As I understand, the Alice decision set up a two-step rule for determining patent eligibility. Subsequent rulings in the U.S. (like McRO, Inc. v. Activision Publishing, McRO v. Bandai Namco Games America, Enfish LLC v. Microsoft, and Amdocs v. Openet Telecom) have been favorable to software patents and have clarified the landscape a bit.

But when drafting claims directed to software inventions, what are the best and/or most used strategies for avoiding Alice-based §101 rejections? I recall reading that a strong post-Alice strategy is to write the claims in means-plus-function language which refers back to specific hardware in the specification. Is this the best approach? Are there other approaches that don't limit the claims as much?

  • 1
    Ask yourself what your software does to the real world and put that info a claim. Or are you asking for more specific advice? Because I don't think there is any (yet). But I'd love to be proven wrong there anyways ;)
    – user18033
    Jan 20, 2018 at 16:31
  • @DonQuiKong, thank you for the suggestion! I'm looking for more specific advice (if it exists). I'm imagining strategies like (a) use means-plus-function, (b) make sure any described function refers to a specific piece of hardware (more specific than just "a processor"), (c) don't just describe the hardware, but also describe any physical connections/configurations that enable the specific functionality, etc. (As a disclaimer, the last one might not make sense! But those are the sorts of things I'm wondering about.)
    – jdpatent
    Jan 20, 2018 at 16:35
  • @DonQuiKong, by "what your software does to the real world" do you mean "what your software does to the hardware"? That could be useful guidance. It might suggest referring to data structures, etc.?
    – jdpatent
    Jan 20, 2018 at 16:36
  • the opposite (though I'm basically guessing here), I imagine that describing something on bit-level won't help, but rather how it makes something behave on the top level. Like, I don't know, - X happens and obviously it happens by software because it can't happen by hand - instead of - doing X in software on a pc with motherboard and means for storage etc. -
    – user18033
    Jan 20, 2018 at 16:53
  • @DonQuiKong, your comment "X happens and obviously it happens by software because it can't happen by hand" seems to be on the right track, and I think the key is unpacking the implications of "it can't happen by hand." The software must have novel and nonobvious implications at some hardware level--I think that's what the Alice decision intended to imply. At least, I remember reading something that suggested this fairly soon after the decision came out.
    – jdpatent
    Jan 25, 2018 at 22:24

1 Answer 1


In January 2019 the Director of the USPTO issued new guidance to the examining corps that caused them to be much less strict on giving 101 rejections. The 3620, 3680, and 3690 art units, which examine business methods, are still very strict due to the nature of the subject matter (Alice was about a business method).

Many patent applicants were happy with the changes to examination. Some, like I, are concerned that examiners will now allow patents that courts would find invalid under 101. To ensure that your eventual granted patent will be enforceable, it is important to understand how the courts apply 101 law.

In opinions of the Court of Appeals for the Federal Circuit (the top appeals court for most patent cases), an often repeated word is "specificity". Also, "results-oriented" claiming is not sufficient. My summary is, "don't just say what it does, say how it works".

Specificity does not require that the claim is long, and many long claims have been found ineligible under 101. Furthermore, machine claims are not significantly more eligible than method claims. Only 2 cases that I can find have been decided on 101 grounds based on claims in means-plus-function form. One eligible, one ineligible.

I maintain a growing document listing almost all claims analyzed by the Federal Circuit on 101 grounds since Alice and excerpts of the court's explanations for why they are or are not eligible. Though you might use some eligible claims as a template for your own drafting, I think that the document is mostly useful for study.

You can download my document here and a similar document for European Patent Office Board of Appeals cases on subject matter eligibility here.

Multiple resources are available here. The "October 2019 update" is a fairly small update to the January 2019 guidance. That is available here.

  • If you can provide links to the actual guidance documents that would be useful. Otherwise I appreciate your answering older unanswered questions.
    – Eric S
    Nov 23, 2020 at 17:15
  • This is a very helpful answer! I was thinking about it more from the prosecution side, but of course the enforcement considerations do bear on the prosecution strategy. To make an analogy, the PTO issued (at various points) instructions on the machine-or-transformation test as a result of the Bilski rulings. As a direct result, there were clear patterns in the types of 101 rejections that PTO examiners issued and the language patent attorneys used to get around them. It sounds like there's nothing analogous for the Alice case, and instead, the only cohesive guidance is specificity?
    – jdpatent
    Nov 24, 2020 at 2:11
  • In other words, I'd also be very interested in an analysis of allowed claims and the amendments they underwent to overcome a 101 Alice-based rejection. The advantage of inspecting patterns on the prosecution side are: (a) this practical knowledge will help a prosecuting attorney, and (b) there are far more granted patents than litigated patents. The disadvantage, as you say, is that the courts may not always agree with the PTO. (Then again, it seems that the patterns I'm asking about don't exist given the new guidance you reference.)
    – jdpatent
    Nov 24, 2020 at 2:16
  • I'm thinking about this more--if you could add a link to the new guidance, I'd very gladly accept this as the correct answer.
    – jdpatent
    Nov 24, 2020 at 2:23
  • @EricS Multiple resources are available here. uspto.gov/patent/laws-and-regulations/examination-policy/… The "October 2019 update" is a fairly small update to the January 2019 guidance. That is available here. uspto.gov/sites/default/files/documents/peg_oct_2019_update.pdf Nov 25, 2020 at 3:26

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .