In chemical and materials prosecution, the rejections aren't simple. You get 103 combinations pulling from four or five references, stitching together fragments from different parts of each patent, with an optimization rationale tying it together. These are hard to overcome and expensive to fight. And when the examiner issuing them has an allowance rate in single digits while their art unit peers sit above 50%, you start asking different questions.

A Patent Docs study put numbers to what most of us already feel: “It has become generally accepted in the field that the examiner that is assigned to an application can be as determinative on its outcome as the application itself.” The study found 24 examiners averaging a 6.4% allowance rate, while peers in the same art units above 52%. Most had been at it for more than a decade. None of that surprises you if you've lived it. But it should change how you prosecute.

Why the System Rewards Rejection

The USPTO's quality assurance process audits examiner work primarily for "clear errors" in allowances, or claims that shouldn't have been granted. If you get flagged, and prosecution reopens, your record takes a hit. Bad rejections receive far less scrutiny. A Government Accountability Office (GAO) report confirmed the broader picture: examiners told auditors they prioritize output over quality, and the GAO incentivizes production but offers nothing comparable for quality.  

From the examiner's chair, the math is straightforward. Allow incorrectly, and you face real consequences. Reject incorrectly and nothing happens, unless the applicant pays to appeal. That's not bad faith. It's a rational response to a lopsided system. And in a field like ours, where 103 rejections involve multi-reference combinations and optimization arguments that are genuinely hard to unwind, it gives the examiner very little reason to engage. 

What I Have Learned Works, and What Does Not Work

I run examiner analytics on every application. It is not as a nice-to-have. Rather, it drives the strategy at each stage. Allowance rate, interview effectiveness, appeal reversal history, rejection patterns. That data informs whether I recommend an interview, an Request for Continued Examination (RCE), a pre-appeal brief conference, or going straight to appeal. It shapes how I draft the spec to steer toward a favorable art unit in the first place.

What I've found is that the conventional advice doesn't always hold in our space. I've proposed specific amendments to interview agendas and had examiners show up without having looked at them. I've tried escalating to supervisors, but this has never moved the needle. Pre-appeal brief conferences share the same story in my experience. The FY2026 Performance Appraisal Plan (PAP) now requires Supervisor Patent Examiner (SPE) review of the senior examiner's first actions. In theory, this is a meaningful change for experienced primaries who've had minimal oversight for years. In practice, I have not seen it yet.

What has worked is building the record for appeal from day one. In chemical cases, that means nailing down unexpected results, going hard at motivation to combine, and getting every argument in writing, not just talked through in an interview. This means a long response. It also means being straight with clients about why I'm recommending a path. I've started pulling examiner-specific appeal data and reversal rates into my strategy calls, and it changes the conversation. Clients see the logic behind the spend, not just the next invoice. 

Where That Leaves Us

The FY2026 PAP changes are a step in the right direction, and I'm watching to see how they play out. But the core asymmetry, that allowance errors carry more professional risk than rejection errors, is still baked into how examination quality gets measured. Until that shifts, the smart move is to plan around it. Build for appeal from the start. Use the data to pick your battles. Be straight with your clients about the terrain.

The examiners are not the problem. They are responding to the incentives in front of them. The better we understand those incentives, the better we prosecute.