PTABInstitutionStrategy

PTAB discretionary denial

By Ipiry TeamPublished August 31, 20267 min read

Short answer

Discretionary denial is the PTAB declining to institute a petition even though it meets the statutory threshold. The main grounds are parallel district court litigation likely to reach trial first, serial or duplicative petitions against the same patent, and reliance on art the examiner already considered. Institution rates ran roughly 65% in October 2024 and about 37% by February 2026, driven by discretionary practice rather than petition quality.

PTAB discretionary denial is the Board declining to institute a trial that meets the statutory threshold.

Institution is discretionary under the statute, which means satisfying the merits standard establishes eligibility rather than entitlement.

And denials are largely unappealable. A decision that ends the challenge cannot be reviewed, which concentrates a great deal of significance in a single order.

The rates have moved sharply. Roughly 65% institution in October 2024, about 37% by February 2026 — a change driven by discretionary practice rather than by petitions getting worse.

Why the rate matters before the merits

	If instituted	If denied
Preponderance standard available	Yes	No
Stay of parallel litigation	Likely	Unlikely
Estoppel	On a final written decision	None
Petition cost	Spent	Spent
Validity fight	At the Board	In district court, clear and convincing

Denial costs the petition and changes nothing else. The dispute returns to court at the higher standard.

Which makes institution probability a real input to whether filing is worth it, alongside the strength of the prior art.

The main grounds for denial

Ground	Concern
Parallel litigation	District court will resolve validity first
Serial petitions	Repeated attacks on one patent
Parallel petitions	Several filed at once against one patent
Previously considered art	The Office already looked at this
Petition deficiencies	Inadequate showing on the merits

Parallel litigation has been the most consequential, because most petitions are filed by parties already being sued.

The Fintiv factors

The framework for weighing parallel litigation.

Factor	Favours institution when
Stay granted or likely	A stay exists or is likely
Trial date	Court trial is after the Board's deadline
Investment in the parallel case	Investment is low
Overlap of issues	Overlap is limited
Same parties	Parties differ
Other circumstances, including merits	Merits are strong

The trial date factor drives most outcomes. If the district court will reach trial before the Board's final written decision, the efficiency argument for instituting weakens considerably.

Which makes venue relevant to a PTAB decision, since districts vary widely in time to trial. A defendant sued in a fast district faces a harder institution question than one sued in a slow one.

A stay changes everything. If the court has stayed the case pending the Board's decision, the parallel proceeding concern largely disappears.

Stipulations

Stipulation	Effect
Not to pursue the same grounds in court	Reduces duplication concern
Not to pursue any ground that could have been raised	Stronger — approaches full estoppel voluntarily
No stipulation	Overlap factor weighs against institution

A broad stipulation is a real concession and it works. Agreeing not to raise in court anything you could have raised at the Board removes most of the duplication objection.

It is also expensive. You are volunteering estoppel before knowing whether the petition will be instituted at all.

Serial and parallel petitions

	Serial	Parallel
Meaning	Filed sequentially against one patent	Filed at the same time
Board's concern	Repeated attacks, roadmapping	Burden and duplication
Expectation	Later petitions need justification	One petition per patent
Requirement	Explain why not raised earlier	Rank them and justify

Roadmapping is the underlying worry with serial petitions. A petitioner who sees how the patent owner defended the first challenge and files a second, better one, gains an advantage the Board considers unfair.

Parallel petitions require ranking. Where more than one is filed against a patent, the Board expects a ranked list and an explanation of why a single petition was insufficient.

Claim count is the usual justification. A patent with a large number of asserted claims may genuinely require two petitions to address properly.

Previously considered art

Situation	Risk of denial
Art the examiner never saw	Low
Art cited but not applied	Moderate — explain what is new
Art the examiner applied and the applicant overcame	High
Same argument the examiner rejected	High

The question is whether the Office erred materially. Re-presenting art that was considered and distinguished, without showing what the examiner got wrong, invites denial.

Art from a different field is the strongest position. A reference the examiner could not reasonably have found, teaching the claimed element directly, presents no previously-considered problem at all.

Which puts weight on the prior art search. See patent invalidity search.

Worked example: two petitions, same patent

Petition A — denied

Factor	Position
Filed	11 months after service, near the bar
Parallel case stage	Fact discovery closing
Trial date	4 months before the Board's deadline
Stay	Not sought
Stipulation	None
Art	Two references cited during prosecution
Outcome	Discretionary denial

Nearly every factor pointed the same way. Late filing, an advanced parallel case, an earlier trial date, no stipulation and previously cited art.

The merits were reasonable. They did not matter.

Petition B — instituted

Factor	Position
Filed	3 months after service
Parallel case stage	Pleadings
Trial date	Not yet set
Stay	Motion filed
Stipulation	Broad — will not raise these grounds in court
Art	A foreign publication the examiner never saw
Outcome	Instituted

Filing early was the decisive difference. With the parallel case barely started, the efficiency argument favoured the Board deciding validity.

The stipulation and the fresh art removed the other objections. Nothing was duplicated and nothing had been considered before.

Same patent, similar merits, opposite outcomes. The variables were timing and preparation.

Reducing denial risk

Action	Effect
File early, not near the one-year bar	Strongest single factor
Seek a stay in the parallel case	Removes the duplication concern
Offer a stipulation	Reduces overlap
File one strong petition	Avoids parallel petition problems
Use art the examiner never saw	Avoids previously-considered denial
Address the discretionary factors in the petition	Do not leave them to the owner

Filing early is worth more than anything else on that list. The one-year bar is a deadline, not a target, and petitions filed close to it face the worst discretionary posture.

Address the factors affirmatively. A petition that ignores parallel litigation leaves the patent owner to frame it, and they will.

What denial leaves you

Route	Still available?
District court invalidity defence	Yes — clear and convincing standard
Ex parte reexamination	Yes — no deadline, no estoppel
A further IPR petition	Possible, but serial petition concerns apply
Post-grant review	Only if within nine months of issue

No estoppel attaches to a denial, because there is no final written decision. The documentary case survives intact.

Ex parte reexamination is the natural fallback. It has no deadline, no estoppel and a much lower threshold for being ordered. See ex parte reexamination.

Reading the rate honestly

Period	Approximate institution rate
October 2024	~65%
February 2026	~37%

Quote it with the date, always. The figure moves with policy and practice, and an undated number will be wrong within months.

It is an input, not a verdict. Even at 37%, institution brings the preponderance standard and a likely stay, and the petition cost is far below the litigation alternative.

Model it as a probability. Petition cost against institution probability against the value of winning, compared with fighting validity in court at the higher standard.

PTAB discretionary denial: the checklist

File early. Timing relative to the parallel case is the strongest single factor.
Do not treat the one-year bar as a target. It is the last possible date, not the intended one.
Move for a stay in the parallel litigation, and say so in the petition.
Consider a broad stipulation on grounds you will not raise in court.
File one petition per patent unless claim count genuinely requires more, and rank them if it does.
Search for art the examiner never saw, particularly foreign and non-patent literature.
Explain what is new where you rely on previously cited references.
Address the discretionary factors in the petition, rather than leaving them to the patent owner.
Price institution probability into the decision to file, using current dated figures.
Plan the fallback. A denial leaves no estoppel, and ex parte reexamination remains open.

Common questions

What is PTAB discretionary denial?

The Board declining to institute a trial even where the petition meets the statutory threshold. Institution is discretionary under the statute, so meeting the merits standard establishes eligibility rather than entitlement. Denials are generally not appealable.

Why would the Board deny a meritorious petition?

The main reasons are parallel district court litigation likely to resolve validity first, serial or duplicative petitions against the same patent, and petitions relying substantially on art the examiner already considered during prosecution.

What is the Fintiv analysis?

A framework for weighing whether parallel litigation justifies denial. It considers whether a stay has been granted, the trial date relative to the Board's deadline, investment in the parallel case, overlap of issues, whether the parties are the same, and other circumstances.

How much have institution rates changed?

Substantially. Rates ran roughly 65% in October 2024 and fell to about 37% by February 2026. Always quote the figure with its date, because the movement reflects shifting discretionary practice rather than any change in petition quality.

Can a discretionary denial be appealed?

Generally no. Institution decisions are largely unreviewable, which concentrates significant consequence in a decision that cannot be challenged. A request for rehearing is available but rarely successful.

What are serial petitions?

Multiple petitions filed against the same patent by the same or related parties, usually sequentially. The Board weighs factors including whether the petitioner knew of the art earlier, why it was not raised first, and the elapsed time between filings.

What are parallel petitions?

Two or more petitions filed at the same time against one patent. The Board generally expects one petition per patent and requires a ranking and explanation where more are filed, granting multiple petitions only where genuinely justified.

Does previously considered art lead to denial?

It can. Where a petition relies substantially on the same art or arguments the examiner already considered, the Board weighs whether the Office erred in a material way. A petition that simply re-presents considered art is vulnerable.

How do I reduce the risk of discretionary denial?

File early rather than near the one-year bar, seek a stay of the parallel case, file one strong petition rather than several, use art the examiner did not consider, and address the discretionary factors expressly in the petition.

Does a stipulation help?

It can. A petitioner who stipulates not to pursue in district court the same grounds raised in the petition reduces the duplication concern, which weighs against denial on parallel litigation grounds.

What happens after a discretionary denial?

The petition fee is spent, no estoppel attaches because there is no final written decision, and the validity dispute continues in district court under the clear and convincing standard. Ex parte reexamination may remain available.

Should I still file if denial rates are high?

It depends on the alternative. Even at reduced rates, the preponderance standard and the prospect of a stay make institution valuable. The calculation is the cost of the petition against the probability of institution and the value of winning.

PTAB: the Patent Trial and Appeal Board

What the PTAB is, the proceedings it hears, who the judges are, how a trial runs, what it costs, and why institution rates have moved so sharply.

Inter partes review: what an IPR is and when to file one

IPR is the PTAB proceeding to cancel patent claims over prior art. The one-year bar, the institution decision, what estoppel costs you, and why institution rates have collapsed.

Patent litigation: process, cost and timeline

How a patent case actually runs — pleadings, claim construction, discovery, expert work and trial — what each stage costs, why most cases settle, and the alternatives worth exhausting first.

What is your patent worth?

Free AI valuation in 60 seconds, benchmarked against 652,000 KPSS patent value records and the Ipiry Patent Survival Curve.

Value My Patent Free →Takes 60 seconds. No credit card required.