Evidence Based Prosecution: More Claims Filed Equates with Longer Time in Prosecution

I am looking generally at how the contents of a patent application may impact the prosecution process. In this study, I looked at a sample of 50,000 original patent applications (excluding continuations) that have issued as patents in the past few years. This first graph shows the average time in prosecution (filing to issue) as a function of the total number of claims in the original patent application. The result is that the claim count positively correlates with time in prosecution. This correlation is strongest when the claim count is less than 30. As an example, patent applications with 30 claims took about 33% longer to issue than those with only 2 claims. The correlation makes sense: more claims would generally mean more work for the PTO to reject each claim, and more work for the applicant to respond in kind.

The second graph shows a similar pattern for independent claims.

24 thoughts on “Evidence Based Prosecution: More Claims Filed Equates with Longer Time in Prosecution

  1. 24

    I agree that the results need to be correlated to art unit and also distinguish time to first office action from time from first action to issuance.

    There is a wide range of times even to first office action among the various art units, and it may be that some technical fields will tend to have more claims, skewing the results.

    The time to first office action depends upon factors outside of a patent drafter’s control (number of applications filed in that field, number of claims of those other applications, etc.), whereas time from first office action to issuance includes factors (e.g., difficulty of examination and response) that may relate more closely to the number of claims presented. If I present a case with just one claim in a field that tends to have many claims, I still have to wait just as long to get an office action. There is a group effect of the overall average number of claims of all applications in a particular technology that effects pendency mainly before first action, and an individual effect of the particular number of claims in a given application that affects only that portion of the pendency time after first action. These effects should be separated. Patent drafters want to know how the number of claims in their particular application will affect time to issuance.

  2. 23

    “Also, are there no data for prosecutions of less than 2 years”

    This is because y axis shows average prosecutionn times.

  3. 22

    This needs to be analyszed by art unit.

    Also, is this time to first issuance or last dividonal issued (if it is the latter, this means little as more divisionals take more time).

  4. 21

    should you use these graphs as a basis for prosecution and combine it with an unwillingness to narrow the claims, you have a recipe for disaster. A better understanding of an application’s pendency can be acquired by:

    1) comparing # claims and # independent claims within an art unit (as suggested by others above)
    2) normalize the results from 1) to the backlog for an art unit
    3) eliminate anomalies (e.g. accelerated apps, special apps, etc.,)

    Also, generally the opposite is true; with less claims it is easier to find prior art and maintain a rejection (there maybe 50k 2 claim app’n alone which did not issue).

    What is key is the number of elements in an independent claim and the degree of specificity in claim scope, and not the number of claims. With 30 claims, it is quite possible the applicant did not make independent claims specific enough and extended critical elements into dependent claims or other independent claims (meaning more work for the examiner less work for the applicant).

    There’s another problem with the correlation which suggests a contradiction, some applications may have 29 dependent claims and one independent claim, some may have 5-10 claims all in independent form. It is not clear whether an applicant filing fewer claims (or more claims) is more or less willing to narrow the scope (applicant may restrict, elect and later rejoin claims which places some serious doubts on the conclusion contrued).

    AK

  5. 19

    Dennis, 2 suggestions:

    1. standard deviations
    2. correlation coefficients

    BEFORE conclusions.

    Also, are there no data for prosecutions of less than 2 years? No, seriously, I’m not joking. I have gotten Quayled in under 18 months a few times. OK, a very few times. But out of 50,000 cases, there should be a small but visible population of sub-2 year prosecutions.

  6. 18

    Dennis, 2 suggestions:

    1. standard deviations
    2. correlation coefficients

    BEFORE conclusions.

    Also, are there no data for prosecutions of less than 2 years? No, seriously, I’m not joking. I have gotten Quayled in under 18 months a few times. OK, a very few times. But out of 50,000 cases, there should be a small but visible population of sub-2 year prosecutions.

  7. 17

    MaxDrei

    My guess is that it was too confusing for the PTO to reject some but not all of the multiple dep claims which depended from other multiple dep claims and also too complicated to figure out the extra claims fee.

  8. 16

    ROW = rest of the world.

    Multiple on multiple = bedrock important, everywhere in the world, except the USA (something you will learn, 6K, when you one day instruct foreign associates). It is not just an EPO thing. Readers, I suggest you ask your Chinese patent counsel.

  9. 15

    Thanks pds. I thought so. So, no problem with clarity then. Can any reader please tell me, what is the reason for the requirement in the CFR, unique to the USA, for no dependency of a multiple-dependent claim on a multiple-dependent claim. It is this which necessitates recitation of a single technical feature “multiple times”, and makes US claim sets multiple times longer than anywhere else in the world. Reciting a word string “multiple times” is otiose. Why, alone in all the world, does the US government insist upon it? Is this an isolated occurrence, of a US Government requirement for redundancy, or are there other such examples?

  10. 14

    “Tell me 6K, would ROW-style multiple dependency make your job harder, or easier?”

    I’ve been trying to figure out what ROW means since you’ve been using it but I haven’t bothered to look it up. I assume you mean multiple on multiple for the present conversation, and frankly they seem rather pointless to me, at least in my art. I’ve only ever had maybe 3 claims that were even multiples in all my thousand? claims examined. Harder/easier I couldn’t comment on but I doubt it would change much. If you gave US prosecutors multiple on multiple and they were inclined to use it they’d simply make 30 multiple on multiples. Claims are made for the sake of making claims, not claiming what the inventor believes to be the invention. Didn’t you hear Max?

  11. 13

    What I think the numbers show is that after 6/30, the likelihood that any additional claim will recite something different than that already claim is very small.

    IMHO, if you writing more than 6/30 claims, the additional claims are likely repeats of old claims but perhaps directed to different types of statutory classes (e.g., method and device) or the inventive device in combination with something else (e.g., “a new windshield wiper, comprising …” and “a car comprising a windshield wiper …”).

  12. 12

    “would ROW-style multiple dependency make your job harder, or easier?”

    You don’t need 6K to answer that question. The answer is neither harder nor easier.

    Once a limitation has already been presented in the claims (either once with a multiple-dependent claim or multiple-times in US-style claims), an examiner need only do the analysis just one time. It takes a nominal amount of effort to “cut and paste” the analysis from one claim to another.

  13. 11

    “This correlation is strongest when the claim count is less than 30”.

    This may be correct, but I have doubts that that is what you really meant as to state that would require some work. First you would have to calculate two different correlation values (either side of 30), and then make an additional calculation to show that the difference in those values is statistically significant. If you did this, great! Could you post the numbers?

    Or did you mean “the effect is greatest when the claim count is less than 30”?

  14. 10

    When I was an Examiner we were told to examine cases based on how old they were. However, on occasion, I was known to examine a case with fewer claims and a smaller specification before a large case because it was easier and I got the same amount of credit (two counts per balenced disposal i.e. one count for first action and one when the case was finished). Perhaps some of the effect you see is simply Examiners doing easy cases first in which case there is a causal effect.
    I also agree with the post above that a large number of claims often are submitted in cases when the applicant thinks the invention is important and, furthermore, I think it is an indicator that the applicant has substantial resources. When an applicant thinks the invention important, the applicant is less likely to compromise on claim language and if the applicant has money, the applicant can afford to appeal a case. Therefore a large number of claims could simply be an indicator of an important case of a rich applicant and thus a case which will take a long time to prosecute.
    Numerous other thoughts come to mind such as the amount of claims may also be dictated by the type of art which, in turn, is one with a large pendancy. E.G. Business method applications tend to have many claims and are also in an area with long prosecution times.

  15. 9

    Steven S.

    Exactly my thoughts. Number of claims goes up over time, years in prosecution goes up over time, etc.

    Many correlations here, and I’m not sure any are due to a causal relationship.

  16. 8

    No other country but the USA forbids multiple dependent claims from dependence on such a claim. Thus, every other country can claim in 10 claims what the USA needs 30 claims, or even more (can’t be faffed now, to work out the geometric progression). No other country tries patent infringement by jury. Are these facts somehow connected, or is it really that multiple on multiple is too difficult for the PTO examining cadre to handle. Tell me 6K, would ROW-style multiple dependency make your job harder, or easier?

  17. 7

    Obviously MM, once your case is designated as a “huge one” it doesn’t matter if you have 50 or 300 claims you’re still going as far back as you can be pushed.

  18. 6

    One take home: if you have the cash and you’re already filing 30 claims, might as well file a dozen more. It won’t slow things down much.

  19. 5

    The graphs are meaningless without tying them to a particular art unit. Event simple mechanical inventions with 20 or less claims is taking 3 or more years to be examined. Some software applications are out to 60 months with no more than 20 claims.

  20. 4

    Impressive research. How are you getting the raw data? Other possible analyses:
    – correlating number of claims to technology
    – number of claims to size of the applicant/assignee
    – trends over time, based upon filing date or issue date
    – impact of changes in claim count, filed vs granted
    – spikes or dips, correlated to cases and economic conditions

  21. 3

    Ummm – – duhhhhh?!?! – – We get it already Dennis, you’re very proficient with Excel. Now enough already with these lame-a$$ figures!!

  22. 2

    That’s nice to see what we all knew in black and white. I personally would have thought the trend much more pronounced. What was the sample size and what AU’s were used?

  23. 1

    I am not surprised at all by your findings. It also may be the applications with more claims are considered to be more valuable to the Applicant and thus the Applicant is unwilling to settle for narrow (easy to design around) claims. Also, I have found that examiners often do not give each of 30 claims the same examination care as each of 5 claims, for example, thus requiring additional Office Actions to allowance. (Additionally, don’t forget that when claims exceed 3/20 there are extra claims fees.)

Comments are closed.