Update April 2019: You can now lookup your examiner’s AFCP stats in BigPatentData Examiner Analytics.
The After Final Consideration Pilot Program (AFCP) 2.0 has again been extended for another year. It would be nice if the extension came with some useful data about how the program is doing and why it was extended, but that does not seem to be the case. If anyone is aware of any stats or analysis that was published in support of extending the program I would to grateful for the link. All I have found on the USPTO’s website are these somewhat inconclusive stats from way back in February:
Although its not clear, the second column in the graph above appears to include all purported AFCP 2.0 responses (i.e., AFCP 2.0 requests where: (1) the amendment was considered, (2) the amendment was not considered because it was “too extensive;” and (3) the request was deemed improper). That data shows AFCP 2.0 responses overall actually did significantly worse than traditional after final responses (26% allowances vs. 36% allowances). But, if we look at just the cases where the claim amendments were actually considered – i.e., the last column of the graph – we see a very strong allowance rate of 38% (i.e., 38% of the time a notice of allowance followed the after-final response).
Keeping in mind that AFCP 2.0 requests must include a narrowing claim amendment, it would be ideal if I could compare these stats to traditional after-final responses having a narrowing claim amendment. Unfortunately I have no such data set. But I don’t think I am going too far out on a limb by guessing that the allowance rate for traditional after-final responses having a claim amendment is way lower than 38% or even 26%. If this was the Showcase Showdown my guess would be 0% because even the most trivial amendments after final always seem “require further search or consideration.”
So, considering that we are talking about a 38% allowance rate for after-final amendments, the AFCP 2.0 may in fact be saving us many unnecessary RCEs caused by the dreaded “the amendments require further search or consideration.” But its hard to say that with confidence because the overall stats have not changed much since the AFCP 2.0 went into effect in 2013 (AFCP 2.0 requests were submitted with almost half of all after-final responses in 2017):
If we nevertheless assume that the AFCP 2.0 is saving us from unnecessary RCEs when the amendments are not “too extensive,” that begs the question: “When is an amendment “too extensive?” Well, I don’t have an answer for that per se, I can give you some stats on how often it is the case:
Since 2015, ~50% of AFCP 2.0 amendments have been considered, ~30% were “too extensive”, ~2% of requests were improper, and ~18% are not available / unknown.