Patent Application Pendency: Percent of Applications Still Pending

The large backlog of cases at the PTO has added greatly to the problem of long patent application pendency. There are other contributors to the problem – such as applicants and examiners unwilling to reach a compromise solution (often based on the demands of superiors).

To provide some measure of pendency, I created a database of 20,000+ randomly selected patent applications and issued patents filed 2000 - 2007. For each of case, I tallied the final status as of mid-July 2009. (Status is either patented, abandoned, or pending). Using that data, I created a chart of the percent of utility applications still pending as of July 2009. This data can be helpful for someone considering whether their still-pending application is an anomaly. Over 84% of the applications filed in 2007 are still pending (+/– 2% at 95% CI) , while less than 1% of applications filed in 2000 are still pending (+/– 1% at 95% CI).

ScreenShot037

The second chart standardizes the same data according to the amount of time in prosecution.

ScreenShot034

Notes:

  • This data only includes applications that were either (1) published or (2) patented. I do not know the percentage of cases unpublished cases that are still-pending versus those that are abandoned.
  • The data presented here does not distinguish between technology centers, continuations, etc. The still-pending rates varies considerably within those different categories.
  • The pendency rate for applications filed Q4 2004 appears unexpectedly high and the rate for Q1 2005 appears unexpectedly low.. That is at least partially explained by a data collection artifact. My data sample includes a relatively large number of applications filed in late December 2004 and early January 2005 rather than being more evenly distributed throughout the respective three months.

38 thoughts on “Patent Application Pendency: Percent of Applications Still Pending

  1. 99% of the time I receive an OA with a 112, 2nd rejection, it asserts that the Exr does not understand a term that anyone that has taken Bio 101 understands. Every time, I lose a little more sympathy for the PTO’s backlog.

    And yes. It happens every single time, and my sympathy is gone. Whether or not e-commerce patents are legitimate, and whether or not Exr’s are overwhelmed, there is no excuse for that crap.

    The “brilliance” that they show in analyzing the prior art is never reflected in their analysis of your claims or spec. Enough said. The PTO’s problems are self-inflicted, so let them fix it. I’ll just keep counting my PTA (since they don’t do that, either).

  2. 6 – seriously, please stop discussing actual cases…

    Everything dealing with prosecution is supposed to be between the Office and the applicant, not some random patent guys on the internet (even if you completely ignore them).

    Since another blogger inside the Office has already (allegedly) determined your identity, if that got out it would be trivial to determine which case you were discussing.

    For what it’s worth, you are a passionate contributor, just please be more careful.

    EMG

  3. “More frustrating is that its effective filing date is 2000 so the patent term is almost half over.”

    They “hypo” example above happens to have priority back to 01.

  4. I am prosecuting a continuation/divisional application (not an e-commerce application). Filed almost three years ago, and still no first OA, despite the examiner’s many promises and missed deadlines. More frustrating is that its effective filing date is 2000 so the patent term is almost half over.

  5. In response to one comment above, here are a few reasons why ignoring the original [claimed] filling date of a continuation or divisional [treating them as if they were “new” applications] re “pendency” is not a good idea.
    First, their patent term runs from their original filing date, not their re-filling date, so if the PTO does not treating them in their true priority date order it can reduce their patent term. [Except for applications still pending [inexcusably] since 9/9/95, which, even worse, will get a 17 year patent term on top of their entire gross pendency].
    Secondly, a continuation or divisional is not a new application just because some or all of the original claims are amended. [If merely changing claims made for a new application, every substantial voluntary amendment would also generate a new pendency statistical run!] A continuation or divisional cannot validly add any “new matter”, and all of its claims must have full 112 support in the original specification, including enablement and “a written description of THE invention” unless it is a CIP. [But a CIP loses the original filing date benefit for any claims not fully 112 supported in the original application.] A continuation or divisional is, or becomes, presumptive prior art and prior invention against all later-filed applications and patents as of its original filing date, not its re-filing date. Thus it has the same effect against the public as an originally filed application date, except that it may contain surprise new or extensively amended claims added much later on during the longer total pendency obtained by doing so in those re-filings, rather than originally, which is why an extended true pendency for a continuation or divisional can be much more publicly dangerous.
    Thirdly, the almost unlimited ability to “chain” serial, sequential, continuations and/or divisionals for 25 years or more years of pendency, as in several famous and nationally costly Lemelson and Hyatt patents, for example, effectively demonstrates the illogic of looking at only the most recent such re-filing date, with the latest set of clams, in evaluating application pendency statistics and proper PTO processing priorities.

  6. 6: so sorry, but I don’t understand, so I can’t answer your question.

    Never mind. I have anyway the feeling it is others you want to hear from, rather than me. Good luck with that.

  7. “A(br)B(br)C(sp) and A(br)B(sp)C(br)”

    FINALLY! The attorney’s (in general) point of view comes out.

    The question at hand however, can be phrased differently, as I put forth above. ABb’C, and ABCc’. Or, in the language of restrictions, b’ is “A”, C is “Bbr” and Cc’ is “Bsp”.

    The answer thus, appears to me, that this restriction would be proper. Thanks however for showing me why attorneys have such a hard time seeing this, and for showing off attorneys natural disposition to not like combination/subcomination rejections. It will make it easier for me to write the restriction up.

    In a point of fact, there are some additional considerations in the actual “hypo” app. The applicant moved the limitation b’ down into a new dependent claim. Thus we are left with the classical situation of ABsp (new claim 3 not previously mentioned), ABbr (old claim 1), and Bsp (new claim 1 as mentioned). Wherein, as the MPEP informs us, it is proper to restrict.

    Additionally, and this is where it gets strange, applicant filed a continuation which will die under ODP (the claims are basically ABbrCbr, AevenbroaderBbrCsp, and Ahugelybroad,Bbr,Cbr, by the guy above’s way of determining ABC) if the claims in this app don’t die or get a restriction.

    Oh and hey smart arse, I can recite

    link to uspto.gov

    by heart and have it faved in my Insight.

    Max, I’ll be seein won’t I?

    Looks like crystalcity agrees with me.

  8. Hi Dennis,

    Quick question regarding these numbers – I know you don’t have a breakdown by TC, but do you know if any of these cases are design applications, and if so, any idea what percentage design apps represent here?

    I am putting together a post getting to some of the reasons behind this data, but the design app info could be skewing some of the later filed applications.

    Thanks,
    Erin-Michael

    http://www.gillip.com

  9. Freehly: I’m intrigued. You mean the Application as filed disclosed a contribution to the art but then claimed less than all of it? Is it normal, under US law, that you as PTO Exr have to draft the claims that Applicant should have drafted if he had been awake, and then search them instead of the ones filed by the dozy Applicant? Fascinating.

    Oh and BTW, we in Europe are not limited to the same claim set either. And there is no blanket prohibition on broadening the claims after filing (eg by filing a divisional). Our requirements are i) support in the originally filed app ii) no new subject matter iii) clarity and iv) claim set must be “concise” and confined to a single inventive concept. Thus, you can have co-existing in a single claims set any which way claim categories (apparatus, method, product, use, etc) you like, but (normally) no more than one independent claim in the same “category”. Example: “means for X-ing, Y-ing and Z-ing co-existing with a claim reciting the method steps of X-ing, Y-ing and Z-ing, all in the same app. Why not? It is all the same inventive concept isn’t it?

    Slow: One and the same. That’s me. But what has PSA to do with this particular thread?

  10. Applicant has cheekily presented you with subject matter that did not form part of his app as filed.

    Posted by: MaxDrei | Jul 17, 2009 at 02:46 AM

    Max, the hypo said the claim was amended. It did not say there was no support for the newly claimed subject matter in the specification. In the US, it very well may be considered previously searched matter (albeit since the prior search was necessarily very broad).

    Are you sure you’re the same MaxDrei that used to post on and on about the advantages of PSA? Your comments lately are much more off the point.

  11. Max,
    6 didn’t say anything about the new claim not being supported by the spec. If it were not, he would give the applicant a 112 rejection.

    The question 6 is asking is whether claims to A(br)B(br)C(sp) and A(br)B(sp)C(br) (br=broad, sp=specific) can be restricted as a combination/sub-combination. They cannot. Perhaps he could restrict them under some other restriction rationale, but not combination/sub-combination. He could possibly do an election of species if no embodiment is disclosed that comprises both B(sp) and C(sp), and searching both presents a search burden.

    Regarding your point (ii), as I’m sure you’re aware, applicants in the US are not limited to the original claim set. As such, USPTO examiners cannot use (ii) as a recourse. If they could, they would be using the hell out of it.

  12. yes, IF there would be a “serious search burden” if restriction were not required as evidenced by separate classification, status, or field of search.
    “As a European, I’m fascinated, appalled, even flabbergasted.” Ein Sturm im Wasserglas? ya?

  13. “Will it really take until 2016 before an ecommerce patent app filed today gets looked at?”

    “God, I hope so.”

    Even that is too soon IMO.

  14. 6, you characterize Applicant’s amendment to the independent method claim:

    “Thus further specifying actions to be taken in step B and striking the recitations of the specificities of step C. ”

    ….and your correspondents tell you to get on and examine it. As a European, I’m fascinated, appalled, even flabbergasted. Applicant has cheekily presented you with subject matter that did not form part of his app as filed. He was too dumb or lazy to write a proper claim on day one and now wants to widen it, during prosecution. Applicant demands that you now accommodate that sloppy behaviour, get on and do a new search and a new examination, without his paying any more fees to the PTO. Didn’t some CAFC or SCOTIUS judge recently admonish PTO Applicants with the caution “Be Careful What You Ask For”?

    At the EPO, Applicant would simply be told either:

    i) New claim contains new matter (ie offends your “written description” requirement). Not allowable. OR

    ii) New claim is directed to unsearched matter. Go file a divisional.

    And all that would take an EPO Examiner about ten minutes, I should think.

    Can’t you do the same?

  15. “Is the examiner justified in performing a combination/subcombination restriction and withdrawing new claim 1 by original presentation?”
    No. No subcombination has been claimed.

    “Bonus: Would the examiner have been justified in performing a combination/subcombination restriction had the newly amended claims been presented along with the previously submitted claims in the application as originally filed?”
    No. Consult the MPEP (for once)

  16. Combination/subcombination restrictions should never be justified IMHO. Just go ahead and examine the claims and don’t spend time and cause delay with restrictions like this.

  17. After such an overwhelingly positive response to a different question in another thread, let’s go for another round.

    Applicant recieves action on the merits for invention Group 1.

    1. A method comprising step A, step B, and step specific C.

    Applicant amends his claims in a response to read:

    1. A method comprising step A, more specific step B, and step C.

    Thus further specifying actions to be taken in step B and striking the recitations of the specificities of step C.

    (That is, original claim 1. ABb’C, new claim 1. ABCc’ where the lower case letters represent narrowing limitations)

    It just so happens that there are subclasses for B, C, and A, as well as an overall class (which is itself technically a subclass) which each of those subclasses falls under. Previously, the examiner had focused his search in the subclass for A (because it would be a very broad subclass and likely to get lots of hits) and the overall class(subclass) which encompasses all three subclasses. All of the subclasses are large.

    The specificities of C (which are now striken) were previously addressed in the office action under caselaw as being an obvious range of values.

    The new specificities of B are going to require going to require searching the subclass for B whereas the general step B which was previously incorporated happened to get caught in the more general search previously done.

    The point at which this application is in prosecution is amendment after non-f.

    Is the examiner justified in performing a combination/subcombination restriction and withdrawing new claim 1 by original presentation? Brief explanations of your reasoning are encouraged.

    Bonus: Would the examiner have been justified in performing a combination/subcombination restriction had the newly amended claims been presented along with the previously submitted claims in the application as originally filed?

  18. I mean: what makes you so sure that the problem isn’t being manufactured? Perhaps this problem is being manufactured to solve a bigger problem (e-commerce patents?). Perhaps it is being manufactured by those whom you are wanting to fix the problem.

    I think I know who you are irl curious. Don’t worry, I know the id’s of lots of anon’s here.

    “Honestly, what is really stopping the PTO from contracting out some of the work causing this backlog.”

    The law :(

  19. Paul Morgan said: “Of course any application pendency study is misleading unless pendency is counted from the ORIGINAL filing date for which benefit is claimed for an application, not the filing date of a continuation or divisional. Was that done here?”

    How can it be “misleading” Paul?

    Examiners examine CLAIMS – not a priority disclosure. The search and examination activities that are analyzed commence only after the specific CLAIMS are submitted and if, for example, a Divisional application is filed two years after the parent, the new claims are by definition directed to different aspects of the disclosure. As a PTO performance measure, your suggestion for adding two years to the actual pendency would be meaningless and even more “misleading” because the examiner had nothing to examine in that period. The same argument applies for Continuations that, unlike RCE’s, are not a repeated attempt at the parent’s claims. In contrast, RCE’s are by definition a repeated attempt at earlier-filed claims and therefore their pendency should be (and indeed is) counted from the date of the earlier application. RCE’s do not receive a new application serial number and therefore the charts that Dennis posted correctly reflect pendency based on the proper serialized application date.

  20. “Honestly, what is really stopping the PTO from contracting out some of the work causing this backlog.”

    They’re out of money.

  21. Ron Katznelson and I just went over the latest data from the Official Gazette on the pendency to first office action of ecommerce patent applications.

    It looks like the USPTO has substantially increased the rate of first office actions in the past year and pendency to first office action may have leveled off at 4 years.

    In all fairness then, I would now expect first office actions in 2013, not 2016, for e commerce applications filed today. If they can keep the high rate of first office actions up, then the delay should come down even more.

  22. Honestly, what is really stopping the PTO from contracting out some of the work causing this backlog. Taking a simplistic view here, why can’t for example, they set up a special programme where experienced patent attorneys or other technically qualified individuals (thereby creating jobs) are recruited to work for the PTO (in exchange for cash or credits- which if its an issue of money, can then be redeemed in some other way i.e. free filings of applications by said attorneys. This could probably help clear away part of the backlog, and i’m sure there are many who would sign up to such a programme.(Of course there are a lot of factors that would require careful consideration, such as preventing conflict of interest situations – which any university would be happy to teach u how). Its simply unbelievable whats causing such a backlog. Its certainly not that there aren’t enough competent and technically qualified individuals who are more than able to do the job in the US? Even if it was an issue of who foots the bill, there are plenty of innovative ways around such things?

  23. “Will this kind of situation arise in 10% of filings though?”

    Between cons, divs, and possibly even RCE’s, the answer is probably yes.

  24. “PTO management, please explain why and what you are doing to fix this problem.”

    What makes you so sure this isn’t an artificial problem?

  25. Paul writes: “Of course any application pendency study is misleading unless pendancy is counted from the ORIGINAL filing date for which benefit is claimed for an application, not the filing date of a continuation or divisional. Was that done here?”

    My GUESS is that it was from the filing date of the continuation/divisional and not their effective filing dates? Dennis?

  26. Mark writes: “an ecommerce patent filed today won’t get its first office action until 2016.”

    That is absolutely inexcusable.

    PTO management, please explain why and what you are doing to fix this problem.

  27. Is there a reason why the 4Q 2007 is only 90% pending and not 100% pending? The only thing I can imagine happening is where the original gets restricted but the examined claims are almost immediately allowed, then divisionals are filed for the non-elected claims and they are also almost immediately allowed. Will this kind of situation arise in 10% of filings though?

  28. “Will it really take until 2016 before an ecommerce patent app filed today gets looked at?”

    God, I hope so.

  29. This proves a point I have repeated made to the PTO and elsewhere. The number of [often dangerous] “submarine” [grossly long pendency] applications is small enough that it could be dealt with by better PTO management docket supervision and reassignments to examiners who would get appropriate credit for promptly dealing with them, without a budget increase.
    —–
    Of course any application pendency study is misleading unless pendancy is counted from the ORIGINAL filing date for which benefit is claimed for an application, not the filing date of a continuation or divisional. Was that done here?

  30. Will it really take until 2016 before an ecommerce patent app filed today gets looked at?

    My clients are usually most concerned about when a patent application will get its first office action. Based on Official Gazette data, and assuming trends continue, it looks like an ecommerce patent filed today won’t get its first office action until 2016. Here’s a link to a graph and short article explaining why.

    link to bit.ly

    I welcome counter opinions.

  31. Michael, you can bet that they will increase, and things will get worse. Cost consciousness is increasing among almost all of my clients, especially foreign ones. Increases in PTO costs will simply decrease filings even more.

    After the PTO burns through the backlog, the lower ranks, this time examiners themselves, will work to make it difficult to obtain a patent. This happened while I was there almost 20 years ago. Examiners start scrambling for cases and will try to keep cases on their dockets as long as possible (by not allowing, restricting etc.).

    I continue to recommend appealing as soon as legally possible.

  32. Post-First-Office-Action-on-the-Merits versus pre-FOAM might be insightful: As new filings and the percentage of four-year and eight-year maintenance payments drop off, one might speculate filing fee increases (on the order of PCT US/ISA) just over the horizon to cover the two-year and older FOAM cases within this data.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>