Today I am looking at statistics on continuing applications. This post was motivated by two questions that I wanted to explore:
Question 1: Should I take not-so-great allowable claims to “break the seal”?
We’ve all been there, the examiner has rejected our independent/broader claims (on what we believe to be dubious grounds) but indicated some other, way-narrower, claims as allowable. Should we keep fighting and tear apart her flimsy rejection? Or should we offer an olive branch by taking the allowable claims and returning to the broader claims in a continuation?
I often offer up the olive branch. I do so at least partly because I think a giving the examiner these “easy” counts (one for the allowance and another for a relatively low-effort first office action in the continuation) seems to puts the examiner in a cooperative frame of mind for the continuation. In other words, I believe taking the allowable claims “breaks the seal” and allows better claims to start flowing in continuations.
The question here is this: is my “break the seal” strategy backed by the data, or is it just a figment of my cognitively-biased imagination?
Question 2: Do continuations get examined as fast as they seem to?
A corollary to question 1 is that, in my experience, continuations often seem to get examined very quickly. In fact, unless a continuation is truly urgent, I usually advise against filing it Track 1 because of how fast track 1 applications seem to get examined. Now, I know continuations go on the “special docket,” but here I want to quantify what that means in practice.
Tangent: From a policy perspective, I don’t understand why continuations are put on the special docket. With hundreds of thousands of new applications awaiting examination, why are the “incumbents” getting their second bite at the apple before the newcomers get their first? Is the PTO is trying to discourage applicants from keeping disclosures pending indefinitely? If so, it is ineffective at achieving that goal. As I mentioned in The Record for Most RCEs in a Single Application, deep-pocketed applicants can afford to game the system to keep disclosure pending whereas small entities cannot. To me it would make more sense to give the applicant the option of whether a continuation goes on the regular or special docket.
I looked at utility applications filed since 2001 and classified the applications as follows:
- “continuing application” – an application which is one or more of: a continuation, a continuation-in-part, a divisional, a reissue, a reexamination, a substitute1. Note that this definition says nothing about whether an application claims priority to one or more provisionals, to one or more foreign applications, or is a national stage entry of a PCT.
- “has an allowed parent” – An application with at least one allowed parent.
- “has no allowed parent” – An application with no allowed parents.
- “continuation” – An application which only has a relationship of “continuation of” with its parent(s).2
- “continuation-in-part” – An application which only has a relationship of “continuation-in-part of” with its parent(s).2
- “divisional” – An application which only has a relationship of “divisional of” with its parent(s).2
Continuing Applications Rising
Before I jump into the examination metrics, first a look at the general state of continuing applications in the U.S.
In Continuation Nation, I showed the share of U.S. utility applications which are continuing applications has been rising for a decade.
Tangent: I assume the primary reason for this is that continuations are the closest thing we have to a time machine — rather than having to predict how a product or market will develop in the future, we can simply keep the disclosure pending through continuations and then rewrite history with some cleverly-drafted new claims.
What the chart below shows that the charts in Continuation Nation did not, is a breakdown based on whether the continuing application has an allowed parent. As you can see, most continuing applications have an allowed parent, which makes sense since an RCE is usually recourse for applications struggling to obtain an allowance.
Tangent: I wonder how many of these continuations were filed instead of an RCE in the hopes that they would get a different examiner? (Post on that coming soon…subscribe so you don’t miss it!)
The next figure shows them broken out by priority type.
Since I was lazy and gave you non-interactive charts, here is the breakdown for applications filed in 2016:
- 13% of utility applications filed in 2016 were continuations with an allowed parent (up from 4% in 2001)
- 5% of utility applications filed in 2016 were continuations with no allowed parent (up from 3% in 2001)
- 4% of utility applications filed in 2016 were divisionals with an allowed parent (down from 5% in 2001)
- 1% of utility applications filed in 2016 were divisionals with no allowed parent (no change from 2001)
- 2% of utility applications filed in 2016 were continuations-in-part with an allowed parent (down from 3% in 2001)
- 1% of utility applications filed in 2016 were continuations-in-part with no allowed parent (down from 2% in 2001)
Allowance Rate by Priority Type
Here we see that continuing applications with an allowed parent have an allowance rate that is 9% higher than average. Continuations without an allowed parent have an allowance rate that is 14% below average.
Breaking it down by priority type shows that continuation allowance rate is 22% higher when there is an allowed parent vs. when there is not. This 87% allowance rate is 11% higher than the USPTO overall allowance rate of for utility applications.
So continuation allowance rate is one data point in favor of my theory that allowances are “easier” to get for continuations. Of course, this says nothing about the quality or scope of the claims in the parent and child, but still, it at least directionally supports my perception that taking some crappy claims in a parent application can help you get good claims in a continuation.
To be continued . . .
In the next post I will continue exploring this topic with some more metrics including: responses to allowance and time to first action. Subscribe here to get it.
1 I actually had to look up what a substitute is. The MPEP defines it as: “in essence the duplicate of an application by the same applicant abandoned before the filing of the later application . . . [which] does not obtain the benefit of the filing date of the prior application.”
2 The result of these definitions is that applications with mixed priority chains (e.g., an application which is a continuation of one parent and a divisional of another parent) are excluded. This represented about 4% of utility applications in the dataset.