Prosecution Pendency Thoughts

The first chart below shows one view of how patent prosecution pendency is generally improving.  In particular, the chart shows the percentage of patents that took more than three years from filing to issuance.  As is apparent, that percentage has plummeted in President Obama’s patent office (Kappos-Rea-Lee).  Unlike a few years ago, most US patents now issue within three years of filing (not counting priority claims).

The second chart tells a different story by looking at the oldest prosecution files and seeing that those have grown substantially over this same time period.  In particular, the second chart shows the percentage of patents that took more than seven years to issue (filing date to issue date).  For the most part, these long-pendency prosecutions involved PTAB appeals.


39 thoughts on “Prosecution Pendency Thoughts

  1. 6

    A much more telling set of graphs would show pendency in each technology type: mechanical, chemical, electrical, computing, business methods. I think we can all guess how those graphs would come out.

    1. 6.1

      Much the same as at the EPO, David, I would guess.

      The EPO has a shortage of good tri-lingual examiners in EE and computing. Where are all the trained tri-lingual engineers who could have been examining? Inventing, I suppose.

      What would interest me is whether there is any match between long pendency and increased numbers of requests for expedited examination. For, at the EPO, one can ask for accelerated examination and normally get what you ask for, even though there are no fee to pay and no reason need be given. Anybody then who wants early issue only has to ask.

      Of course, there are many art areas where early issue is expressly not wanted by the Applicant. In such areas, the dilemma for the PTO is whether (in the public interest) to give Applicant early issue, whether it wants it or not.

      1. 6.1.1

        Well, the situation is different here. We have plenty of EE and CS examiners – the corps seems fully staffed – it’s not a lack of manpower. Rather, the USPTO exhibits an insistence on protracted examination of software, where the claims are rejected over and over again in view of different combinations of references. Meanwhile, in EE and other areas, the claims sail through – as long as they recite any novel element, you get your allowance after one or two office actions, and sometimes a first-action allowance.

        Despite the vast size of U.S. patent law, not one letter of the law states whether an application with claims that are distinguished over the references of record should be allowed, or whether the examiner should perform a new search. It is 100% discretionary, defined solely by the culture of the art unit: i.e., where the director and SPE want to set the allowance rate. High allowance rate = easy allowances; low allowance rate = protracted examination.

        The only thing more surprising than the complete subjectivity of this crucial decision, is that everyone just accepts it.

  2. 5

    In other news, one of the worst patentees just saw five of its junk patents bite the dust. Along the way a Federal judge needed to explain to the patentee that an old television is no different from an old computer from the perspective of section 101. Nobody should be surprised by that, of course. What’s surprising is that a patentee could float the counter-argument in 2015 and not be sanctioned.

    link to

    Most satisfyingly, at least one of those patents has been on my short list for many years as exemplary of the sort of completely insane (and sanction-worthy) “arguments” that the defenders of the worst patents ever love to trot out, e.g., that identifying a subset as “not X” is “completely different” from identifying the remainder of the set as “X” (LOLOLOL). In this case, the subset in question was “unwatched videos.”

    “We are disappointed in, and strongly disagree with, the Court’s decision finding the five patents invalid and plan to appeal that decision,” said Rovi’s chief of IP Samir Armaly in a statement to investors.

    Yo “investors”: Samir is blowing sunshine up your behinds. These claims are junk and they aren’t going to be resurrected. Kiss them goodbye and find a better place to spend your money.

    1. 5.1

      And another update on everybody’s favorite federal district court:

      link to

      Two weeks after online retailer Newegg filed a petition complaining about “excessive and unreasonable” delays in getting a final judgment in its patent case, the judge in that case has handed Newegg a big win.

      In an order published yesterday afternoon, US District Judge Rodney Gilstrap ruled (PDF) that Newegg doesn’t infringe a patent belonging to TQP Development, notwithstanding a 2013 jury verdict that granted $2.3 million to TQP, a “patent troll” with no business other than patent licensing….

      …Gilstrap appears to blame a clerk’s error for the oversight.

      1. 5.1.1

        Would you please stop clogging up threads that are unrelated to 101 with your links to Arstechnia and other off topic nonsense? Go start your own blog if you want to rant about 101 every day.


          threads that are unrelated to 101

          I’ll bet you that quite a bit of that “backlog” is related to 101.

          More importantly, if you are really worried about “clogging” threads, Bluto, I’d suggest that you stop whining about two bits of news and avert your eyes downward. Good luck.


            I’ll bet you that very little of it is.

            Also, tra*sh is tra*sh. Whether it’s two bits or 100, it’s still an eyesore.

      2. 5.1.2

        “While SiRF may remain good law even in light of Alice, it seems apparent that a machine must do more than simply “play a significant part in permitting the claimed method to be performed” in order to supply the required inventive concept. Given Alice’s rejection of a “general purpose computer” as sufficient to establish patentability,

        a patentee must show that the machine itself is a particular machine, and not just that a generic machine is being used for a particular purpose.

        As discussed above, if identifying a particular function of a machine were enough to establish patentability under the “machine or transformation” test, then any patentee could evade invalidity by using specific-sounding language to describe a general purpose computer. For instance, the
        Alice patentee could describe the recited general purpose computer as a “settlement risk mitigating computer,” without which the claimed method could not be performed. To allow a patentee’s creative description of his claimed computer to govern patent eligibility
        would be to turn the § 101 analysis into a draftman’s art. Such an approach directly contradicts the purpose underlying § 101.”

    2. 5.2

      Indeed, MM. It now looks like the fastest and cheapest way to knock out 101 junk is to take it to court. The courts are fully on board and they “get it.”

  3. 4

    Graph with 3year lag compared to graph with 7year lag. Shape of 2nd graph matches first half of 1st graph. Writer describes it as “telling a different story.”

    The same story (that applies to both graphs) is that the number of filings increased in 2001/2002 and the office didn’t start huge hirings until 2007/2008.

    Who wants to bet that the 2016 number for the second graph will be lower than 2015, just like the 2012 number is lower than the 2011 number for the first graph? I mean this is kind of basic logic – patents that don’t pend at the 3yr mark can’t pend at the 7yr mark (though the difference in magnitude of the two graphs 30-60% versus 1-5% means it won’t track exactly). We’re engineers, right?

    1. 4.1

      “Graph with 3year lag compared to graph with 7year lag. Shape of 2nd graph matches first half of 1st graph. Writer describes it as “telling a different story.”

      The same story (that applies to both graphs) is that the number of filings increased in 2001/2002 and the office didn’t start huge hirings until 2007/2008.”

      The differences in lag IS the different story.

      If applications were being handled on a first come first server basis, then the 7 year pendency would be dropping faster than the 3 year pendency. If they being handled independent of pendency then the drop rates would be even. What these graphs are showing is a last come, first served sort of prosecution.

      1. 4.1.1

        If applications were being handled on a first come first server basis, then the 7 year pendency would be dropping faster than the 3 year pendency.

        Not necessarily. It’s reasonable to assume that, generally speaking, the longer-pending apps (and their owners) are burdened with issues that slow the entire process down.

        The idea that all applications are created equal or even capable of being treated equally is a fantasy.

      2. 4.1.2

        What these graphs are showing is a last come, first served sort of prosecution.

        What? It’s not showing that at all. There is only one year included in the 2nd graph (2015, describing cases filed 2008ish) where the number goes down, and that’s because 2008 is the first year that the backlog came down (the first year the corps’ production outpaced filings). So you CANT make a statement as to the trend of the 7 year pendancy dropping because there is no trend expressed yet.

        Contrary to what is being said the graphs have only 6 years of overlap. The first graph’s 2005-2011 describe the same pool of cases that the second graph’s 2009-2015 describe. The shape of those graphs are the same – i.e. An application that hasn’t been finally disposed of in 1 or 2 rounds (and is still pending at the 3yr mark) has a 10%ish chance of not being finally disposed of in 3-5 rounds (or 2-3 rounds with a ptab appeal and thus pending at the 7yr mark), and that is true whether the application was filed in 2002, 2003, 2004, 2005, 2006, 2007 or 2008.

        The only thing the graphs are showing is that the higher backlog meant less cases had gone throw multiple rounds of examination prior to 2008.

        If applications were being handled on a first come first server basis, then the 7 year pendency would be dropping faster than the 3 year pendency. If they being handled independent of pendency then the drop rates would be even.

        Your logic doesn’t make any sense at all. First come first served applies to the backlog, not to cases in active examination. Are all cases at the 3yr pendancy mark past the backlog and in active examination? Yes. Are all cases at the 7yr mark in active examination? Yes. So, just as you say, they are both being handled independent of pendancy and you would expect the change in rates to be even.

        What you would need to show that cherry picking is occurring is a graph of first action or first round allowances, or something like a 1yr pendancy graph. But by 3 years everything, good and bad apps, has gone into examination and you can’t tell if cherry picking is going on at all.

      3. 4.1.3

        BTW this is a common example of how people can get confused because they’re presented with poorly described information and don’t understand the inner workings of the office.

        When an application is filed a computer determines which art unit the subject matter belongs in. When I run out of cases in my docket my boss takes the oldest cases that belong to our art unit and gives me a dozen. At that point nobody has judged how “easy” the cases are. Now I get these dozen cases and I can bother to check which is easy and cherry pick from them. I can do the last case if I want (I won’t get the docket management credit for doing it, but maybe I don’t care about that). Here’s the thing – that last case, since its only 12 cases after the first case, has a pendancy difference of a few months max. And thats for a slow art. You know how many applications are filed in some art areas daily? It’s not impossible to get an entirely new docket where you can’t cherry pick, not cause you don’t want to but because literally every application has the same filing date.

        This idea that I can figure out an easy case was filed 6 months ago and pull it and do it because I want to is pure fabrication. I have a dozen cases to pick from and they’re all within a few months of each other.


            Well what I do is sneak into my boss’ office, give myself all the Google assigned cases and send out first action allowances. Then I spend the rest of the quarter in the Bahamas with Michelle, smoking cigars lit with $100 bills and preparing improper 101 Abstract Idea rejections to IBM while playing poker with our secret SAWS II contingent.

  4. 3

    Longer pendency should result in even fewer 12-year maintenance fees being paid to the PTO. Why pay $7500 for only 1-2 two years of patent term?

  5. 2

    I think the idiom that applies here is “picking the low hanging fruit”.

    Someone wants to improve the numbers, so they are allowing that appear allowable with out much investigation to get them out the door quickly. At the same time they are getting hammered for “low quality patents” and so anything that will take some time to investigate they are just saying no and dumping it on the Appeals Board.

    We call that a “no brainer.”

    1. 2.1

      And if they are picking the low hanging fruit? Why is that bad?

      Can,t an applicant choose how low to hang his fruit ?

      1. 2.1.1

        Its only bad if you are the guy with that invented something they don’t immediately understand and your application is made to wait while others are addressed out of turn.


          “they” Les? Who are you using to write your patent application Les, that is unable to write text that a specialist PTO examiner can understand. How much are you paying, for an hour of his time? It ain’t worth it.


              The Pollyanna Principle is a positivity bias, right? What bias do you discern? Are you suggesting that examiners at the PTO are routinely incapable of understanding the specifications of as filed patent applications in their designed art areas and that I’m naïve to suppose that they do ?


                Idk about in the whole USPTO bro but yes, over in my area the inventions and often the newer art is above a lot of heads over here. It is one of the most complicated arts on the face of the planet though so I mean, it can be forgiven to an extent, and nobody is all that much complaining that I know of. The applicants are aware of the complexity and at least some examiners are aware of the complexity and everyone just makes do best they can.


          LOL – you beat me to it.

          Play the game (subjectively) instead of follow the PUBLISHED objective rules (applied universally) runs to – at least the appearance – of impropriety and BAD things.

          This applies to government programs as well as blogs, just in case anyone was wondering.


          I don’t recall anybody suggesting that SAWS affected more that about 0.5% of applications.

          News flash: a tiny, tiny, tiny number of screeching, paranoid, patent-worshipping ignorami isn’t “evidence.”

          Nice try, though.

          Maybe next time consider the fact that some of those screeching ignorami have admitted or are widely known to engage in habitual appealing of every possible issue right up to the point where the Federal Circuit smacks them down with a Rule 36 affirmance of the PTAB. See, e.g., MediaQueue etc.

      2. 2.1.3

        Max, I think Les was referring to the fact that the PTO gives EXAMINERS way too much rope for picking easier applications to work on, out of order, and in particular out of their true original filing date order.
        The extreme cases of that are messy and really old serial application prosecution files with lots of claims, like Jerome Lemelson’s or Gilbert Hyatt’s, which the PTO had allowed to be ignored for dozens of total years of prosecution delays.
        Of course there are also some applicants that use plural re-filings to delay issuance, one reason being to be able to amend claims covering subsequent commercial products of others. But that delaying tactic would not work if the PTO required them to be acted on in their true claimed filing date order.

        Also, there is reportedly a big prosecution backlog difference between different examining groups, business methods in particular.


          I see that Paul. But doesn’t it depend on what you mean by ‘easier ‘? My point is that it is the Applicant who fixes how ‘easy ‘ the case is to examine.


            I don’t think so Max. I think it is the subject matter of the invention that determines how easy the case is to examine.

            If you mean to suggest that some applications are written in a manner that is difficult to fathom, I think you have to attribute a great deal of blame to the courts for penalizing applicants for every modicum of specificity or clarity. Refer to “the invention” in the specification and be limited to what ever embodiment was being addressed. Refer to an aspect as “a step” and have your claim limited to the precises order of operations recited in the claim.


              That would suggest to me Les that in any specific art unit it will be foreign originating cases that are ‘easier’ and will therefore be examined first. Could that be so?

    2. 2.2

      The graphs are the same, one is just four years behind the other. The first graph shows an increase in pendancy of cases filed 2002-2007. The second graph shows an increase in pendancy of cases filed 2001-2007. Why is this surprising?

      Starting in 2001ish, the number of cases increased and the office did not hire examiners to match. Now the office has overhired Examiners and the time is coming down. The idea is that attrition will eat away at the overhiring as time goes on, and in the meantime the backlog will be cleared out, resulting in a properly-sized examination corp with a decreased backlog.

      1. 2.2.1

        I also should point out that the graphs are misleading because the first is 0-70% and the second is 0-5%. There’s always been a statistically insignificant number of terrible cases. I bet you’d find a similar <5% number of abandonments after 7 years.

  6. 1

    Seeing these graphs in the context of my experience at the EPO, I wonder if there is concerted action in the world’s first division Patent Offices, to give out patents swiftly, to those who “play ball”, to the detriment of those applicants who insist on disputing every point?

    You scrath my back and I’ll scratch yours, say the British. One hand washes the other, as the Germans say.

Comments are closed.