The Category Is Not Dead, but the Odds Are Brutal

A business method patent claims a way of doing commerce: a pricing mechanism, a matching process, a settlement scheme, a workflow for approving something. US law does not exclude methods of doing business by name. What kills them is the judicial exception for abstract ideas, tightened decisively by the Supreme Court in Alice Corp. v. CLS Bank.

The practical result is visible in the numbers. Allowance rates in the art units handling business data processing sit far below the office-wide average, in some groups under 20 percent, while mechanical art units run above 70. If your invention is a business method, you are not filing an ordinary application with ordinary odds. You are filing into the hardest subject matter at the USPTO, and you should price and plan accordingly.

How the Eligibility Test Is Applied

Examiners run a two-step analysis. Step one asks whether the claim is directed to an abstract idea, which the office reads broadly to include fundamental economic practices, methods of organizing human activity, and mental processes. Hedging risk, intermediated settlement, matching buyers to sellers, and scheduling resources all land here.

If the answer is yes, step two asks whether the claim contains something more: an inventive concept that transforms the abstract idea into a patent-eligible application. Reciting a generic computer does not qualify. Neither does saying the method is performed over a network, on a mobile device, or in the cloud. The often-quoted formulation is that doing an old thing on a computer is not an invention.

So the entire question becomes whether your claim recites a specific technical improvement rather than a business outcome achieved with off-the-shelf computing.

The Dividing Line: A Technical Effect

Claims survive when the improvement is to the operation of a machine or to a technological process, not to a commercial result. Concretely, the surviving categories look like this:

  • Improvements to computer function. A data structure that lets a database answer a class of query without a full scan. A caching scheme that cuts memory footprint by an order of magnitude. The improvement is measurable in the machine, not in revenue.
  • Improvements to a technical process outside the computer. A scheduling method that reduces energy consumption in a physical plant. A routing method that measurably increases throughput of physical vehicles under stated constraints.
  • Specific technical implementations. A particular cryptographic protocol, a particular sensor fusion pipeline, a particular error correction scheme, claimed at the level of how it works rather than what it achieves commercially.
  • Business methods tied to real hardware. A vending, dispensing, kiosk, or logistics system where the claim recites structural elements and their interaction, not just software steps.

What does not survive: a better way to price insurance, a novel loyalty program, a marketplace with a new fee structure, a method of underwriting, a way to organize an approval chain. All of those may be excellent businesses. None of them is likely to be a patent.

How to Draft an Application That Has a Chance

The drafting choices are made before the first claim is written, and they matter more here than in any other technology area.

Lead with the technical problem. The specification should open by describing a deficiency in existing systems in technical terms: latency, storage, bandwidth, accuracy, failure rate, power. If the opening paragraphs describe a market inefficiency, you have framed the invention as abstract before the examiner reaches claim 1.

Claim the mechanism, not the goal. Recite the specific steps, the data transformations, the thresholds, and the architecture. Generic language such as determining, analyzing, and providing invites a step-two rejection. The general principles of claim construction that apply here are set out in patent claims explained.

Quantify the improvement in the specification. Concrete performance numbers give your attorney something to point at during prosecution and support the argument that the claim delivers a technical benefit.

Recite meaningful hardware. Not a generic processor and memory, but the sensors, actuators, interfaces, and physical elements that the method genuinely requires. Where the invention is essentially software, the framing options are covered in can you patent software in the US.

Expect a subject matter rejection anyway. Answering it is a specialized argument, and the mechanics of the exchange are described in how to respond to a patent office action. Budget for at least two rounds and possibly an appeal, which puts a realistic total in the $25,000 to $50,000 range rather than the $12,000 a mechanical case might cost.

Other Jurisdictions Are Not Friendlier

The European Patent Office excludes schemes, rules, and methods for doing business as such, and applies a technical character requirement that is, if anything, stricter than the US test. Features that only contribute to a business purpose are excluded from the inventive step analysis entirely. Chinese and Japanese practice similarly demand a technical solution to a technical problem. Filing the same business method broadly across territories multiplies cost without materially improving odds; the European cost structure alone is set out in the European patent application process and costs.

Even a Granted Business Method Patent Is Fragile

Grant is not the end of the eligibility question. A defendant can raise subject matter eligibility as a defense, and district courts have invalidated granted business method claims at early stages of litigation. Post-grant proceedings at the USPTO give challengers another route. A patent that would not survive a serious challenge has limited deterrent value and limited value in a licensing negotiation with a well-advised counterparty.

When There Is No Patent Path

Most business method inventions are better protected by other means, and the honest advice is usually to stop filing and start building the moat elsewhere.

Keep the algorithm, the model weights, the scoring logic, and the operational playbook as trade secrets, with the controls and tradeoffs described in trade secret vs patent. Register the brand, because in a service business the name carries more enforceable value than the method; start with how to trademark a product name in the US. Then compete on the things a patent could never have given you: data accumulation, network effects, integration depth, switching costs, and execution speed.

Decide Before You Spend

Projects House helps clients test whether a concept has a defensible technical core or whether the value sits in execution, and whether pairing it with real hardware opens a filing route that pure software does not. Describe your system through our contact form.