The PTAB Institution Collapse: Rethinking IPR Strategy

Inter partes review was supposed to be the reliable arrow in the patent litigator's quiver. From the time the America Invents Act created the proceeding in 2013 through roughly 2022, the Patent Trial and Appeal Board granted IPR petitions at approximately 67 percent — a figure that made institution almost a default assumption when sophisticated defendants mapped their defense strategy. Serve a complaint, wait twelve months, file a well-constructed petition. The Board would likely take the case.

That calculus is gone.

By April 2026, according to data reported by IPWatchdog, the PTAB's overall IPR institution rate had fallen by 43 percent from recent baselines. Under USPTO Director John Squires, who assumed personal control of institution decisions in October 2025, the decline reached its logical endpoint: 34 consecutive petitions denied, a reported zero-percent institution rate on the Director's own docket. Practitioners who built litigation strategies around IPR as a near-automatic proceeding are operating with a map that no longer matches the terrain.

How the Collapse Happened

The legal mechanism is discretionary denial — the PTAB's authority under 35 U.S.C. §§ 314(a) and 324(a) to decline to institute a trial even when a petition presents a facially valid unpatentability argument. The Board has always possessed this authority. For most of the AIA era, it used it sparingly.

The shift arrived in two stages. In February 2025, then-Acting Director Coke Morgan Stewart rescinded the June 2022 Director Guidance Memo that had substantially curtailed one of the PTAB's primary discretionary denial frameworks — the Fintiv factors. That framework, established in the PTAB's 2020 precedential decision Apple Inc. v. Fintiv, allows the Board to weigh six factors tied to the status of parallel district court proceedings when deciding whether to grant institution. The 2022 memo had effectively sidelined Fintiv for three years. Rescinding it brought the full six-factor analysis back.

The six Fintiv factors include the proximity of the parallel district court trial date to the PTAB's projected final written decision, the extent of investment already made in the parallel proceeding, and the overlap between issues before the PTAB and issues in district court. In the current environment, a petitioner facing a district court case with a trial date inside eighteen months, substantial prior claim construction proceedings, and overlapping invalidity issues faces a strong Fintiv headwind regardless of how compelling the prior art is.

By October 2025, Director Squires centralized institution decisions entirely and began denying petitions at a rate without precedent in PTAB history. One tool that had partially blunted Fintiv — the Sotera stipulation, in which the petitioner commits in writing to not raising in district court any ground that reasonably could have been raised at the PTAB — no longer works as a shield. In March 2025, the PTAB announced that a Sotera stipulation is no longer dispositive on discretionary denial and cannot override the Director's independent judgment. Even compelling petition merits alone cannot ensure institution.

The Estoppel Problem Is Still Alive

Nothing about the current environment reduces the severity of estoppel exposure for petitioners who do get instituted. Under 35 U.S.C. § 315(e)(2), a petitioner who receives a final written decision from the PTAB is estopped in any subsequent civil action from asserting that a claim is invalid on any ground "raised or reasonably could have raised" during the IPR. The "reasonably could have raised" standard sweeps broadly — the Federal Circuit has held it covers any ground a skilled searcher conducting a diligent search would have been expected to discover.

The consequence is permanent. A petitioner who runs an IPR on a set of prior publications and receives a final written decision, even a favorable one on some claims, cannot return to district court with a different invalidity theory built on prior art patents or publications it chose not to raise, or that a competent prior art search would have surfaced.

There is a recent carveout worth knowing. The Federal Circuit's May 7, 2025 decision in Ingenico Inc. v. IOENGINE, LLC narrowed the estoppel boundary in one specific dimension: § 315(e)(2) does not estop invalidity theories based on prior use, prior sale, or public use by others — the physical-world invalidity grounds under § 102(a)(1) — even when the same patents or publications were used in the IPR. The court held that § 315(e)(2) estoppel is limited to invalidity grounds rooted in "patented or described in a printed publication." Arguments based on a prior commercial product, a prior use in commerce, or a pre-filing public demonstration survive an IPR outcome.

For petitioners, this means the invalidity mapping before filing matters more than ever. If the most powerful invalidity argument rests on a physical prior product, a commercial sale, or a documented prior use, that argument should be reserved for district court — not because IPR is unavailable, but because filing the IPR potentially estops the printed-publication-based invalidity case while the physical prior art case remains live either way.

The One-Year Bar Still Runs

Nothing in the PTAB's current posture changes the statutory filing deadline. Under 35 U.S.C. § 315(b), an IPR petition must be filed no later than one year after the date of service of a complaint alleging infringement of the challenged patent. The clock starts on service. The deadline is jurisdictional and absolute.

This creates a tension the current environment makes sharper. Filing to preserve the option is prudent; filing a weak petition that receives a fast discretionary denial can, depending on the claim construction development in district court, leave the petitioner in a worse litigation position than if it had never filed. The discipline required is to treat the one-year window as preparation time, not as a filing deadline for an unfinished petition.

A petition that reaches the Director's desk with genuinely compelling prior art — multiple strong independent grounds, tight claim construction, and a clear record of why the prior art reads on the claims — still stands a better chance than a thin or redundant petition. The Director's denials have concentrated on petitions that appear weak on the merits, redundant of district court proceedings, or timed to create delay rather than to challenge validity. Filing a compelling petition remains the best argument available against discretionary denial, even if it is no longer a guarantee.

When IPR Still Makes Sense — and the Alternative Worth Reconsidering

For patents where the prior art case is genuinely strong, the district court proceeding is at an early stage, and the invalidity case rests on printed publications rather than physical prior art, a well-constructed IPR petition still offers something district court cannot: a twelve-month statutory timeline from institution to final written decision, a technically expert tribunal, and an invalidity record built under an adversarial PTAB proceeding. None of that has changed.

Where IPR is the wrong tool — because the parallel litigation is too far advanced, because Fintiv makes institution unlikely, or because the strongest invalidity arguments rest on physical prior art — practitioners are taking a fresh look at ex parte reexamination. Reexam is not subject to the one-year bar, does not require a pending district court complaint, and is not vulnerable to Fintiv-style discretionary denial the way IPR is. The tradeoff is structural: reexam is a proceeding between the requester and the USPTO; the patent owner participates in prosecution but there is no inter partes record; and unlike IPR, a reexam outcome does not carry § 315(e)(2) estoppel into district court.

For defendants who want to challenge claim validity without locking in estoppel, for cases where the district court proceeding is too advanced for Fintiv to be manageable, or for situations where the patent owner is not yet asserting the patent in litigation, ex parte reexam is worth serious evaluation. It was widely used before IPR made it obsolete for many purposes. The current PTAB posture is beginning to reverse that obsolescence.

What Practitioners Should Be Doing Now

The firms that adapt fastest will treat IPR as one instrument in a portfolio strategy rather than as a default. Before filing, that means: a rigorous prior art assessment distinguishing printed-publication-based invalidity from physical prior art invalidity; a realistic Fintiv read based on the parallel litigation timeline; a genuine merits evaluation of whether the petition is compelling enough to survive the Director's current review; and a deliberate choice about which invalidity theories to preserve for district court versus commit to the PTAB record.

For patent litigation and prosecution strategy, the era of filing IPR petitions as reflexive defense moves is over. The PTAB's institutional behavior has changed. The strategies built around it need to change with it.