Statistical Interlude: First 100 Published Applications

Interesting sideline results from some empirical work that I’m doing:

In 2001, the PTO began publishing patent applications 18 months after filing unless the applicant filed a nonpublication request.  To date, more than 1.2 million pending applications have been published. Of the first 100 published applications, 73 applications have issued as patents and 27 were abandoned.  None of the originally filed applications are still pending, and only one of the abandoned application still has a pending continuation. 19 of the abandoned applications have at least one family-member that is an issued patent.  Thus, leaving only 8 that are “fully dead” (i.e., with no known issued patents or pending family members).


8 thoughts on “Statistical Interlude: First 100 Published Applications

  1. “The claim is the name of the game,” as Judge Rich told us so often. When the PTO began publishing applications, the PTO also began publishing claims. Given that the claim is in fact the name of the game, wouldn’t it make more sense to come up with statistics on how many of the first 100 published INDEPENDENT CLAIMS have been issued without being narrowed by amendment?

  2. I wrote a paper that was published by JPTOS in May that is related to this topic. I agree with the comments posted by Dennis Crouch that it is difficult to post a single number that precisely reflects “patenting rate” or “grant rate” or “success rate.” Actually, it is impossible to do, in my judgment, at least if one is intent on being accurate to the precise percentage point.

    As far as I know, Cecil Quillen and Slim Webster (both formerly from Kodak) were the first to engage in a statistical analysis regarding the impact of continuation filings on USPTO-reported “grant rates.” They have written a series of two or three papers on this subject, which precipitated papers from others and, to a certain extent, drew ire that was undeserved, in my opinion.

    I think it is important to not get too caught up in a debate about whose “grant rate” calculation is precisely correct, or “most” correct, etc. Instead, it is important to acknowledge that USPTO-reported “grant rates” do not reflect reality and mislead the public in that reported grant rates suggest that patent applicants are being less successful than reality. Having said that, in my opinion, it is probably not intentional on the USPTO’s part but is instead the result of a bureaucratic dynamic that has evolved within the USPTO over a period of many, many years.

  3. With the publication of the applications, Dennis’ work may provide a clearer statistical picture of just what the hell is going on inside the secret USPTO, at least with respect to the group of those patent applicants who intend to foreign file at the time they file a US patent application and those who just plain didn’t file a nonpublication request. His work also may encourage others to statistically look at the same problem instead of sitting around waiting to be spoon fed the answers and then crying about that the spoon is insufficient.

  4. Would Dennis’ research, if refined to avoid “skewing” and extended in light of Heath’s question and comment, make clear whether the PTO’s proposed new continuation limitations are well founded? Can anyone tell me whether the PTO has any data of the sort that would result from such research? And, if the research showed that they were not well-founded, might our patent reform legislators be willing to consider adding some “remedial” provisions to their bills?

  5. Any info on the amount of family members still alive in the issued cases (i.e., how many of the issued patents have divisionals/continuations that were filed and still are pending)? That would be interesting information in view of the patent office’s assertion that continuation practice is killing patent office resources.

  6. You should realize that these first 100 applications are somewhat skewed. They were all filed in 1999 immediately after the publication rule came into effect. Many people were afraid of publication and thus opted for non-publication (if they could). In addition, there was a large bubble of filings just before the rule change in late 1999.

  7. Kevin, The answer is that both percentages provide us with a bit of information.

    The patent system is complex enough that I don’t think we can convey enough information with a single number to represent a “patenting rate.” Anyone who is interested in the topic (for more that just flame wars) will be able to handle a bit of nuance rather than just one number.

  8. So, Dennis, what would you say is the “patenting rate” – 92%? 73% Something in between? And if there have been multiple patents from these applications (for example, divisionals due to PTO restriction practice), how do you think they should be counted?

Comments are closed.