The Timing of Patent Grants

by Dennis Crouch

To create the chart below, I merged the my database of all utility patents issued since January 2005 with my database of all patent applications published since January 2005.  I then looked to see how many of the published applications have issued as patents and grouped those according to the application filing date. This chart necessarily excludes information related to unpublished applications.  

From the chart, about 10% of the applications filed 18-months ago have already issued as patents, and about 50% of the applications filed four years ago (January 2009) have issued. The chart is intended to provide some sense of how long it might take to obtain patent protection.  With the advent of fee-based accelerated examination, an applicant can largely direct its patent timing.  The question for quickly issued patents is whether that speed came at the sacrifice of adequate scope.   

The second chart shows two different series: One for "original" non-provisional applications and another for applications that claim priority to a prior US non-provisional application (i.e., continuation applications). As is apparent from that chart, the continuation applications tend to issue much more quickly. 

Timing of Patent Grant

PatentlyO197

The seeming artificat in the continuation series (red) stems from a rush of continuation-application filing in the run-up to USPTO's (failed) implementation of continuation rules.  Because the rules were never implemented, many applicants abandoned those hastily filed continuations. 

 

 

20 thoughts on “The Timing of Patent Grants

  1. Except that it is not – not only for the reasons already stated, but also because this snapshot says nothing about the future.

    Lets try to think about stats and what they actually mean, ok?

  2. The picture is probably not very useful to people who filed 3.5 years ago, but it may still be useful to people filing today.

    For your question, I’m sure you already know to tell your clients what you think their chances are based on the office actions you’ve already received.

  3. If you filed 3 1/2 years ago, what do you tell your client?

    You filed 3 1/2 years ago and you’re looking for a chart to help you figure out what to tell your client? Now that’s ridiculous.

  4. “When will I know whether I’m getting an issued patent or not?”

    “Most cases are either allowed or abandoned within 4 years. Your choices can affect that, but that’s the average.”

    That’s one way to use this information.

  5. Actually, now that I think about it, it’s the difference between timing and volume. The continuations have a very different track as shown by Dennis’s new graph, but there’s not enough of them to move the entire chart more than 5%.

  6. Thanks!

    Interesting. So the inflection points are the same (backlog queue), but the issued percentage gets knocked down by about 5% at each point in time.

    Not as drastic as I had expected, but the difference between continuations and originals is pretty drastic.

  7. I should have said “may” in the first post. That’s the point. without the data, there’s I can’t know.

  8. bja,

    Your first post declares “also greatly skew” and your response to me backtracks mightily.

    I suspect your response below will also do the same.

  9. bja,

    Feel free to answer the questions in my hypothetical instead of merely saying “that’s ridiculous.”

    Your answers will show whether or not “that’s ridiculous.”

    Nothing wrong wth stats – but knowing what they say and what they don’t say is rather important too.

  10. “MAY”

    Yeah, I doubt that the continuations affect the chart THAT much, but it’d be nice to know if they have a perceivable effect or not.

  11. That’s ridiculous.

    To those of us that work with individual inventors and startups, this is GREAT information. All of the information that Dennis posts is directly related to the questions that these guys ask:
    how long?
    how much?
    What costs how much when?
    What are my chances?

    I frequently find myself relaying the statistical information that Dennis develops as rough estimates to my clients.

  12. If you filed 3 1/2 years ago, what do you tell your client?

    That it’s fifty-fifty that he is holding a patent in his hands? That his odds are 1 in 5 that he has abandoned his application? That the odds are 3 in 10 that he may (at some unknown future date) still receive a patent? For that 30%, how much (on average – or otherwise) time remains until either a patent is received or abandonment is reached?

    The pretty picutre is pretty useless.

  13. Not sure you can actually use the data to predict how long it will take to recieve a patent.

    1) each patent issues (or not) on its own merits (these are not widgets).

    2) Success of one application reaching the state of being granted has no relation to any other application being granted(albeit continuations may have an easier go and as bja notes, might heavily skew timing).

    3) how many of the remaining applications published (for example the approximately 38% of January 2005) can yet become patents, and how many no longer stand a chance? (is that a linear relation between the two arrow noted levels) When (if) those that can still become patents actually do, how much will that affect the average?

    4) I would also take issue with the “applicant can largely direct its patent timing” statement. The accelearated options may give you a different track, but again, the notion of each application not being a widget means that “your timing may vary” can quite obscure any estimate that can be derived from the pretty picture.

  14. Continuations also greatly skew these numbers. when you look at the topic again, maybe you could try to control for continuations so that we have a better idea of time to issuance for new non-provisionals.

Comments are closed.