Skip to content
PatentBrief

§ 101 Eligibility · Software & Business Methods

The Alice Test

The two-step test that determines whether software, business methods, and algorithms can be patented. Most § 101 rejections in software patents trace to Alice — understanding it is essential for any software patent strategy.

The test in two sentences

Step 1: Is the claim directed to an abstract idea? Step 2: If yes, does the claim add an inventive concept that transforms the abstract idea into something patent-eligible — a specific technical improvement, not just “do it on a computer”?

The origin

Alice Corp. v. CLS Bank International (2014)

Alice Corporation held patents on a computer-implemented method of using a third-party intermediary (a “shadow ledger”) to reduce settlement risk between two parties in a financial transaction. The Supreme Court unanimously held the patents invalid under § 101.

The Court built on Mayo Collaborative Services v. Prometheus Laboratories (2012), which had established the same two-step test for laws of nature. Alice extended it to abstract ideas and made the framework the universal test for all three § 101 exceptions.

The holding (unanimous, 2014)

Claims directed to the concept of intermediated settlement — using a computer to implement a well-known financial practice — are patent-ineligible under § 101. Adding generic computer components (“a computer,” “a processor,” “a data storage unit”) to an abstract idea does not supply an inventive concept. “Simply appending conventional steps, specified at a high level of generality, [is] not enough to supply an inventive concept.”

The framework

The Alice/Mayo two-step test, step by step

Step 1

Is the claim directed to an abstract idea, law of nature, or natural phenomenon?

YES → proceed to Step 2 · NO → patent eligible

The court or examiner asks whether the claim, considered as a whole, is directed to one of the three judicial exceptions: (1) laws of nature, (2) natural phenomena, or (3) abstract ideas. If yes — and most software/business method claims involve an abstract idea — the claim moves to Step 2. If no, the claim is patent-eligible and the analysis ends here.

Examples

  • Abstract idea: 'intermediated settlement' (Alice Corp. v. CLS Bank, 2014)
  • Abstract idea: 'hedging risk' (Bilski v. Kappos, 2010)
  • Abstract idea: 'organizing information into categories' (Digitech Image Technologies v. Electronics for Imaging, 2014)
  • NOT abstract: 'specific database structure that stores data more efficiently' (Enfish, 2016)
Step 2A Prong 2

Does the claim integrate the abstract idea into a practical application?

YES → patent eligible · NO → proceed to Step 2B

Under the 2019 USPTO Revised Guidance, the next question is whether the additional claim elements integrate the abstract idea into a practical application — imposing meaningful limits such that the claim amounts to more than the abstract idea plus 'apply it.' This is the new gateway that saves many claims before they reach the demanding inventive concept analysis.

Examples

  • Practical application: claim recites a specific technical improvement to computer operation (Enfish self-referential table)
  • Practical application: claim applies the abstract idea with particular machines that are integral, not merely incidental (DDR Holdings — specific technical solution to internet-specific problem)
  • NOT practical: 'use a computer to implement intermediated settlement' — adding a computer to an abstract financial idea
  • NOT practical: claim covers all practical applications of the mathematical formula (preemption concern)
Step 2B

Does the claim add an inventive concept 'significantly more' than the abstract idea?

YES → patent eligible · NO → patent ineligible under § 101

If no practical application is found, the examiner asks whether the additional elements — individually or as an ordered combination — provide an inventive concept beyond what is well-understood, routine, and conventional in the field. This is the hardest step to satisfy. Generic computer components (processor, memory, network) performing their ordinary functions are never inventive concepts.

Examples

  • Inventive concept: specific, unconventional placement of filtering at the ISP rather than end-user (BASCOM Global, 2016)
  • Inventive concept: specific unconventional rules for morpheme-based lip sync animation (McRO, 2016)
  • NOT inventive: 'store data in memory and process it with a processor' — entirely conventional
  • NOT inventive: adding 'on the Internet' or 'on a smartphone' to an otherwise abstract method

Alice outcomes by technology

What survives — and what fails

Software / algorithms

✓ Survives Alice

  • Specific algorithm improving computer network security architecture (SRI International v. Cisco)
  • Specific data structure enabling more efficient database queries (Enfish)
  • Specific video compression method using unconventional hardware arrangement

✗ Fails Alice

  • 'Using a computer to analyze data and generate a report'
  • 'Storing records on a server and transmitting them over a network'
  • Machine learning model trained on data — without specific technical improvement to the model itself

Business methods / fintech

✓ Survives Alice

  • Specific cryptographic protocol implementing a new escrow mechanism (not just 'use encryption for escrow')
  • Fraud detection method using specific unconventional combination of data sources at a non-obvious point in the transaction flow

✗ Fails Alice

  • Intermediated settlement — Alice Corp. itself (2014)
  • Hedging commodity risk — Bilski v. Kappos (2010)
  • Reward points system implemented on a computer

Medical diagnostics / life sciences

✓ Survives Alice

  • Method of treatment using a specific drug at a specific dosage determined by a specific biomarker (eligible even if the biomarker discovery is a law of nature, if the treatment step adds a meaningful limitation)
  • Specific new combination therapy regimen not merely correlating a natural relationship

✗ Fails Alice

  • Method of detecting a natural phenomenon (correlation between metabolite levels and drug dosage) — Mayo Collaborative Services v. Prometheus (2012)
  • Comparing patient DNA to reference sequence to diagnose a condition — Athena Diagnostics v. Mayo Collaborative

Drafting strategy

How to draft claims that survive Alice

1

Frame the problem as a technical problem

Alice-vulnerable claims describe business or human problems. Surviving claims describe technical problems: latency, storage inefficiency, bandwidth, error rates, security vulnerabilities. The specification should lead with the technical problem the invention solves — not the business goal.

2

Claim specific technical means, not functional results

Claiming 'a system that identifies fraudulent transactions' is abstract. Claiming 'a system that compares transaction vectors using a trained convolutional neural network applied to a specific feature vector of [8 named features] to generate a confidence score' describes specific technical means. The specificity is what survives.

3

Avoid 'using a computer' as the only technical limitation

Adding 'on a computer,' 'via the internet,' or 'using a processor' to an otherwise abstract method is the hallmark of an Alice-failing claim. The computer must be integral to the invention — not just the medium of implementation.

4

Tie the abstract idea to a specific technical improvement

The Federal Circuit consistently upholds claims that improve computer functionality itself: faster processing, reduced memory, improved network efficiency, better security architecture. If the specification can show that the claimed method improves the technical functioning of a computer (not just uses a computer to improve a business process), the claim is much stronger.

5

Describe the algorithm at a level of detail that works

For software patents, the specification should describe the algorithm in enough detail that a POSITA could implement it — pseudocode, flowcharts, detailed prose description. Vague functional descriptions ('analyze the data,' 'apply machine learning') leave the claim vulnerable to both Alice and § 112(b) indefiniteness challenges.

FAQ

Alice test questions

What is the Alice test for patent eligibility?

The Alice test is a two-step framework for determining whether a patent claim is directed to patent-eligible subject matter under 35 U.S.C. § 101. It comes from Alice Corp. v. CLS Bank International (2014), where the Supreme Court held that software implementing an abstract idea of intermediated settlement was patent-ineligible. Step 1 asks whether the claim is directed to a patent-ineligible concept — an abstract idea, law of nature, or natural phenomenon. Step 2 (sometimes called the 'inventive concept' step) asks whether the claim adds 'significantly more' than the abstract idea itself — something that transforms the claim into a patent-eligible application. A claim that fails both steps is invalid under § 101.

What is an 'abstract idea' for purposes of the Alice test?

The Supreme Court has declined to define 'abstract idea' precisely, saying only that it includes 'fundamental economic practices long prevalent in our system of commerce,' 'methods of organizing human activity,' and 'mathematical relationships and mathematical formulas.' The USPTO's 2019 Revised Guidance provides three categories of abstract ideas: (1) mathematical concepts — mathematical relationships, mathematical formulas or equations, mathematical calculations; (2) certain methods of organizing human activity — fundamental economic principles, commercial or legal interactions, managing relationships or interactions between people; (3) mental processes — concepts performed in the human mind, including observations, evaluations, judgments, and opinions. A claim that is directed to one of these categories moves to Step 2.

What counts as an 'inventive concept' that saves a claim under Step 2 of Alice?

An inventive concept under Alice Step 2 is an element or combination of elements that is 'sufficient to ensure that the patent in practice amounts to significantly more than a patent upon the [ineligible concept] itself.' Well-understood, routine, and conventional activities — even if not in the prior art — do not qualify. What has been found to satisfy Step 2: (1) specific technical improvements to computer functionality (Enfish — a self-referential table that improves database efficiency); (2) specific algorithms that are not purely mathematical but have a technical character (McRO — specific rules for animating lip sync, not just 'use a computer'); (3) unconventional combinations that achieve a technical result in an unexpected way. What does NOT satisfy Step 2: using generic computer components to implement an abstract idea; applying an abstract idea 'with a computer'; limiting the abstract idea to a particular technological environment.

What types of software claims survive the Alice test?

Software claims that consistently survive Alice are those directed to specific technical improvements in computer functionality, data processing, or computing architecture — not merely using a computer to do what humans did before. Claims that have survived: (1) Enfish (2016) — a specific database structure (self-referential table) that stores data more efficiently than conventional databases; (2) BASCOM (2016) — specific placement of internet content filtering at the ISP level, not a conventional location; (3) McRO (2016) — specific rules for automating 3D character animation lip-sync, not just 'automate lip sync with a computer'; (4) Berkheimer (2018) — specific software architecture that reduced redundant processing and required less storage (but the inventive concept question was a fact issue). The pattern: claims survive when they recite a specific technical solution to a technical problem, not just a functional result achieved by a computer.

How does the USPTO's 2019 Revised Guidance change the Alice analysis?

The USPTO's January 2019 Revised Guidance restructured how examiners apply Step 2 of Alice into two sub-steps: Step 2A, Prong 1 asks whether the claim recites one of the three enumerated abstract idea categories (mathematical concepts, certain methods of organizing human activity, mental processes). Step 2A, Prong 2 asks whether the claim integrates the abstract idea into a practical application — i.e., whether the abstract idea is applied in a way that imposes a meaningful limit on the claim. If Step 2A Prong 2 is satisfied (practical application), the claim is patent-eligible without needing to reach Step 2B. Step 2B (formerly Step 2 of Alice) asks whether any additional elements individually or in combination provide an inventive concept beyond what is well-understood, routine, and conventional. The practical effect: more claims survive under the 2019 guidance because Prong 2 of Step 2A is an additional gate before the demanding Step 2B inventive concept analysis.

Related guides

Patent Eligibility (§ 101)Landmark § 101 CasesHow to Patent SoftwareKSR ObviousnessNovelty (§ 102)Non-Obviousness (§ 103)Patent Specification (§ 112)Landmark Patent CasesWhat Can Be Patented?