pursuing. Stumble at any one and the application stalls. Before you ever reach an examiner, the single best thing you can do is prepare a thorough, organized record of your invention; our guide How to Prepare an Invention Disclosure for Your Patent Attorney walks through exactly what to gather so your attorney can draft strong claims and spot problems early.

What Can and Cannot Be Patented: A Practical Map

Putting the gates together, here is the practical lay of the land.

You generally can patent: new and useful machines and devices; manufacturing and industrial processes; methods of treating diseases; chemical compounds, pharmaceuticals, and materials; many (but not all) software-implemented inventions, provided they are framed as concrete technical improvements rather than abstract ideas done on a generic computer; ornamental product designs (via design patents); and new asexually reproduced plant varieties (via plant patents).

You generally cannot patent: abstract ideas, mathematical formulas, and pure algorithms standing alone; laws of nature and naturally occurring phenomena (including, after Association for Molecular Pathology v. Myriad Genetics, Inc., 569 U.S. 576 (2013), naturally occurring DNA segments merely isolated from the body — though synthetic cDNA that does not occur in nature can be eligible); products of nature in their natural state; printed matter whose only novelty is the content it conveys; a mere idea or suggestion unaccompanied by a concrete, fully described embodiment (you cannot patent the wish for a teleporter, only a working teleporter you can actually describe and enable); and, by statute, atomic weapons.

The fuzzy middle ground — software, business methods, diagnostic methods, and AI-assisted inventions — is exactly where the most litigation and uncertainty live, all of it downstream of Alice and Mayo. If your invention lives in that zone, this is the moment to talk to a patent attorney rather than rely on a generic primer. The way the claims are written can be the difference between a valuable patent and an invalid one. For inventions in fields with their own regulatory wrinkles — cannabis, life sciences, and the like — the eligibility and utility analyses can take on additional dimensions that no overview can fully capture.

The Journey of an Application: From Idea to Issued Patent

Now let's walk the actual path an invention travels through the USPTO. The agency itself is sizeable — over 11,000 employees, the majority of them patent examiners with technical and legal training, receiving on the order of half a million utility applications a year. The examiners are organized into Technology Centers, each specializing in a field of technology, so that a battery invention is examined by someone who actually understands batteries. The examiner's bible is the Manual of Patent Examining Procedure (MPEP), a freely available USPTO compendium that spells out, in granular detail, how applications are examined — and which experienced applicants consult to anticipate how their claims will be treated.

Step One — The Invention Disclosure and the Search

Before filing, most inventors (or their attorneys) conduct a prior art search to gauge whether the invention is likely new and non-obvious. You can search yourself using the USPTO's public search tools or at a Patent and Trademark Resource Center library, but a preliminary search is exactly that — preliminary. The examiner will conduct a far more thorough search during examination and may reject claims based on prior art the inventor never found. A professional patentability search and opinion can save you from spending thousands of dollars prosecuting an application that was dead on arrival, and it gives your attorney the raw material to draft claims that thread between the closest references.

Step Two — Choosing Your Filing: Provisional vs. Nonprovisional

This is one of the most consequential early decisions, and it confuses people constantly, so let's be precise.

A provisional application is a lower-cost, placeholder filing introduced in 1995. It requires a written description of the invention (and drawings where needed to understand it), a cover sheet, and a fee — but no claims and no oath or declaration are required, and it is never examined on the merits. Its job is to plant a flag: it establishes an early effective filing date and lets you mark your invention "Patent Pending." A provisional automatically expires twelve months after filing and cannot be renewed. To capture its benefit, you must file a corresponding nonprovisional application (or a PCT application) within that twelve-month window, and the nonprovisional gets the provisional's filing date only for subject matter the provisional actually and adequately described under Section 112. (Provisionals are not available for design inventions, and — importantly — the twelve-month provisional pendency does not count against the eventual 20-year patent term, because the term runs from the nonprovisional filing date.)

Provisionals are popular for good reasons — they are cheaper, faster, and buy you a year to refine the invention, gauge market interest, and raise money before committing to the full cost of a nonprovisional. (For founders, that timing interacts directly with fundraising; see Preparing Your Startup for Capital Raising.) But there is a famous trap: a sloppy, thin provisional that doesn't really describe what you later claim provides no benefit at all. The priority date you think you locked in evaporates if the provisional didn't support the claims, and worse, your own intervening public disclosures may then become prior art against you. Treat the provisional seriously, not as a throwaway.

A nonprovisional application is the real thing — the application that actually gets examined and can mature into a patent. It must include a specification (the written description plus the claims), drawings where necessary to understand the invention, an oath or declaration signed by each inventor (a declaration, unlike an oath, need not be notarized), and the filing, search, and examination fees. The specification ends with one or more claims, presented in independent form (standing alone) or dependent form (referring back to and narrowing an earlier claim). The claims define the protection, so they are the heart of the document.

Step Three — Filing, Fees, and Entity Status

Applications are filed electronically through the USPTO's patent filing system, and filing a regular nonprovisional utility application on paper triggers a substantial extra non-electronic filing fee — so e-filing is effectively mandatory in practice for utility applications. The fee structure rewards smaller inventors. Small entity applicants (independent inventors, small businesses, and nonprofits that qualify under 37 C.F.R. § 1.27) pay most fees at a 60% discount, and micro entity applicants (who meet stricter income and prior-filing limits under 37 C.F.R. § 1.29) pay at an 80% discount — discounts that were deepened by the Unleashing American Innovators Act of 2022. Watch the claim counts, too: filing more than 20 total claims or more than three independent claims triggers "excess claims" fees. USPTO fees change periodically (adjustments typically take effect in the fall), so always check the current fee schedule rather than relying on a number you saw in an old article.

Once a complete application is received, the USPTO assigns it an application number and a filing date and sends a filing receipt. The filing date is the date the office receives a specification (with at least one claim, for a nonprovisional) and any necessary drawings — and locking in that date is the whole point of the AIA's first-inventor-to-file race.

Step Four — Publication

Under the American Inventors Protection Act of 1999, most utility and plant applications are published 18 months after the earliest effective filing date, whether or not a patent has issued. 35 U.S.C. § 122(b). After publication, the file wrapper becomes public. Publication has a silver lining for the applicant: it can trigger provisional rights, allowing a patentee who later obtains a patent with claims substantially identical to the published claims to seek a reasonable royalty for infringement that occurred between publication and grant — but only if the infringer received actual notice of the published application. 35 U.S.C. § 154(d). An applicant who is filing only in the U.S. and wants to keep the application secret can request non-publication at filing, but that option disappears the moment you file a counterpart abroad or under the Patent Cooperation Treaty, and you must notify the USPTO if you later do so.

Step Five — Examination and Office Actions

This is the long part. An examiner reviews the application, searches the prior art, and almost always issues an Office Action — a written rejection or objection identifying problems (most commonly novelty rejections under § 102, obviousness rejections under § 103, eligibility rejections under § 101, or clarity/written-description issues under § 112). This back-and-forth is called patent prosecution, and despite the scary name it has nothing to do with crime; it is simply the negotiation between the applicant and the examiner over what claims will be allowed.

The applicant responds by amending the claims, arguing against the rejection, or both. There may be a first (non-final) Office Action, a response, and then a "final" Office Action — though "final" is something of a misnomer, because there are still options after it, including a Request for Continued Examination (RCE) to keep prosecuting, an appeal to the Patent Trial and Appeal Board (PTAB), an examiner interview, or filing a continuing application (a continuation, divisional, or continuation-in-part) to pursue different or broader claims off the same disclosure. Watch out for a doctrine that quietly governs this stage: every amendment and argument you make becomes part of the prosecution history, and arguments you make to escape a rejection can come back to narrow your claims later under the doctrine of prosecution history estoppel, capping how far the doctrine of equivalents can stretch in litigation. Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722 (2002). Responding well to Office Actions is its own craft; we cover it in Responding to Patent Office Actions — Strategies for Overcoming Rejections. The whole process commonly takes a couple of years from filing to a final outcome, varying widely by technology center and backlog.

A vital warning during prosecution: applicants and everyone substantively involved owe the USPTO a duty of candor and good faith, including a duty to disclose material prior art they know about, usually through an Information Disclosure Statement. 37 C.F.R. § 1.56. Violating it can render the entire patent unenforceable for inequitable conduct — sometimes called the "atomic bomb" of patent law because it can destroy not just the tainted patent but related patents in the same family. The Federal Circuit tightened the standard in Therasense, Inc. v. Becton, Dickinson & Co., 649 F.3d 1276 (Fed. Cir. 2011) (en banc), requiring proof of both but-for materiality and specific intent to deceive. See Inequitable Conduct in Patent Prosecution — Navigating the Atomic Bomb of Patent Law and the evidence-gathering companion, Finding Evidence of Inequitable Conduct in Patent Prosecution.

Step Six — A Word on Double Patenting

One more prosecution hurdle deserves a mention because it surprises so many applicants and frequently surfaces in pharmaceutical and software portfolios. You cannot obtain two patents covering the same invention or its obvious variants — that is the doctrine of double patenting. There are two flavors: statutory (same-invention) double patenting under 35 U.S.C. § 101, and the judge-made obviousness-type double patenting (ODP), which bars a second patent claiming an obvious modification of what an earlier commonly-owned patent already claimed. The usual cure for an ODP rejection is a terminal disclaimer under 35 U.S.C. § 253, in which the applicant disclaims any term of the later patent that would extend beyond the earlier one and agrees the two will remain commonly owned. The point of the doctrine is to prevent an applicant from improperly stretching exclusivity past a single invention's rightful term. For the mechanics, see Double Patenting.

Step Seven — Allowance, Issue, and the Granted Patent

If the examiner is persuaded the claims are allowable, the office issues a Notice of Allowance, the applicant pays the issue fee, and the patent grants. At that moment the application's promise becomes an enforceable property right — the right to exclude others, for the remainder of the patent term, from making, using, selling, offering to sell, or importing the claimed invention in the United States. Issued patents enjoy a statutory presumption of validity, and anyone challenging the patent in court must prove invalidity by clear and convincing evidence — a meaningfully higher bar than the usual civil standard. 35 U.S.C. § 282; Microsoft Corp. v. i4i Ltd. Partnership, 564 U.S. 91 (2011).

How Long Does It Last? Term, Adjustments, and Maintenance Fees

A utility or plant patent's term is 20 years from the earliest effective U.S. filing date (not from the grant date). 35 U.S.C. § 154(a)(2). A design patent's term is 15 years from grant. 35 U.S.C. § 173. Because the 20-year clock for utility patents starts at filing but runs while the application sits in examination, delays at the USPTO can eat into the effective monopoly. Congress addressed this with two distinct mechanisms. Patent Term Adjustment (PTA) adds days to the term to compensate for certain USPTO delays during prosecution. 35 U.S.C. § 154(b). Patent Term Extension (PTE) restores time for products — like pharmaceuticals and medical devices — that lose marketable patent life to regulatory review by agencies such as the FDA. 35 U.S.C. § 156. These are not interchangeable; PTA is about office delay, PTE is about regulatory delay, and a single drug patent can claim both.

Keeping a utility patent alive is not free. The patentee must pay maintenance fees at three points — due at roughly 3.5, 7.5, and 11.5 years after grant (with a six-month grace period and surcharge) — or the patent expires early, falling into the public domain. 35 U.S.C. § 41(b). (Design and plant patents, notably, require no maintenance fees.) The fees escalate over time, a deliberate policy nudge: the system effectively asks patentees to "pay rent" to keep an invention exclusive, and to let go of patents that are no longer worth the cost so the public can use them sooner. Docketing these deadlines is a core function of any IP practice, because a single missed maintenance fee can vaporize a valuable patent, and reviving a lapsed patent (which requires showing the delay was unintentional) is neither cheap nor guaranteed.

Challenging a Patent After It Issues: PTAB and Reexamination

A patent is presumed valid, but that presumption is not the last word. One of the AIA's most consequential reforms was building out a suite of administrative procedures that let the public, and accused infringers in particular, challenge an issued patent at the USPTO rather than (or in parallel with) a federal court. These proceedings are heard by the Patent Trial and Appeal Board (PTAB) and have become a central front in patent disputes. The Supreme Court upheld the constitutionality of this entire scheme in Oil States Energy Servs., LLC v. Greene's Energy Grp., LLC, 138 S. Ct. 1365 (2018), reasoning that because a patent is a public franchise the government grants, the government may also provide for its reconsideration.

Inter partes review (IPR) is the workhorse. Any third party may petition the PTAB to cancel one or more claims of any patent as unpatentable, but only on novelty (§ 102) or obviousness (§ 103) grounds, and only based on patents and printed publications. 35 U.S.C. §§ 311–319. Post-grant review (PGR) is broader — it can attack a patent on essentially any ground, including eligibility under § 101 and disclosure defects under § 112 — but it is available only during the first nine months after a patent grants and only for patents subject to the AIA's first-inventor-to-file rules. 35 U.S.C. §§ 321–329. A specialized, sunset transitional program for covered business-method patents has expired, but the IPR/PGR architecture remains. Older and simpler, ex parte reexamination survives the AIA: anyone (including the patent owner) can ask the USPTO to re-examine claims in light of new prior art patents or publications, though after requesting it a third party largely steps aside. The AIA also added supplemental examination, which lets a patent owner ask the office to consider, reconsider, or correct information — a tool that, used properly, can immunize a patent against later inequitable-conduct attacks based on that information. 35 U.S.C. § 257.

Why does this matter to an inventor or business with a patent? Because owning a patent no longer means surviving only a court challenge. If your patent reads on a competitor's profitable product, expect that competitor to consider an IPR — proceedings that are faster and cheaper than district-court litigation and that have invalidated a meaningful share of challenged claims. Strong prosecution, a clean prior-art record, and well-supported claims are your best insurance. For the strategic interplay of PTAB challenges and infringement suits, see Comprehensive Guide to Patent Infringement Litigation.

Protecting Your Invention Beyond U.S. Borders

A U.S. patent is, by its terms, effective only within the United States, its territories, and its possessions. It does not stop anyone from making or selling your invention in Germany, Japan, or anywhere else. Patents are territorial, and there is no such thing as a single "international patent" that covers the world.

What does exist is machinery that makes seeking protection in many countries far more manageable. The most important is the Patent Cooperation Treaty (PCT), administered by the World Intellectual Property Organization (WIPO). A PCT application is a single "international" filing that, in effect, preserves your right to seek a patent in over 150 member countries while you decide where you actually want protection. It also yields an international search report and a preliminary (non-binding) opinion on patentability, useful intelligence before you commit real money. Crucially, the PCT does not grant a global patent — at around 30 months from your priority date, you must enter the "national phase" in each individual country (or regional office, like the European Patent Office) where you want a patent, and each examines your application under its own laws and grants its own patent. The PCT mostly buys you time and a unified, streamlined process for the early stages.

Two other treaties matter for timing. The Paris Convention gives you a 12-month priority window: file in one member country, and you have a year to file in others while claiming the benefit of that first filing date — which is one more reason the "file before you disclose" rule is so important, because that first filing date is also what neutralizes intervening prior art abroad. And the Hague Agreement offers an analogous centralized filing route for design protection internationally. The interplay of all this with the absence of a foreign grace period is exactly why international patent strategy needs to be planned before any public disclosure, not after. For the litigation side of going global, see Global Patent Litigation Strategies and the case studies in Global Patent Wars.

Who May Apply, and Do You Need an Attorney?

By law, the inventor — or a person or company to whom the inventor has assigned or is obligated to assign the invention (typically an employer) — may apply for a patent. When two or more people contribute to the conception of the invention, they apply as joint inventors. Here is a distinction people get wrong constantly: someone who only contributes money, or who merely manages the project or builds what the inventor designed, is not an inventor and cannot be named as one. Inventorship is about contributing to the conception of the claimed invention — the definite and permanent idea of the complete and operative invention. Naming inventors incorrectly — adding a boss who didn't invent, or omitting an engineer who did — can create real validity problems, though innocent errors can usually be corrected. If an inventor has died, a legal representative may apply; if an inventor refuses or can't be found, a co-inventor or the assignee may proceed via a substitute statement.

Because inventorship turns on conception, the question of whether an artificial-intelligence system can be a named inventor has become genuinely live. For now the answer in the United States is no: the Federal Circuit held in Thaler v. Vidal, 43 F.4th 1207 (Fed. Cir. 2022), that an "inventor" under the Patent Act must be a natural person, and the USPTO has since issued guidance on patenting AI-assisted inventions that requires a significant human contribution. We unpack that fast-moving area in AI-Generated Inventions — Who Owns What the Machine Creates. And because so many inventions are made by employees, who actually owns the resulting patent often turns on an assignment agreement signed years earlier; see Employee Invention Assignment Agreements — Drafting for Enforceability Across Jurisdictions.

Can you file a patent yourself, "pro se"? Legally, yes — an inventor may prepare and prosecute their own application. Practically, it is one of the riskier do-it-yourself projects in all of law. Patent drafting demands fluency in both the relevant technology and a dense body of statutes, rules, and case law, and the claims — the part that determines what your patent is actually worth — are notoriously hard to write well. A patent obtained by an untrained inventor may technically issue but protect almost nothing, because the claims are too narrow, too broad, or fatally flawed. As the USPTO itself cautions, there is "no assurance that the patent obtained would adequately protect the particular invention."

The professionals who do this work come in two flavors. Patent attorneys are licensed lawyers who have also passed the separate USPTO registration exam (the "patent bar") and hold the technical or scientific degree required to sit for it; they can both prosecute patents before the USPTO and handle litigation, opinions, and contracts. Patent agents have passed the same patent bar and can prepare and prosecute applications before the USPTO, but they are not lawyers and cannot litigate in court or, in most states, draft contracts like assignments and licenses where state law treats that as practicing law. Both must demonstrate good moral character and the necessary scientific and technical qualifications, and both are subject to USPTO discipline under 37 C.F.R. Part 11. The USPTO maintains a public directory of registered practitioners, but — by policy — it will not recommend or vouch for any particular one. A related caution: the agency cannot give you legal advice, opine on whether your idea is patentable before you file, or tell you whether you might be infringing someone else's patent; for that, you need your own counsel. Wondering which kind of legal professional you actually need for a given problem? Our directory Types of Lawyers — A Guide to Legal Specialties can help you sort it out.

A final, more somber warning that the USPTO itself emphasizes: the world of "invention promotion" and "invention marketing" firms is riddled with scams that prey on independent inventors — charging large fees for glossy "evaluations" and promises to get your invention to market, while delivering little of value. These firms are not registered to represent you before the USPTO, and federal law (35 U.S.C. § 297) requires them to make certain disclosures and gives defrauded inventors a cause of action. Be skeptical of anyone who guarantees a patent or a licensing deal, and do your homework before signing anything.

What a Patent Actually Lets You Do — and Not Do

Here is the single most misunderstood concept in patent law, so we will state it as plainly as possible: a patent gives you the right to exclude others, not the right to practice your own invention.

In the words of the statute and the grant itself, a patent confers "the right to exclude others from making, using, offering for sale, or selling" the invention throughout the United States, or "importing" it into the United States. 35 U.S.C. § 154(a)(1). Notice what's missing: the patent does not affirmatively give you permission to make, use, or sell anything. That sounds like a paradox until you see the example.

Suppose Acme Hydration Corp. patents an improved water-filter cartridge — but the cartridge necessarily uses a basic filter housing that Beta Filters, Inc. patented years earlier and that is still in force. Acme's patent lets Acme stop everyone else from copying Acme's improved cartridge. But Acme still cannot legally make and sell its own cartridge without a license from Beta, because building it would infringe Beta's still-valid housing patent. Each company holds a right to exclude the other; neither holds an automatic right to practice. This is why patent portfolios in crowded fields become a web of cross-licenses and negotiations, and why "I have a patent" never means "I'm free to sell this." It means "I can stop others from selling my claimed invention." Sorting out whether you are clear to launch a product is the job of a freedom-to-operate analysis; see Conducting Freedom-to-Operate Analysis for New Products. And if you want to understand the affirmative side — turning your right to exclude into revenue — see How to License Your Patent — From Valuation to Term Sheet.

The other half of this principle is that the USPTO has no role in enforcement. The office grants the patent and then steps back entirely. If someone infringes, it is up to the patentee to enforce the patent — by negotiation, by licensing, or by filing a civil lawsuit in federal court. The government will not police it for you.

Marking and "Patent Pending"

If you sell a patented product, you generally want to mark it — to put the public on notice of your patent rights — because marking affects how much money you can recover from an infringer. Under 35 U.S.C. § 287, a patentee who fails to mark patented articles generally cannot recover damages for infringement that occurred before the infringer received actual notice of the patent. In other words, marking lets the law presume the world was on constructive notice; failing to mark can shrink your damages to the period after you personally warned the infringer. (Note a wrinkle: the marking statute applies to patented articles; a patent with only method claims, which cover no physical product to stamp, is generally not subject to the marking requirement.)

Modern law allows virtual marking — placing the word "patent" or "pat." along with a freely accessible web address that lists the covered patent numbers, rather than stamping every number on the product itself, which is a boon for products covered by many patents and a way to keep notices current as patents issue or expire. Beware the flip side: false marking — deliberately marking an unpatented product as patented to deceive the public — is unlawful under 35 U.S.C. § 292, and competitors who suffer a competitive injury can sue. We cover the details and the traps in Understanding Patent Marking Requirements — A Practical Guide for Practitioners and Entrepreneurs and the broader notice question in Understanding Patent and Trade Dress Marking Requirements.

What about "Patent Pending"? That phrase simply means an application has been filed and is awaiting examination. It is a true and useful signal — it warns competitors that a patent may be coming and lets them factor that risk into their plans — but, importantly, it confers no enforceable rights on its own. You cannot sue anyone for infringing a "patent pending"; the exclusionary right exists only once the patent actually issues (with the narrow exception of the provisional-rights royalty discussed above, which still requires the patent ultimately to grant and the infringer to have had actual notice of the published application). And the old folk remedy of mailing yourself a description in a sealed envelope — the so-called "poor man's patent" — has never substituted for filing an application, and means even less now that the U.S. is first-inventor-to-file. If you want priority, file at the USPTO.

Enforcement Basics: What Happens When Someone Copies You

A patent is only as strong as your willingness and ability to enforce it. Patent infringement is governed by 35 U.S.C. § 271, which recognizes several flavors: direct infringement (making, using, selling, offering to sell, or importing the claimed invention without authorization, § 271(a)); induced infringement (actively and knowingly encouraging someone else to infringe, § 271(b)); and contributory infringement (supplying a material component especially made for an infringing use and not a staple article of commerce, § 271(c)). Inducement and contributory infringement both require knowledge of the patent and of the infringement, a point the Supreme Court emphasized in Global-Tech Appliances, Inc. v. SEB S.A., 563 U.S. 754 (2011).

The infringement analysis proceeds in two steps. First, the court engages in claim construction — figuring out what the claim language means as a matter of law. The Supreme Court held in Markman v. Westview Instruments, Inc., 517 U.S. 370 (1996), that claim construction is a question for the judge, not the jury, and the Federal Circuit's en banc decision in Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005), set out the methodology: read the claims in light of the intrinsic evidence (the claim language, the specification, and the prosecution history) before turning to extrinsic evidence like dictionaries and expert testimony. Courts resolve these disputes at a so-called Markman hearing, and the construction the court adopts frequently decides the entire case. For the mechanics, see Patent Claim Construction.

Second, the court compares the construed claims to the accused product or process. Under the all-elements rule, an accused product infringes a claim only if it contains every element of that claim, either literally or under the doctrine of equivalents (which catches products that make only insubstantial changes to dodge the literal words of a claim) — subject to the prosecution-history estoppel limits discussed earlier. Remedies can include injunctions (no longer automatic after eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006), which requires a four-factor equitable showing), monetary damages (at minimum a reasonable royalty, and potentially the patentee's lost profits), and — for willful infringement — enhanced damages of up to three times the award under Halo Electronics, Inc. v. Pulse Electronics, Inc., 579 U.S. 93 (2016). On the other side of the "v.", accused infringers commonly defend by arguing the patent is invalid (it never should have issued under §§ 101, 102, 103, or 112), that they do not infringe, that the patent is unenforceable (e.g., for inequitable conduct), or that they are licensed or protected by exhaustion. They will also frequently file an IPR at the PTAB to attack the patent on a parallel track.

This is a primer-level treatment; for the doctrine, see What Constitutes Patent Infringement?, and for the full arc of a patent lawsuit, see Comprehensive Guide to Patent Infringement Litigation — From Summary Judgment Denial to Post-Trial.

Patent vs. Trade Secret: The Other Choice

Before you race to the patent office, it is worth pausing on a strategic fork that surprises many inventors: you do not have to patent at all. The alternative is to keep the invention a trade secret.

The two regimes are mirror images. A patent requires you to publish your invention to the world in exchange for a time-limited 20-year monopoly. A trade secret requires you to conceal your invention and take reasonable measures to keep it confidential, in exchange for protection that can last forever — but only so long as the secret holds. Trade secrets are protected under the federal Defend Trade Secrets Act of 2016 (18 U.S.C. §§ 1836 et seq.) and state law (most states' versions of the Uniform Trade Secrets Act), and unlike patents they require no government filing and no examination. The catch: the moment a trade secret is lawfully reverse-engineered or independently discovered, it is gone, and you have no recourse against those who reach it honestly. The classic illustration is the formula for Coca-Cola, kept secret for well over a century — far longer than any patent could have protected it, but only because it could not be easily reverse-engineered from the finished drink.

The practical rule of thumb: if your invention can be readily reverse-engineered from a product you sell (most mechanical and electronic devices), a patent is usually the better choice, because secrecy won't survive the product's release. If your invention is a process or formula that customers can't deduce from the end product (and you can keep your workforce and systems locked down), a trade secret may be more durable and far cheaper. Note one AIA-era subtlety: because the on-sale bar can be triggered even by secret commercial use, an inventor who keeps an invention secret too long while exploiting it commercially may forfeit the patent option later — yet another reason to make the patent-versus-secret call early and deliberately. For a deep dive on the secrecy path, see Protection of Trade Secrets and Building a Trade Secret Protection Program From Scratch.

Key Takeaways

  • A patent is a constitutional bargain (U.S. Const. art. I, § 8, cl. 8): you disclose your invention to the public, and in return you get a time-limited right to exclude others — not a right to practice the invention yourself.
  • There are three types: utility (functional inventions, 20 years from filing), design (ornamental appearance, 15 years from grant), and plant (new asexually reproduced varieties, 20 years from filing).
  • Every invention must clear four statutory gates: § 101 eligibility and usefulness (no abstract ideas, laws of nature, or natural phenomena — see Alice and Mayo), § 102 novelty (it must be new), § 103 non-obviousness (more than an obvious tweak — see Graham and KSR), and § 112 disclosure (describe and enable it well enough to teach the world and define your claims).
  • The U.S. is first-inventor-to-file, and most of the world has no grace period — so the cardinal rule is file before you publicly disclose.
  • A provisional application is a cheaper placeholder that establishes a priority date and "Patent Pending" status for up to 12 months but is never examined; a nonprovisional is the real, examined application.
  • An issued patent is presumed valid but can still be challenged — in court, or at the PTAB through inter partes review and post-grant review.
  • Utility patents require maintenance fees at 3.5, 7.5, and 11.5 years, and patent term adjustment/extension can offset USPTO and regulatory delay.
  • Patents are territorial; the PCT, the Paris Convention, and the Hague Agreement help you preserve and stage international filings, but there is no single worldwide patent.
  • The USPTO grants patents but does not enforce them — enforcement is on the patentee, in federal court.
  • Trade secrets are the strategic alternative: forever (in theory) but only as long as the secret holds.
  • The claims are everything. Get experienced patent counsel before you file — and steer clear of invention-promotion scams.

Frequently Asked Questions

How long does it take and how much does it cost to get a patent? Plan on a multi-year journey — commonly two to three years (or more) from filing to a decision, depending on the technology center and backlog. Costs vary widely with complexity. A simple design patent can be relatively inexpensive, while a complex utility patent in a crowded field, prepared by an attorney and prosecuted through several Office Actions, can run well into five figures by the time it issues, plus government filing, search, examination, and issue fees (discounted 60% for small entities and 80% for micro entities), and later maintenance fees. Always check the current USPTO fee schedule. For quick, plain answers to cost-and-timing questions, see Patent FAQs.

Do I need a working prototype to file? No. You need to be able to describe the invention fully enough to enable a skilled person to make and use it (the § 112 requirement). A working model is almost never required by the USPTO. That said, building a prototype often helps you understand and describe the invention better — and the law does require that the invention actually work (be "useful" under § 101), which is why perpetual-motion machines fail.

Can I patent software, an app, or a business method? Sometimes — and this is the hardest area in modern patent law. After Alice Corp. v. CLS Bank, 573 U.S. 208 (2014), you cannot patent an abstract idea merely by saying "do it on a computer." But software that delivers a concrete technical improvement — to computer functionality, data processing, network operation, or a technological process — can be eligible if the claims are drafted carefully (see Enfish and DDR Holdings). The wording of the claims is often decisive. See Patent Eligibility After Alice and Protecting Your Mobile App — A Comprehensive IP Strategy Guide.

What's the difference between a patent, a copyright, and a trademark again? A patent protects functional inventions (how something works); a copyright protects original creative expression (the words of a book, the notes of a song, lines of code); a trademark protects brand identifiers (names, logos, slogans that tell consumers who made something). They come from different statutes and different offices. A single product can be covered by all three at once. See Copyright vs. Trademark — What Is the Difference? and Trademark Basics.

If I get a U.S. patent, am I protected worldwide? No. Patents are territorial. A U.S. patent only covers the United States, its territories, and its possessions. To protect your invention abroad you must seek patents in other countries — the Patent Cooperation Treaty and the Paris Convention's 12-month priority window make that process more manageable, but you still end up with separate national patents, each granted under its own country's law.

Does "patent pending" give me any rights? On its own, no. "Patent pending" simply tells the world you've filed an application. It's a useful warning to competitors, but you can't sue anyone for infringing a pending application. Enforceable exclusionary rights begin when the patent actually issues (with a narrow provisional-royalty exception that still requires the patent to ultimately grant and the infringer to have had actual notice). And the "poor man's patent" — mailing yourself a sealed description — has never been a substitute for filing.

Can someone challenge my patent after it issues? Yes. Even after grant, a competitor can ask the USPTO's Patent Trial and Appeal Board to cancel your claims through inter partes review (on novelty/obviousness grounds using patents and printed publications) or, within nine months of grant, post-grant review (on essentially any ground). They can also raise invalidity as a defense if you sue them. That is why a clean prosecution record and well-supported claims matter so much.

Should I file the patent myself to save money? You legally can, but it's risky. The claims determine what your patent is actually worth, and drafting them well requires real expertise in both the technology and patent law. A self-prepared patent may issue yet protect almost nothing. At minimum, consult a registered patent attorney or agent before filing, and start by organizing your invention with How to Prepare an Invention Disclosure for Your Patent Attorney.

What happens after my patent issues — am I done? Not quite. For utility patents you must pay maintenance fees at roughly 3.5, 7.5, and 11.5 years after grant, or the patent expires early. You should also mark your products to preserve damages, watch for post-grant challenges, and police and enforce the patent yourself — the USPTO won't do it for you. Think of an issued patent less as a finish line and more as an asset you have to manage.

Related Articles

This article provides general information about U.S. patent law and is not legal advice. Patent law is technical and fact-specific, and important rights can be lost through delay or a single misstep; consult a qualified, registered patent attorney about your particular invention and circumstances.