Patent prosecution is the negotiation between an applicant and a USPTO examiner that turns a filed application into an issued patent, or into an abandonment. It is not a court proceeding and there is no opposing party. It is an ex parte exchange of written positions: the examiner searches the prior art and explains why your claims are not allowable, you amend or argue, and the process repeats until the claims are allowed, abandoned, or appealed.
Two things about it surprise people. First, rejection is the normal opening move, not a verdict; the large majority of applications that eventually issue are rejected at least once. Second, the record you build along the way is permanent and public, and it constrains what your patent means for the rest of its life.
The sequence is fairly consistent, whatever the technology.
You may begin with a provisional application, which is a 12-month priority placeholder. It is never examined, needs no claims, and never itself becomes a patent. Its only job is to fix an early filing date that a non-provisional filed within 12 months can claim, and it does that only for subject matter it actually describes well enough to satisfy 35 U.S.C. § 112.
The non-provisional is the real application, and it enters an examiner’s queue by technology art unit. The first action on the merits is the examiner’s first substantive response, usually landing somewhere around 14 to 20 months after filing. It typically rejects most or all claims under § 102 for anticipation, § 103 for obviousness, § 112 for indefiniteness or lack of enablement, or § 101 for ineligible subject matter.
Before that, the examiner may issue a restriction requirement under 35 U.S.C. § 121, saying the application claims two or more independent inventions and you must elect one. The non-elected claims are not lost; you can pursue them in a divisional application, which keeps the original priority date. Section 121 also contains a safe harbor: if the divisional is filed before the other application issues, neither resulting patent can be used as a reference against the other.
The first rejection is non-final. You have a broad right to respond by amending the claims, arguing that the examiner misread the reference, or both.
If the examiner is unpersuaded, the next rejection is usually final. This is a term of art and it misleads almost everyone. It means prosecution on the merits is closed, so the examiner no longer has to enter your amendments. It does not mean the application is dead. From a final rejection you have several live paths:
Everyone substantively involved in prosecution owes the USPTO a duty of candor and good faith under 37 C.F.R. § 1.56. You do not have to search for prior art, but you must disclose material information you know about, normally by filing an information disclosure statement.
Breaching that duty is inequitable conduct, and the remedy is brutal: the entire patent becomes unenforceable, not merely the tainted claims, and the taint can spread to related patents in the family. For decades the defense was pleaded in nearly every infringement case, an epidemic the Federal Circuit itself called a plague.
Therasense, Inc. v. Becton, Dickinson & Co. (Fed. Cir. 2011) (en banc) tightened it substantially. An accused infringer must now prove but-for materiality, meaning the USPTO would not have allowed the claim had it known the truth, and specific intent to deceive as the single most reasonable inference from the evidence. Gross negligence is not enough, and materiality and intent are separate elements that cannot be traded off against each other. The court preserved a narrow exception for affirmative egregious misconduct, such as filing a false affidavit, which is material per se.
This is the doctrine that makes prosecution strategic rather than clerical. Everything you file is published in the application’s file wrapper and is public. Courts read it.
Under prosecution history estoppel, when you narrow a claim for a reason related to patentability, you surrender the ground between the original claim and the amended one. You cannot later recapture it through the doctrine of equivalents, which otherwise lets a patentee reach things that are insubstantially different from the literal claim language.
Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722 (2002), is the controlling authority. The Supreme Court rejected an absolute bar, holding instead that a narrowing amendment creates a rebuttable presumption of surrender. The patentee can rebut it by showing the equivalent was unforeseeable at the time, that the amendment’s rationale bore only a tangential relation to the equivalent, or some other reason it could not reasonably have been described.
Arguments carry the same weight as amendments. Telling an examiner your invention “requires a rigid housing” to distinguish a reference will be read back at you by a district court construing that claim years later. This is why experienced practitioners say as little as possible on the record and prefer amendments that are precise over arguments that are colorful.
Applications publish 18 months after the earliest priority date under 35 U.S.C. § 122(b), whether or not a patent ever issues. A withdrawn or abandoned application that has already published has handed your disclosure to the public for nothing. You can request nonpublication, but only if you certify you will not file abroad in a country that publishes at 18 months.
Because the 20-year term runs from the earliest non-provisional filing date, prosecution time is subtracted from your patent’s life rather than added to it. Patent term adjustment under 35 U.S.C. § 154(b) gives some of it back for USPTO delay, such as missing the deadline for a first action, though applicant delay is subtracted from the award. Pulling in the other direction, a terminal disclaimer filed to overcome obviousness-type double patenting deliberately gives up term, cutting the later patent’s expiration back to match an earlier related one and tying the two together in common ownership.
If the timeline itself is the problem, Track One prioritized examination buys speed for a substantial extra fee, targeting final disposition in roughly 12 months.
What is the patent prosecution process? Patent prosecution is the back-and-forth between an applicant and a USPTO examiner that turns a filed application into an issued patent or an abandonment. The examiner searches the prior art and issues office actions rejecting claims; the applicant responds by amending claims or arguing. It is an ex parte negotiation, not litigation, and it typically runs two to three years.
How long does patent prosecution take? The USPTO’s first action on the merits usually arrives roughly 14 to 20 months after filing, and total pendency to grant or abandonment commonly runs two to three years. Track One prioritized examination targets final disposition within about 12 months for an extra fee. Delay caused by the USPTO can be added back to the patent term as patent term adjustment under 35 U.S.C. § 154(b).
What is the difference between a non-final and a final rejection? A non-final rejection is the examiner’s opening position, and the applicant has a broad right to amend in response. A final rejection closes prosecution on the merits: amendments are entered only at the examiner’s discretion. Final does not mean the end. The applicant can file a request for continued examination, appeal to the PTAB, or file a continuation.
Why does what you say during prosecution matter later? Everything in the prosecution history is public and binds the patent afterward. Under prosecution history estoppel, narrowing a claim to get around prior art surrenders the territory you gave up, so you cannot reclaim it under the doctrine of equivalents. Festo Corp. v. Shoketsu Kinzoku (2002) governs this. Arguments made to the examiner also shape claim construction in court.
Going further: How to Patent an Idea, step by step .
This page is general legal information, not legal advice, and it does not create an attorney-client relationship.
The Supreme Court held that a Section 145 applicant may introduce new evidence with no special limits, and the court must then find those facts de novo.
The Federal Circuit's en banc ruling held that gross negligence alone cannot prove the deceptive intent required for inequitable conduct.
The Federal Circuit held that a first-filed, first-issued, later-expiring patent claim cannot be invalidated for double patenting by a later-issued, earlier-expiring family member, rescuing patent-term adjustment.
The Federal Circuit held that obviousness-type double patenting is measured against a patent's post-PTA expiration date, reshaping prosecution strategy for patent families.
The Federal Circuit affirmed that decades of unreasonable, unexplained delay in prosecuting continuation applications can render the resulting patents unenforceable.
The Federal Circuit held that a later-issued, earlier-expiring patent can serve as a double-patenting reference, anchoring the doctrine to expiration dates in the post-URAA world.
The en banc Federal Circuit rebuilt the inequitable-conduct doctrine around but-for materiality and specific intent to deceive, narrowing the 'plague' of unenforceability defenses.
