by Dennis Crouch

I created a database of utility patents running up through April 8, 2012 and tallied the number of patents that list more than twenty claims. More claims tend to create a greater likelihood that at least one of the claims will be found both valid and infringed. On the other hand, more claims raise the cost of prosecution, increase the odds of the patent being held unenforceable, and correlates with greater prosecution time.

The underlying data is available here: Feel free to use the data with acknowledgement.

26 thoughts on “

  1. 26

    Patent applications are available to the public. If it is important to keep the details of your invention a secret, it may be better to not apply for a patent.

  2. 25

    Thank you. We once again find it necessary to ask him to subtract out the agenda from the reported data.

    I was able to find breakdowns by GS level for a few years (real world numbers apparently for 2007-2009) here:
    link to

    Within a given GS level, the hours per production unit didn’t change much, but the numbers of examiners at higher GS levels has increased significantly in the last few years.

  3. 24

    “My analysis based on USPTO Annual Report data shows that the average annual examiner production rate increased from 65.1 disposals in FY 2008 to 85.8 disposals in FY 2010. In part, this remarkable achievement is due to improved efficiencies and examiner incentives and due to increased use of examiner overtime pay.”

    You’re not considering the hiring freeze that occurred in early 2009 through 2010. Without new hiring during that period it’s only natural that the disposals per examiner increased as people progress up the GS scale and the dockets of newer examiners matured.

  4. 23

    “How long does it take, for goodness sake, to examine a dependent claim?”

    That depends on what the dep claim is max.

    I don’t really know why we see either of these trends.

  5. 22

    Combine the points made by RK and LaDL and you get a far less sensationalistic view more in touch with reality (and with far more insight) than a simple “20+ claim application rate is crashing!!!!”

    Point of the matter is that there is no such crash.

    “Percentage” has some rational drivers. The actual number of 20+ claims appears to have plateaued. The bulk of the increase (hence lower percentage) is more smaller claim set applicaitons being processed. As Ron notes, this is not a surprise.

    I think that I am more concered with the sold stepdown in claims filed, which to me indicates a form over substance result of Office pressure to change applicant behavior.

    Number of claims is an arbitrary measure.

    “Quality” patents contain enough claims to appropriately capture the invention. Forcing that number to be a pre-determined number – any pre-determined number – is just more of the one-size-fits-all widget mentality that is a root cause of examination disfunction.

    Focus on the merits, let the arbitrary things fall as they may.

  6. 21

    Interesting post from Mr Katznelson. I certainly have the feeling in EPO practice that Examiners are much much quicker to take up for continued examination a set of amended claims that self-evidently can be disposed of quickly. Put simply, the larger the number of independent claims in the amended set, the longer you will wait for a second Office Action.

    Even in the USPTO, the work of examining a set which comprises 20 independent claims (or even four) is (I imagine) much higher than the work of examining a single independent claim and 19 claims dependent on it. How long does it take, for goodness sake, to examine a dependent claim?

    In which case, the fewer the number of independent claims, the lower the juicy Katznelson fruit hangs.

    If so, then if you want to get to issue, perhaps it is more important to minimise the number of independent claims than to reduce the total number of claims to 20 or less.

    Or not? 6, are you there today?

  7. 20

    This is interesting information.
    You are showing the number of claims in issued patents. The number of claims issued in a patent should not be confused with the number of claims filed in patent applications. The decline in the former does not necessarily mean that it is due to any decline in the latter. Indeed, although the average number of claims in applications has taken a unit step down after the December 2004 claim fee increase, the average number of claims filed in applications does not appear to have declined appreciably since. This is shown in your own data published on this blog on Oct 01, 2008 (link to

    If the number of claims in applications does not decline, why would patents issue with a declining number of claims? There are two factors that can explain your observations. First, as the figures in my article at link to show, the trend you show of a substantial decline in the number of issued claims is temporally coincident with the trend of substantial decline in the USPTO allowance rate (Figure 1) and the resulting steep increase in Appeal Brief filings (Figure 3). Whenever there is an increase in the number of rejected claims in applications that are not allowed, you can be sure to find an increase in the number of rejected claims in applications that are allowed (issued patents). Therefore, due to the “reject, reject, reject” mentality of the prior administration, we should expect issued patents to have a declining number of claims.

    A second but perhaps less significant factor is the recent downward biasing effect on claim-count due to the productivity campaign at the USPTO: My analysis based on USPTO Annual Report data shows that the average annual examiner production rate increased from 65.1 disposals in FY 2008 to 85.8 disposals in FY 2010. In part, this remarkable achievement is due to improved efficiencies and examiner incentives and due to increased use of examiner overtime pay. But that is not the only way by which this remarkable productivity was achieved. Under such new incentives, it is only natural for examiners to tackle the backlog by first disposing of the lower-hanging fruit – simpler applications with fewer claims. Indeed your own data at link to show that on average, patents with fewer claims enjoy earlier disposal. Productivity campaigns which, as in here, increase production by such amount are bound to be accompanied by examiner preference for disposing of more patents with fewer claims.

  8. 19

    And ain’t it cool how you pull them in by calling it “data” (rather than that dirty word, “statistics”). Even better how no one notices that the real underlying data isn’t even provided. Heh, heh, heh.

  9. 18

    Well thanks for that Paul. So as I see it, the M&B case really was on my 02:57pm point then. Most interesting.

  10. 17

    On second thoughts this was what was said in Fundamentals about that case:

    The specification did not have specific claims directed to the two exemplified active compounds. It only had generic claims to benzene sulfonamido thiazole derivatives, the genus of compounds claimed being described in words rather than by a structural formula. The description explained that the NH2 group could be substituted amino (e.g. alkylamino, N-acyl-N-alkylamino or aralkylamno) and that the benzene ring could have alkyl substituents. The thiazole ring could also be substituted. The resulting genus covered tens of millions of compounds, and although it covered compounds additional to the two compounds exemplified that had been found to be active, it admittedly also covered compounds that were inactive. The patentees therefore proposed to replace their generic claim by a claim to the two exemplified compounds, arguing that it was apparent on the face of the specification that they were drawing an inference from their specific data and should not be penalised if that inference proved unfounded. In a split (3:2) judgment the House of Lords rejected this approach and said :

    “The argument and the construction of the specification implied in it are open to grave objections. They set aside the plain categorical statements of the specification that the invention consists in the manufacture of new benzene-sulphonamido-thiazole derivatives, and that the new benzene-sulphonamido-thiazoles have especially favourable therapeutic properties. These general statements are no longer to be accepted as statements of fact, and they are to be regarded rather as a scientific hypothesis. It seems to have been forgotten that they were the basis of a patent grant just because they were treated as statements of fact, and that they were put into the specification and followed by the commensurate claims for the purpose of obtaining a monopoly in the large field covered by the claims … Let it be here recalled that the Appellants have acquiesced in the finding of Jenkins J. that on a proper construction of the specification there is an assertion that every substance capable of being produced by the described methods from the prescribed materials was of the especially therapeutic value, and that they have obtained from the same learned judge a finding that they believed what they asserted, that is they committed themselves to this statement as a true generalization from their experimental data. Having so committed themselves they cannot now draw back and say that they are committed only to the truth of their assertions about two specific substances and to an unproved scientific hypothesis.”

    The generic claim was therefore invalid, and in the absence of specific claims to the two exemplified compounds the patent was held to be unenforceable.

  11. 15

    Paul I was thinking more along the following lines: that on the filing date you represented in your appln that one member of the group is as good as any other. But then, later, when it turns out that that one member is the only runner and at least one other group member is old, you change your tune and represent that your invention is the selection of the runner from the group ie you now say that it is NOT the case that one group member is as good as any other. The way I see it, that’s making the invention (and “improving your position”) after your filing date and raises questions whether you are non-obvious on your filing date and whether you have added matter after your filing date.

    May & Baker I have forgotten. Perhaps that case gives UK answers to my doubts, but the jurisprudence on chemical selections varies from jurisdiction to jurisdiction.

  12. 14

    Yes, it also looks to me like the big fee increases for filing or adding large numbers of claims stasting several years ago is showing increasing effects on finally-issued claim numbers.
    But almost unlimited numbers of claims are still available in the U.S. for anyone thinking that their invention is worth paying for all of them. Getting them all allowed will make new reexaminations really expensive for requestors with the greatly increasing fees. But as fish scales also notes, filing lots of clams may well trigger restriction requirements forcing divisional applications. That will cause plural patents with plural issue and maintenance fees.

  13. 13

    Wonder how much is do to more election/restriction requirements?

    (Probably much is do to clients/attorneys being more conservative due to economy.)

  14. 12

    Thanks Paul.


    That was the link that gave the US jurisdiction to go after a Hong Kong-based file transfer business. They had US servers.

  15. 11

    If there is a European patent (DE) it would probably be kind of handy if signals were transmitted from a server in Germany or purchased e.g. over the Internet by users in Germany. If you want to buy an application program nowadays, you do not go to a store, you buy it over the Internet. So being able to claim the program in the form of an electrical signal is useful because that is what is traded nowadays.

    And it is good for US patent specifications to contain language foreshadowing potential signal claims. If there is not language of such generality in the US application as filed, there is no basis for priority of a signal claim in Europe, and it will probably be impossible to add a claim of this type.

  16. 10

    Paul, regarding the signal claim, assume a signal patent in Germany, a transmitter in Russia, and a receiver located on a submarine in the Atlantic: is there infringement in Germany? What if there is proof that the signal is relayed through satellite and satellite every now and then passes over Germany.

    Just when does someone infringe a signal patent?

  17. 9

    Not so hypothetical actually. The actual case went all the way to the UK Supreme Court.

    But tell me more. What was it that prompted your remark. What do you know about the case that I don’t yet know?

  18. 8

    It’s great the impression one can do with graphs.

    Charting the pure number of 20+ claims paints a very different picture. Rather than a cataclysmic drop, the rate of 20+ claims in actual numbers has more approached a steady state.

    One further driver omitted from discussion is the relatively recent propensity to request restriction of larger claim sets. Such a driver has a double effect as a single count of 20+ may produce two counts of under 20.

  19. 7

    you have no idea that taxol really does work, and further, is the only group member that works.


  20. 6

    If you have an individual claim to compound 31, you are in a more difficult position as regards subsequent selection patents. But that is a price that you have to pay to get an early date.

    The question of creating a new invention by striking-out is an interesting one. The decision of the House of Lords in May & Baker’s patent, reported if my memory is correct in 1950 R.P.C. was on point, although this decision is somewhat forgotten nowadays.

  21. 5

    You have a claim to a genus, and dependent claims of one sort or another to the 60 individual compounds you have made within the genus. You hope that the genus is novel and inventive. Thousands of patents in that general format issue every week.

    But suppose post-grant the novelty of the genus goes through hypothetical compound 32 being known. Post-grant lack of unity is not an objection since it is a matter of pre-grant administration and fee collection. So you still have your surviving 59 compound claims. It has never been thought right that a group of inventors who make 60 chemically related compounds should have to apply for 60 patents.

  22. 4

    Good arguments Paul but I’m not yet totally convinced, at least when it comes to selection inventions, and your blockbuster situation.

    Suppose your subject matter is a stent, coated with something that will inhibit restenosis. For the coating you teach an anti-angiogenic agent, and you recite and claim all (say) 60 known members of the group. At the time of filing, you have no idea that taxol really does work, and further, is the only group member that works.

    But then, some time later, still nothing else works, taxol turns out indeed to be the blockbuster. Further, there is an earlier filing that teaches to coat a stent with an anti-angiogenic substance, indeed to inhibit restenosis.

    New situation now. Let’s balance fair protection for the inventor with reasonable certainty for the public.

    To the EPO, for a central limitation procedure. Belatedly, claim taxol alone.

    What if a claim to taxol alone had been in there from the outset? Is it a more valid claim than the one you make by central limitation? In an ideal world, should the hundred claim filer be rewarded with an injunction while the central claim cropper gets a revocation? If so, why?

    I think it is crazy enough that specifications in chemical cases run to hundreds of pages of diagrams of the molecule of every individual member of the claimed group. For there then to be, on top of that, a claim to every individual group member strikes me as horrific. But then, I’m not a chemist.

  23. 3

    Paul, why not 60 different patents? One can get only on RR per patent no matter how many different claims are infringed. If there are multiple inventions disclosed, it would be better to put the claims in separate patents.

  24. 1

    Although seemingly beneficial to the public the limitation on the number of claims which has been imposed both in the UK and in the EPO in recent years is an extremely retrograde step.

    Consider a pharmaceutical company which has generted a genus of compounds and made some 60 individual compounds whose prepartion and properties are described. Prior to the limitation on claim numbers the advice of any patent attorney would have been to make each compound the subject of an individual claim. Now that we have pressure on claim numbers, the practice is to include them as alternatives in a single claim. But such a single claim is not legally the same as a claim to each of the compounds individually. Suppose compound 31 turns out to be a blockbuster new drug and compound 32 happens to have been anticipated iin an obscure PhD thesis from the 1930’s. A stand-alone claim to the compound will be novel and a good vehicle for enforcement, whereas a claim to the group of 60 is compromised because one of its members is not novel and therefore the claim as a whole is not novel. The examiner work is the same because all the compounds are members of the genus and the Examiner needs to investigate the novelty of all of them whether in individual claims or in a group. But by exerting financial pressure on applicants to claim the compounds as a group, unfair pressure is placed to present the claims in a legally less advantageous form.

    Similarly in electronics in the case of a transmitter/receiver system. It is normal to include claims to the transmitter, the receiver, media recording the signal and in Europe the signal per se [YES, signal claims ARE allowable ni Europe]. That cannot be done adequately in 15 or 20 claims in a system of any complexity.

    It is suggested that reconsideration would be in order, on both sides of the Pond.

Comments are closed.