How Patent Infringement Works: Claim Types and Infringement Theories

Patent infringement is measured claim by claim. A product or process infringes a claim only if it includes every element of that claim, and different kinds of claims are infringed by different acts. A device claim is infringed by making or selling the device. A method claim is infringed by performing the steps. A system claim may involve components owned by different parties. Understanding those differences is the reason well-drafted applications include several claim types, each aimed at the parties and activities that matter.

This is a high-level overview for founders, engineering leaders, and business teams. It is general information, and whether any particular product infringes any particular patent depends on the facts.

The basic rule: every element of the claim

A patent claim is a list of elements. To infringe, an accused product or process must include each one, either literally or, in some circumstances, through an equivalent. Infringement under that second route is known as the doctrine of equivalents. Courts often ask whether a substitute performs substantially the same function in substantially the same way to achieve substantially the same result, or whether the differences are insubstantial, but no single formulation is the exclusive test. The doctrine’s reach can be limited by what was said and amended during prosecution, among other limits.

Missing even one element generally means no literal infringement of that claim. That is why claim drafting focuses so closely on which elements to include. Each added element narrows the claim and gives a competitor another way around it.

Three ways to infringe

Direct infringement. Under 35 U.S.C. 271(a), anyone who, without authority, makes, uses, offers to sell, or sells a patented invention in the United States, or imports it into the United States, directly infringes. Direct infringement does not require knowledge of the patent.

Induced infringement. Under 35 U.S.C. 271(b), a party that actively induces someone else to infringe can also be liable. Inducement generally requires knowledge of the patent and knowledge that the induced acts infringe, and there must be an underlying act of direct infringement by someone. Instructions, documentation, and product design that lead customers to perform a patented method are common examples of conduct examined under this theory.

Contributory infringement. Under 35 U.S.C. 271(c), selling or importing a component that is especially made or adapted for use in an infringement, that is not a staple article suitable for substantial non-infringing use, and that is a material part of the invention can be contributory infringement, again with knowledge and an underlying direct infringement.

There are other provisions as well. For example, importing or selling in the United States a product made abroad by a process patented in the United States can infringe under 35 U.S.C. 271(g).

Device and apparatus claims

A device or apparatus claim covers a physical thing: a circuit, a sensor, a phone, a controller. It is infringed when someone makes, uses, sells, offers to sell, or imports the device.

For product companies, these claims are often the most straightforward. They reach the manufacturer, and also importers and sellers, and infringement can often be shown by examining the product itself. The main drafting challenge is identifying which structural features are essential and which are incidental, so that the claim covers competitors’ versions rather than only your own.

System claims

A system claim covers a combination of components that work together, such as a client device, a server, and a database. Systems are natural for networked and software inventions, but they raise a practical question: who infringes when the components belong to different parties?

Courts have addressed this for some acts, and the answer can differ depending on whether the accused act is making, using, selling, or offering to sell the system. For use, the Federal Circuit held in Centillion Data Systems, LLC v. Qwest Communications International, Inc., 631 F.3d 1279 (Fed. Cir. 2011), that a party uses a system when it puts the system as a whole into service, controlling the system and obtaining a benefit from it, even if it does not own every component. Making a system, by contrast, generally requires that one party assemble all of its elements. As a result, a system claim spanning a customer’s device and a provider’s servers can be enforceable in some situations and difficult in others. Drafting system claims around components one party controls can make them easier to enforce, though the facts of each accused implementation still matter.

Method claims

A method claim covers a series of steps. It is directly infringed only when someone performs all of those steps. Selling a product that is capable of performing the method is not, by itself, direct infringement of a method claim, although the seller may face induced or contributory infringement claims if the conditions above are met.

Method claims are common for software and process inventions, and they can be valuable. Two issues need attention when drafting them. First, who performs the steps: if the steps are split between different parties, such as a user and a service provider, direct infringement may be hard to establish. That topic, divided infringement, has its own article in this series. Second, where the steps are performed, since acts outside the United States are generally not covered by a U.S. patent.

Computer-readable medium claims

A computer-readable medium claim, often written along the lines of “one or more non-transitory computer-readable media storing instructions that, when executed by one or more processors, cause the one or more processors to, individually or collectively,” perform specified operations, is directed to the stored software rather than its execution.

The “one or more processors” and “individually or collectively” wording reflects how courts read claim language. In Salazar v. AT&T Mobility LLC, No. 2021-2320 (Fed. Cir. Apr. 5, 2023), the Federal Circuit construed “a microprocessor,” followed by “said microprocessor” performing several functions, to require at least one microprocessor capable of performing all of the recited functions, so a system where different processors handled different functions did not meet the claim. Modern systems routinely split work across cores, chips, and servers, so drafters now commonly write processor and memory elements so that the recited operations can be performed by one or more components, individually or collectively. The same care applies to claims reciting a server, a controller, or a memory.

Computer-readable medium claims can provide protection complementary to method and system claims. Depending on the claim language and the accused activity, they may reach particular acts involving a medium storing software instructions, without requiring proof that a particular actor performed every step of a corresponding method claim. However, whether developing, storing, distributing, downloading, or providing software through a hosted service infringes a particular computer-readable medium claim requires separate analysis. The form of the claimed medium, the accused acts, and territorial limits can all affect the result, which is one more reason to include several claim types rather than relying on one.

Matching claims to the parties that matter

One approach to planning a claim set is to start with the question of who you would actually want to enforce against, and what that party does. For example:

• A competitor that manufactures a device: device and apparatus claims.
• A competitor that operates an online service: method and system claims written around the operations its servers perform.
• A competitor that develops software run by customers: computer-readable medium claims may help, along with method claims, subject to the limits above and the knowledge requirements for inducement.
• A component supplier: claims directed to the component itself, in addition to the larger product.

These pairings are illustrations, not rules. Other approaches may be more appropriate depending on the specific facts, such as the technology, how competitors’ products are built and sold, where activity takes place, the prior art, and the company’s business goals, and the right mix of claim types is best decided case by case.

Most companies prefer not to pursue their competitors’ customers, so claims that reach the competitor directly are often the most useful. That preference shapes which claim types deserve the most attention.

Other considerations

Territory. A U.S. patent covers acts in the United States, including importing. Inventions that involve components or steps in other countries can raise difficult questions, and foreign protection requires foreign patents.

Evidence. A claim’s practical usefulness depends partly on the ability to show that a product meets every element. Claims written around observable behavior are generally easier to investigate and prove than claims written around hidden internal operations.

Validity. Claims that are easy to show infringed but are anticipated or obvious in light of prior art provide little value. Breadth, provability, and validity have to be balanced together.

A practical takeaway

When reviewing a draft claim set, ask your patent attorney or agent which parties each independent claim would reach, what acts would infringe it, and how infringement would be shown. A claim set that includes device, system, method, and computer-readable medium claims, each written with a specific party in mind, is often more useful than several versions of the same claim, where the disclosure supports it.