Skip to content

USPTO Proposal: Accept Applicant’s Independent Prior Art Search as Pre-Examination Search

USPTO Proposal: Accept Applicant’s Independent Prior Art Search as Pre-Examination Search published on No Comments on USPTO Proposal: Accept Applicant’s Independent Prior Art Search as Pre-Examination Search

The USPTO is seeking to implement a pre-examination automated search to assist the start of examination. However, this initiative overlaps with applicants’ earlier search efforts to inform the intake and drafting processes – which the applicant submits as an Information Disclosure Statement… and which examiners routinely disregard.

Instead, the USPTO could identify third-party search vendors that it trusts to provide reliable pre-examination searches. Applicants could choose among these vendors for prior art searching to inform disclosure review and application drafting, and the USPTO could accept the same report (optionally updated based on the filed application) as its pre-examination search. This process yields a variety of advantages – especially the implementation of the USPTO’s pre-examination search with 0% of the cost.



One of the USPTO’s recent proposals for assisting examination, in the interest of patent quality, involves an automated pre-examination search, described as follows:

The USPTO is pursuing automated tools to assist the examiner in prior art discovery that may include advanced natural language and linguistic technologies. The USPTO understands applicants are typically interested in receiving a patent as quickly as possible with the most relevant art considered by the examiner on the record at the earliest point possible during prosecution. Thorough review of prior art, as efficiently as possible, enables these goals.

Automated searching definitely has much potential… with a few contingencies:

  • Automated searching is applicable differently to different domains: a first set of search techniques can be applied to U.S. patents and patent applications; a second set to academic journals; a third set to web forums and conversations. Still other techniques are applicable to resources in foreign languages. The scope of automated searching should be carefully focused in view of the predicted relevance of a particular domain to the subject invention (e.g., automated searching for medical techniques should focus on medical journals), and the proficiency of automatically searching in that domain.

  • Automated searching can vary wildly in quality, and at worst, can bury the examiner in irrelevant references. Moreover, there is undoubtedly a tradeoff between search quality and cost.

  • Automated searching cannot substitute for examiner-initiated searching, but must be considered a supplement and/or adjunct. The automated search may fails to turn up a relevant result due to language differences or other anomalies; the examiner’s follow-up search by orthogonal techniques, such as by classification, should be able to catch it.
Of course, the USPTO’s efforts to implement an automated search require both resources and effort – both of which are vital resources, given the wide range of improvements the USPTO could pursue.

Meanwhile, many applicants are also performing searching – but this occurs at the disclosure review phase, well in advance of the filing of the application (or even drafting it). These searches are vitally important for both the applicant, in determining whether or not to pursue patent protection, and for the practitioner, in drafting the application to distinguish the invention from known references. Relevant references identified in the search report must be submitted to the USPTO to satisfy the applicant’s Rule 56 obligations. However, statistics show that examiners rarely rely on applicant-submitted references.

Why does this occur? The academic paper revealing this trend was unable to provide a substantiated explanation. Anecdotes exist that many examiners mistrust applicant-initiated prior art, as possibly being submitted with nefarious purposes: to bury a relevant reference in a pile of irrelevance; to distract the examiner from aspects with likely prior-art issues; or to shape – and perhaps distort – the examiner’s understanding of the invention.

Regardless of the reason, this disregard and re-searching is a significant source of inefficiency, as revealed by the following odd scenario:

  1. Upon receiving a disclosure, the applicant pays a search vendor to perform a search.

  2. The search vendor provides a search report, which the applicant uses to evaluate and draft an application for the invention.

  3. The applicant submits the search report to the USPTO. The USPTO disregards the IDS.

  4. The USPTO utilizes some resources to perform an automated prior art search.

  5. The examiner uses the automated prior art search to guide examination.
The oddity of this scenario looks even more serious if the USPTO happens to engage the same third party for its own prior art search. This interaction is completely redundant – perhaps performed by the same process, over subject matter presented in very similar documents (the invention disclosure vs. the filed patent application).

An alternative process for this scenario could reflect the fact that if the applicant is already paying for a search by a third party… why shouldn’t that same search serve as the USPTO’s automated prior art search? Particularly where the search is performed by an approved third-party vendor that is disinterested in the outcome of the search, and is singly intent on providing an objective and thorough search?

A revised process could look like this:

  1. Upon receiving a disclosure, the applicant contracts with a USPTO-approved search vendor to perform a search.

  2. The search vendor provides a search report, which the applicant uses to evaluate and draft an application for the invention.

  3. Optionally, the applicant re-submits the application draft to the search vendor for an update based on the particular language of the draft. (May lead to re-drafting or reconsideration of the application if the updated search identifies any new references.)

  4. When the applicant files the application, it references the search report provided by the third party.

  5. The USPTO obtains the search report from the third party, and uses it as the pre-examination search.

  6. In exchange, the applicant receives some advantage, such as (i) increased priority for the application, or (ii) a modest reduction in filing fees.
The advantages are numerous:
  • Avoiding the cost and resource allocation needed to implement the USPTO’s pre-examination automated search.

  • Reusing a third-party search, rather than utilizing resources of both the applicant and the USPTO to search the same invention.

  • Ending the “we require applicant-submitted prior art, but don’t trust or use it” inefficiency.

  • Putting the applicant and the examiner on the same page from the start of examination, since both are looking (initially) at the same base of prior art references.

  • Optionally – for a set of search vendors, allow the applicant to choose between an adequate search vendor with economical rates, and (for more important applications) a pricier but higher-quality search by a more proficient vendor, and directly translating that choice into the amount of deference granted by the examiner over the results.

  • Added value for applicants in exchange for providing search reports that are already being purchased.
This program could be optional – and for applicants that choose not to engage an approved third-party search vendor for a search, the USPTO could charge an examination fee to fund its own pre-examination search. But for applicants that are already paying these costs, the USPTO could leverage their efforts to meet its pre-examination search goals… with 0% of the energy and cost investment.

Leave a Reply

Your email address will not be published. Required fields are marked *