The appeal process within the USPTO is always slow and expensive. The patent attorney must write an appeal brief that takes much more time and effort than a typical office action response. The entire process can easily add two years to prosecution.
In some instances, the examiner will reopen prosecution after the appeal brief has been filed. Reopening prosecution means the examiner issues another office action, and the attorney is back to where he or she was before the appeal was filed despite having spent time and client's money on the appeal. I've even heard of this happening multiple times for a single patent application (final office action, appeal, reopen prosecution, final office action, appeal, reopen prosecution...).
It seems that if an examiner is doing his or her job correctly, there should generally not be grounds for reopening prosecution during appeal. By the time of the final office action, there should be a clear disagreement to be decided by the appeal board.
Accordingly, one would expect that reopening prosecution during appeal should be a relatively rare occurrence. From the Patent Bots data set, it is unfortunately not a rare occurrence.
In the dataset we looked at, about 21% of appeals were reopened by an examiner issuing an office action (we excluded appeals reopened by a pre-appeal conference). To see examiner variability, we created the following scatter plot:
In this scatter plot, each blue dot represents an examiner. The vertical axis is the examiner's grant rate, the horizontal axis is the examiner's reopen rate (the percentage of appeals reopened with an office action), and the size of the dot corresponds to the number of appeals for the examiner. You can click on the plot for an interactive version.
The scatter plot is a bit hard to interpret so I'll supplement with a few statistics:
- 35% of examiners have never reopened prosecution during an appeal
- 26% of examiners have reopened prosecution exactly once during an appeal
- 9% of examiners have a reopen rate of greater than 50% (the right half of the plot above)
- 5 examiners account for 1% of all reopens
Accordingly, a majority of examiners rarely reopen prosecution, and a relatively small number of examiners do it a lot.
I had expected to see a correlation between harder examiners (lower grant rate) and reopen rate, but the data does not show that. The scatter plot doesn't provide much insight into why some examiner's reopen more often than others, but it does help identify the examiners who reopen too much and could use additional training. As a practicing patent attorney, I hope the USPTO will look into this data more closely to increase the quality of prosecution decisions.