The USPTO has roughly 10,000 patent examiners. When the Office changes direction — on allowances, on examiner incentives, on what counts as patent-eligible subject matter — those changes don’t move through the system overnight. The question right now is what the changes will mean and how much has already changed.
That context matters. The previous administration left the USPTO with a record backlog and Section 101 rejection rates that had climbed back to pre-2019 levels — Voice of IP broke that story in June 2024, when 24% of all office actions contained a §101 rejection and the AI art units had hit 77%. The current administration has been working to reverse those trends, issuing new §101 guidance, signaling a more balanced approach to patent eligibility, ramping up the hiring of new examiners, and refocusing the USPTO on its core function of examining patent applications.
For this episode of Clause 8, Eli reached out to Francesca Cruz for her team at Juristat to pull the data together and then invited Clause 8 return guest Clint Mehall to react to the findings in real time.
Who’s actually driving allowances — and why it matters for the backlog
One of the reasons Eli wanted to look at examiner experience data was straightforward: if the administration wants to reduce the backlog, understanding which examiners are positioned to finish examining cases — and under what conditions — is essential.
Some tools that measure examiner behavior use tenure — years at the USPTO — as a proxy for experience. Juristat’s analysis suggests that’s not fully the right metric, and that the number of applications an examiner has actually worked through, at least relative to other examiners in the same art unit or working group, is often a more revealing indicator of their likely approach.
And, as even the USPTO recognizes, primary examiners with full signatory authority operate differently than even “very experienced” examiners who still need supervisor sign-off on allowances.
The findings on how junior, mid-level, and senior examiners actually compare — on office actions, allowance rates, and consistency — are counterintuitive in ways worth digging into. One data point that particularly stood out: it’s often not the newest examiners who tend to produce the longest prosecutions. Clint had a theory for why — invoking the Dunning-Kruger effect — that’s worth hearing in his own words.
The data also highlights something practitioners already sense but rarely see quantified: the variation between individual examiners within the same art unit can be as wide as the variation between entire technology centers. For that reason, examiner-level data is often more useful than art unit averages when thinking about prosecution strategy. Notably, the USPTO's Office of Patent Quality Assurance (OPQA) is now actively working on how to make apples-to-apples comparisons across examiners handling similar technology to identify meaningful deviations. That's a significant development worth watching if the USPTO is able to find a way to address address that variation.
The full breakdown, including how the examiners were classified for the data discussed in this episode will be shared in a dedicated post coming soon.
Interviews after final: what the numbers actually show
The USPTO has recently pointed to data showing that the share of applications receiving at least one interview has grown — from around 30% to roughly 35% since mid-2025. That’s accurate, and interviews are worth doing: Juristat’s data shows allowance rates are roughly 10% points higher for applications where an interview occurred.
Juristat's analysis found that following the recent PAP changes, the number of interviews after final office actions has decreased. However, it's too early to tell whether that remains the case, and if so, what the downstream effects will be.
What happened after the AFCP program ended
When the After Final Consideration Pilot program was discontinued in December 2024, the stated reason was that the cost of the program was too high for the USPTO relative to the benefits. The data confirms that the results of the AFCP program were underwhelming — but discontinuing the program has caused a meaningful shift in examiner and practitioner behavior with potentially negative implications for reducing the backlog.
Allowance rates after final without an RCE have dropped meaningfully since the program ended. At the same time, allowance rates after final with an RCE have edged up. The numbers tell a clear story about how practitioners have adjusted, which Fran and Clint dig into in the episode.
Section 101: early signals, not yet a trend
The August 2025 guidance memo — reinforced when a key PTAB decision became precedential in September — was the clearest signal yet that this administration wants §101 rejection rates to come down. Comprehensive examiner training on the updated guidance is now being planned for rollout later this month, which the USPTO expects will drive more meaningful change in the data over the rest of the year.
The data so far reflects the lag before that training takes effect — overall rates haven’t moved significantly in the aggregate. But there are art-unit-level signals worth watching, particularly in the AI working group, where rejection rates spiked dramatically under the previous administration and have begun to show some movement. Business methods and medical device art units tell a different story, and the episode walks through each of them.
A dedicated Voice of IP post on the §101 data — including art-unit breakdowns and examiner-level outliers — is coming soon.
🎧 Watch the full episode above or listen on your favorite podcast app—and subscribe to the new Clause 8 YouTube channel for bonus content.
📌 Presented by Tradespace – where ideas take flight.
