How the examiner statistics work

Which examiners have a public profile

Public profiles cover 40 examiners, each chosen when first published from those who had issued the most office actions in the previous two years and had a complete record, including at least 150 applications and enough decided applications to rank. Profiles added since October 2026 also needed rejection data for at least half of the applications they acted on. A published profile is refreshed on every run and kept even if it later falls below these thresholds. Examiner names are not identities, since several people can share a name. An examiner whose records cannot be separated from someone else's does not get a profile. The full dataset in AI Studio covers every examiner.

Allowance rate

The allowance rate is the share of decided applications that issued as patents. Decided means patented or abandoned. Applications still pending are left out, because they have no outcome yet. Status comes from each application's current USPTO status. Other sites may count pending applications, measure at a fixed point in prosecution, or use different years, so the same examiner can show a different rate elsewhere.

First-action allowance is the share of decided applications that issued as patents without any rejection. The outcome chart splits all applications in the window into patented, pending and abandoned, rounded so the three add to 100%.

USPTO and art-unit benchmarks

Newer filings have had less time for office actions, RCEs and abandonments, so a flat average would make an examiner with an older docket look harder than they are. Every USPTO figure on a profile is therefore the USPTO's rate for each filing year, weighted by the examiner's own applications in that year (decided applications for the allowance rate, and applications with a first office action for days to first office action).

Examiners also move between art units. The art-unit figure is the allowance rate of the art units that examined the examiner's applications, weighted by the examiner's decided applications in each art unit and filing year. A profile names every art unit holding at least 10% of them.

Difficulty percentile and label

The difficulty percentile ranks the examiner's allowance rate among every examiner with at least 20 decided applications in the same years. “Harder than 68% of examiners” means 68% of those examiners have a strictly higher allowance rate, and the figure is rounded down so it is never overstated. “Easier than” works the same way from the other side. The label follows the percentile. Below 30 is Favorable, 30 to 60 is Average, 61 to 80 is Hard, 81 to 92 is Very Hard, and above 92 is Extremely Hard.

The percentile counts examiners, each once, while the USPTO average counts applications. The two can point opposite ways, so a rate above the USPTO average can still rank harder than half of examiners. When that happens the profile gives the median examiner's allowance rate, now 84.2%.

Office actions, RCEs and interviews

Office actions here are non-final and final rejections. Restriction requirements and allowances are not counted. Average office actions is the mean number of these actions per application, counting each action once however many rejections it carries. Days to first office action runs from filing to the first rejection, among applications that have had one. The RCE rate and interview rate are the shares of applications with at least one RCE or examiner interview.

Rejections and overcome rates

Rejections are counted by statute (§101, §102, §103 and §112) from the rejections in Boundly's data that the examiner issued on applications filed in the window. That data does not yet cover every office action, so a count can run below the examiner's true total. A rejection is overcome, maintained, abandoned on, or still pending. The overcome rate is the share overcome among those with an outcome, so an application abandoned after a rejection counts as not overcome. A statute is shown only once ten of its rejections have an outcome. §112 covers both definiteness and written description or enablement.

The members' note on how winning responses were built counts the responses that overcame the examiner's most-issued rejection and that Boundly has classified as amended and argued, argued without amending, or amended without arguing. When fewer than five responses have been classified, the note gives how many of those rejections were overcome instead.

Interview and RCE results by year

The year-by-year tables compare decided applications with an interview or RCE against those without, one filing year at a time. The tables show up to four filing years, the most recent that qualify, and leave out the newest filing year. A year qualifies only when it has at least ten applications and at least five decided applications on each side. A profile says the data favors an interview or RCE only when every year shown agrees, gives the range of the gap across those years, and quotes the year with the most decided applications that had one. These comparisons describe the examiner's record. They do not show what an interview or RCE would cause in a given case.

Recent outcomes, classes and contact details

Recent outcomes list the examiner's four most recently decided applications filed in the window, ordered by the date of grant or abandonment. The USPC classes are the examiner's largest, each holding at least 3% of their applications. The art unit technology section and its CPC subclasses describe the art unit and are checked by hand against its grants.

Title, grade, years of service and phone numbers come from records outside Patent Center, and the signatory line follows from the grade. A location appears only with a phone and is the USPTO office its area code belongs to. An email appears only when the USPTO's firstname.lastname format maps to the examiner unambiguously. Where neither is listed, the examiner's direct line is in the closing paragraph of their office actions.

Corrections

If a number looks wrong, or you are the examiner and want something corrected, email [email protected]. The statistics are historical patterns shown for general information. They are not legal advice and do not predict the outcome of any application.