Words in patent claims.

By Dennis Crouch

I have no idea what to make of the meaning behind the trends in this chart but I created it while working on another project. The chart shows a time series of the word count of each independent claim in issued patents averaged over all utility patents issued for a given week.

39 thoughts on “Words in patent claims.

  1. 39

    Wow, talk about an over-reaction.

    The simple point is that the data does not support the considerations being put forth.

    If you get so uptight over my pointing out mere conjecture, perhaps it is you that needs to take a break.

    Just a thought.

  2. 38

    Just saying, You might consider that

    In certain arts, practitioners add plenty of elements, all old and to be found in every infringing product, in order to multiply the difficulty of the examiner to find invalidating art and to argue obviousness.

    Of course, that does lead to corresponding infringement proof problems.

    But, the practice is, IMHO, “sharp.”

  3. 37

    anon, get a grip! It was germane to MAD Atty’s comment.

    You appear to have gone crazy with your posting. My advice is to take a week off, completely cold turkey. I don’t know how you find the time to post as much as you do, but try using that time to exercise instead. I bet that in a week’s time, you’ll feel much better.

    I feel for you, really. It’s like some sort of switch flipped in your head, and not for the better. Take care of yourself, anon.

  4. 36

    I understood the attempted differentiation – I only meant to point out that such was not germane to the data set presented.

    How many angels are dancing atop that pin?

  5. 35

    I only meant to differentiate that situation from the situation in where an applicant includes more words per claim at the time of initial filing.


  6. 34

    No need for that assumption IBP – the metric is merely on words in as patented claims (issued patents) – and there is no metric as to any effects of prosecution itself.

  7. 33



    I’m assuming that you are referring to those situations in which words are added to a claim by amendment, after a rejection for obviousness.

  8. 32

    One possibility for the upswing in word count around 2007 is that a ruling came out around that time that changed how the USPTO formulates rejections. For example, KSR v. Teleflex in 2007 made it easier for the USPTO (at least in their minds) to formulate an obviousness rejection without having to apply to TSM standard. It then became necessary for applicants to include more words in their claims to overcome these obviousness rejections.

  9. 31

    Don’t like business method patents? Talk to your Cingressmen. As learned (maybe not) by Justice Stevens, Congress has explicitly allowed Business Methods as patent eligible subject matter (as a category – not categorically, as we all know that there is no process, machine, manufacture, or composition of matter, at all that is CATEGORICALLY eligible).

    To date, the Supremes have not been brazen enough to actually rewrite sections of law and are content to write law by using a magic microscope to find implicit wording.

  10. 29

    OK, so the magic number has now changed from 100 to 75?

    You are awfully defensive about it.

    Just poking fun at you, you know.

    Poke fun at the people who write 200 word independent claims.

  11. 27

    PS DIP–

    I absolutely LOVE that list to which you linked.

    In looking through it, I found an application for a “process of reincarnation”, by George Knauer III of Naples, FL.

    He alleges that “This invention resulted from my combining Einstein’s Theory of Relativity and Newton’s Second Law of Physics.”

    It looks like he was pro se…

    To your question, I’m not referring to the situation where more words leads to more limitationss–I’m referring to the situation where each limitation is described using sufficient verbiage to permit ambiguation, or even just a broader interpretation itself.


    Thanks again for that link, it made my day, as I’m stuck working on Sunday for a Monday deadline.

  12. 26

    I’ll bite on this one… How does more words equate to more prior art? Shouldn’t it be the other way around?

    Generalizing: More words = more limitations = narrower scope = less prior art.

    For example:
    I claim X comprising A, B.
    I claim Y comprising A, B, C, D.

    Prior art disclosing A, B, C invalidates Claim X but not Claim Y.

  13. 24

    OK, so the magic number has now changed from 100 to 75?

    That was sure an easy way to dispense of the base-10 aesthetic, but you seem only to have transferred to the “divisible by 25” aesthetic…or maybe I was wrong about base-10, and it was “divisible by 25” all along!

    Just poking fun at you, you know.

    I agree that there should probably be a maximum economy of language in claims. Like someone else on this thread said, the more words, the more prior art.

  14. 23

    Yes, fight over a word or two and then get trapped in the new interminable wait between an RCE and the next office action.

  15. 22

    As for Just Saying’s comment about the conveniently 100-word limit, it’s just as absurd as Michel’s conveniently $2 billion PTO funding request.

    Only for attorneys that suck.

    A couple of points … I used the word “should” — not “must.” Second, that also applies to computer-related inventions (e.g., software).

    BTW … I as also being generous with 100 words. I think 75 is probably a better goal.

    However, I will say that any independent claim that has 150 words is overly verbose. 200 words … that is obese.

  16. 21

    I didn’t receive the notification that Just Saying was now King of the World.

    You didn’t? What’s your email address. I’ll make sure you are on the distribution list the next time.

    And to quote one of my favorite Mel Brooks’ lines, “it’s good to be the King.”

  17. 19


    You’re quite right about the bias, but it is a bias born not of prejudice, but of experience.

    And I do believe that so-called “software claims” are over-represented among badly-drafted claims, although I have only my anecdotal evidence for support.

    As for Just Saying’s comment about the conveniently 100-word limit, it’s just as absurd as Michel’s conveniently $2 billion PTO funding request.

    Long live the base-10 aesthetic.

  18. 18

    Less claims overall (from one of your previous charts), but more words per claim. Seems to me that will make my technical patent analyst job so much easier. The more someone says the more the chances at finding prior art. Works for me!

  19. 17

    Come now, IPB, software claims are by no means the only such claims, and your singling them out says far more than you think it says.

    Your continued mantra against them only belies your bias.

    Especially, after you just finished ragging on Just Saying.

  20. 16

    Come now, anon, I’m not “anti-software patent”

    I’m anti-poorly-drafted claim, and IMO many so-called software patents are of this species…

    …for the functional reason that if they were more succinct, they would be very easily seen as unpatentable under the various patentability conditions, by a self-interested and incompetent examining corps that needs to be hand-held every step of the way.

  21. 14

    Standard deviation lines on the chart might be helpful.

    Then again, if that showed that indeed this is just noise, what would the self- anointed anti-software patent QQers do?

  22. 13

    didn’t receive the notification that Just Saying was now King of the World.

    Right back at you, mr. anti-software patent.

  23. 11

    P.S. On the same theory [that really short and clear claims are harder to get allowed], it would be fun to see a “shortest claim contest.”
    E.g., the shortest successful claim I remember seeing [excluding maybe some chemistry] was an old patented method of microfinishing brake drums by “tumbling in broken walnut shells” [if I recalled that correctly].

  24. 9

    The old joke among PTO practitioners was the “span test.”
    [If the claim was longer than the examiners hand span, it should be perceived as having enough limitations to be allowable.]
    This in turn leads to “creative claim padding.” [Using longer expressions with more repetitions and other verbage without actually limiting the scope, etc.]

    But since that is nothing new, a more likely explanation for this chart trend is the decline in simple short “means-function” claims as the patent bar gradually caught on to how narrowly the CAFC was interpreting them, and their increased spec. requirements.

  25. 7

    I doubt it. Why would you expect a correlation between the average # of words in claim 1 and the number of patents in your sample?

  26. 6

    “More words = intransigent USPTO”… I agree. But it’s not equal. “More words” is a result of “intransigent USPTO.

    Therefore, Intransigent + PTO = more words.

  27. 5

    The rise from ’04 onwards might be due to more software claims, which are verbose carp due to obfuscation, inarticulation, or both.

  28. 4

    “A nice, tight independent claim should be about 100 words.”

    I didn’t receive the notification that Just Saying was now King of the World.

  29. 3

    I might be interesting to overlay patent issuance trends (# of patents issued for each corresponding week) and litigation trends (# of patent cases filed each week). It might also be fun to calculate the R squared for words vs. # of patents and words vs. # of patent cases.

  30. 2

    After years of the “just say no” inNOvation office stall game, when examiners would give my clients some patent coverage by allowing 2nd or 3rd tier dependent claims rewriting all the limitations from the above claims and placing them in independent form, almost always this has been taken with a continuation filed to continue fighting for full claim scope deserved.

    So in an odd way, blame this on Kappos (even though everyone knows the root cause was the “Quality = reject” mindset).

  31. 1

    More words = intransigent USPTO and/or lesser quality patent practitioners.

    A nice, tight independent claim should be about 100 words. When you get to 200 words/independent claim, you’ve got a lot of words that aren’t necessary to get around the prior art.

    What is happening is that although we are seeing a higher rate of patent allowances, this is because Examiners are allowing dependent claims (or offering them up in Examiner’s Amendments) and clients are taking them instead of fighting for broader independent claims. I see it all the time.

    This is a good illustration as to why you cannot look at any single metric at the USPTO and get a good understanding as to what is really happening.

Comments are closed.