Filing or Defending an Inter Partes Review: A Practical Guide

Document type: Guide Practice area: Intellectual Property — Patents Jurisdiction: United States (federal) Last reviewed: 5 September 2026

An inter partes review is a compressed, document-heavy proceeding in which almost everything is decided by two filings: the petition and the patent owner's preliminary response. Institution follows, or it does not, and the parties spend the next year litigating a case whose shape was fixed in the first four months.

That front-loading is the defining operational fact. A petition drafted under time pressure with art that has not been fully developed cannot be repaired later — the Board will not permit new grounds, and after SAS Institute Inc. v. Iancu, 584 U.S. 357 (2018) a weak ground can sink the whole petition.

This guide walks the work from both sides, in order.

PART ONE — THE PETITIONER'S DECISION

Step 1: Run the framework before doing anything else

Nine questions, in order. The companion article sets them out; here is how to answer them operationally.

1. What is the invalidity case made of? Separate the art into printed publications (usable at the Board) and public use, on-sale activity, and system art (not usable, and not estopped).

2. What would estoppel cost? List the district court invalidity grounds. Cross out everything that is patents-and-printed-publications. What is left is your post-IPR defense. If nothing is left, this is a forum choice with no fallback.

3. Is the primary defense non-infringement? The Board cannot decide it.

4. Are we within the one-year bar, and who are the real parties? See Step 2 — this deserves its own workstream.

5. What is the district court trial date, and will the court stay?

6. Was the best art before the examiner? If so, prepare the § 325(d) answer.

7. Can we stipulate?

8. What does it cost, against what?

9. Who else should join?

Document the answers in a memorandum. The decision gets revisited by the client, and the analysis should exist in writing.

Step 2: Do the real party in interest analysis properly

This is a discrete workstream and it is where petitions die.

Identify every entity that has been served with a complaint on the patent: your client, its parents and subsidiaries, its suppliers, its customers, its indemnitors, and every member of any joint defense group.

For each, ask:

• Does it fund the petition, in whole or in part?
• Does it direct or control the petition — selecting counsel, approving strategy, reviewing drafts?
• Does it have a right to control?
• Is there an indemnity obligation, and what does it entitle the indemnitor to?
• Would it benefit from cancellation of the claims?
• Is there a pre-existing relationship — corporate, contractual, or in a joint defense group — that makes it a privy?

When was each served? If any real party or privy was served more than one year ago, the petition is barred under § 315(b) — and after Thryv, Inc. v. Click-To-Call Technologies, LP, 590 U.S. 45 (2020), the Board's determination is not appealable.

Err toward disclosure. Naming an entity creates estoppel against it, which is a cost. Failing to name one that should have been named can void the petition, which is fatal. The asymmetry favors disclosure.

Document the analysis contemporaneously, with the funding and control facts recorded. If the patent owner challenges the identification, you will need it.

Step 3: Develop the art properly, and early

Search comprehensively. The estoppel standard reaches what a skilled searcher conducting a diligent search reasonably could have found. Do that search now, both because it produces the best petition and because it defines your exposure.

Separate the categories:

• Printed publications — patents, published applications, articles, standards documents, manuals, and theses. Confirm public accessibility for anything unconventional; a manual or a thesis requires evidence of how it was indexed and available.
• System art — products, public uses, offers for sale. Preserve this separately. It is your post-estoppel defense and it should not be discussed in the petition.

Build the combination story. Most IPRs are decided on obviousness, and most obviousness cases are decided on whether there was a reason to combine. A petition that asserts a combination without a technically grounded, evidence-supported motivation will lose.

Engage the expert early. The declaration is the substance of the petition, and an expert brought in three weeks before filing produces a declaration that reads as though it were.

PART TWO — THE PETITION

Step 4: Draft for all-or-nothing institution

After SAS, the Board institutes on everything or nothing.

The consequences for drafting:

• Fewer grounds, better developed. Two strong grounds beat five with two strong ones, because the weak grounds invite the Board to view the petition as a whole less favorably.
• Fewer claims where sensible, though challenging all asserted claims maximizes the simplification argument for a stay.
• Every ground must stand on its own. There is no partial institution to fall back on.

The structural requirements under 37 C.F.R. Part 42:

• Identification of the real parties in interest.
• Identification of related matters.
• Lead and back-up counsel.
• Service information.
• Grounds for standing.
• Identification of each claim challenged, each ground, and the evidence supporting it.
• Claim construction, where a term requires it.
• Word count compliance — and the limit is real; the petition is a compressed document.

The claim chart is the petition. Element by element, with the reference language quoted and cited by page and line, and with the expert declaration tied to specific chart entries.

The motivation to combine section deserves disproportionate attention. Explain, with evidence: what problem a skilled artisan faced, what the references teach about solving it, why a skilled artisan would look from one to the other, and what reasonable expectation of success existed. Conclusory motivation statements are the most common reason petitions fail.

Step 5: Address discretion affirmatively

Do not wait for the preliminary response to raise it.

On parallel litigation:

• State the filing date relative to the complaint. Early filing is your best fact.
• State the district court schedule and any stay motion filed.
• Offer the stipulation, in the petition, in specific terms.
• Address the overlap between the petition's grounds and the district court invalidity contentions.

On § 325(d):

• Identify any art that was before the examiner.
• Explain specifically what the Office did not appreciate: a limitation the examiner never addressed, a teaching in the reference never discussed, a combination never considered.
• Do not assert that the examiner simply got it wrong; explain the error.

Check the current guidance before filing. This area has moved more than once and the current framework governs.

Step 6: The stipulation

The most effective single tool available to a petitioner facing discretionary denial concerns.

The narrow form: the petitioner will not pursue in district court the specific grounds raised in the petition.

The broad form: the petitioner will not pursue in district court any ground it raised or reasonably could have raised in the petition — that is, it accepts the full statutory estoppel as of institution rather than as of the final written decision.

Which to offer. The broad form is substantially more persuasive and it costs less than it appears, because § 315(e) estoppel would attach anyway on a final written decision. What you give up is the ability to walk away from the IPR and preserve those grounds — which is worth something, and less than the stay.

Where it goes. In the petition, and again in the stay motion, and again in any reply on discretion.

Step 7: Move for a stay

Timing. Many courts will not stay before institution. File the petition early, then move promptly after institution — or move before institution where the local practice permits, with a request to renew.

The motion:

• Simplification. How many asserted claims are challenged, and what happens to the case if they are cancelled. Quantify it.
• Stage. Where the case stands: discovery open, no trial date, no substantial investment.
• Prejudice. Address the patent owner's position honestly. A competitor has a real argument; a non-practicing entity's remedy is money.
• The stipulation. Lead with it.
• Burden reduction, where the district applies a fourth factor.

If the stay is denied, reassess. Running both tracks costs roughly 130 percent of one, and the value of the IPR drops considerably.

PART THREE — THE PATENT OWNER'S RESPONSE

Step 8: The preliminary response

Filed within three months of the notice of filing date. It is optional and it should almost always be filed, because roughly a third of petitions are denied and this is where that happens.

Argue in this order:

1. The bars. Section 315(b) is the strongest available argument when it applies.

• Identify every entity served more than a year ago.
• Establish the relationship to the petitioner: funding, control, right to control, indemnity, joint defense, corporate affiliation.
• Request additional discovery on real party in interest if the facts are within the petitioner's knowledge — the Board grants it in appropriate cases.
• Remember that a favorable determination is unreviewable after Thryv; this argument must win here.

2. Discretionary denial.

• Parallel litigation: the district court schedule, the trial date relative to the Board's projected decision, the investment already made, the overlap in issues, and whether a stay has been denied.
• Section 325(d): the art or arguments were before the examiner and the petitioner has not shown material error.
• Address the petitioner's stipulation directly if one was offered — a narrow stipulation leaves substantial overlap.

3. The merits. Why the petition fails the reasonable-likelihood threshold:

• A claim limitation not disclosed by any reference.
• No adequate motivation to combine — the most productive merits argument.
• A misreading of a reference, shown by the reference's own text.
• An expert declaration that recites conclusions without analysis.

4. Petition deficiencies. Because institution is all-or-nothing, a defect in one ground is an argument against the whole petition: incomplete charts, unexplained expert testimony, redundant grounds, inadequate claim construction.

Testimonial evidence. A declaration may be submitted, subject to the rule that a genuine factual dispute is resolved in the petitioner's favor at institution. Use it to establish facts — what a reference discloses, what a term meant to the art at the time — rather than to argue.

Step 9: After institution — the patent owner response

The single most productive line of attack is the motivation to combine. Develop it with the expert and with cross-examination:

• What problem was the skilled artisan actually solving?
• Do the references address different problems?
• Does one reference teach away from the combination?
• Would the combination have destroyed the operation of the primary reference?
• What evidence supports a reasonable expectation of success, beyond the expert's say-so?

Cross-examine the petitioner's declarant. This is the patent owner's principal discovery tool. Establish:

• What the declarant did not review.
• What the reference does not say, in the declarant's own words.
• The steps between the references that the declaration glosses.
• Any inconsistency with the declarant's positions elsewhere, including in district court.

Develop secondary considerations with nexus. Commercial success, long-felt need, failure of others, industry praise, copying, and skepticism. The Board credits these when they are supported and tied to the claimed features — and disregards them when the nexus is asserted rather than shown.

Watch the district court consequences. Every claim construction position taken here will be used against you on infringement. Coordinate.

Step 10: The motion to amend — decide deliberately

Understand the odds. Success rates are low, and the reasons are structural: the substitute claims face the original art plus whatever the petitioner adds, on a compressed schedule, before judges who have read the art.

Understand the cost. Substitute claims that are not substantially identical to the originals create intervening rights, which can eliminate past damages entirely. A patent owner that amends successfully may hold a patent worth far less than the one it had.

Use the preliminary guidance procedure where available. The Board's practice of providing preliminary guidance on a motion to amend, with an opportunity to revise, materially improves outcomes.

The requirements:

• A reasonable number of substitute claims — one per challenged claim, presumptively.
• No broadening.
• Written description support in the original disclosure, identified precisely.
• Responsive to a ground of unpatentability.
• Candor about the prior art of record.

The better alternative, where available. A continuation application filed before the IPR, while the family is pending, produces new claims without intervening rights and without the compressed schedule. Patent owners with active families should be filing continuations as a matter of course.

PART FOUR — RUNNING THE PROCEEDING

Step 11: The schedule

Discovery. Limited and defined:

• Routine discovery — exhibits cited in a paper, cross-examination of declarants, and information inconsistent with a position advanced.
• Additional discovery — only in the interests of justice, on motion, and granted sparingly. The Board applies a demanding standard requiring more than a possibility that something useful exists.
• Depositions — of declarants, on a schedule set by the parties within the Board's framework.

The duty to disclose inconsistent information is real and often overlooked. A party that takes a position at the Board while holding information inconsistent with it must produce that information.

Step 12: The oral hearing

Format. Typically one hour per side, before the three-judge panel, frequently by video.

What it is. Not a trial. No live testimony, no new evidence, no new arguments. Demonstratives must be served in advance and may not contain new material.

What works:

• Know the record by exhibit number. The panel has read the papers and will ask about a specific paragraph of a specific declaration.
• Have a two-minute answer to the case's hardest question, and lead with it if asked.
• Concede what is not in dispute. Panels react well to counsel who narrow.
• Address the panel's questions directly rather than returning to a script.

What does not work: reading the brief aloud, arguing facts not in the record, and attacking the other side's counsel.

Step 13: The final written decision and after

The decision addresses every challenged claim under § 318(a), and it is where estoppel attaches.

Director Review. Following United States v. Arthrex, Inc., 594 U.S. 1 (2021), a party may request review by the Director. It is discretionary, rarely granted, and adds time. Consider it where the panel departed from Board precedent or where a policy question is presented.

Rehearing. Available on a demanding standard — the party must identify what was misapprehended or overlooked, with record citations. Rarely granted.

Appeal. To the Federal Circuit under § 319. Review is de novo on legal questions and for substantial evidence on factual findings, including what a reference teaches and whether there was a motivation to combine. Substantial evidence is a deferential standard, and it is why Board factual findings usually survive.

Standing to appeal. A petitioner without Article III injury may be unable to appeal a loss. A petitioner that is not a defendant, a competitor facing a credible threat, or otherwise injured should assess this before relying on the appeal.

Certificate. After appeal or the expiry of appeal rights, the Director issues a certificate cancelling unpatentable claims and confirming patentable ones. Claims cancelled are gone as against everyone.

PART FIVE — RUNNING BOTH TRACKS

A defendant with an IPR and a district court case has one case with two tribunals, and the failures are coordination failures.

Step 14: Appoint a consistency owner

One named person reviews every filing in both forums before it goes out, checking:

• Claim construction positions. A narrow construction argued at the Board to distinguish art will be quoted against you on infringement; a broad one argued in court will be quoted at the Board.
• Expert statements. The same expert frequently appears in both, and inconsistent testimony is discoverable and devastating.
• Characterizations of the prior art. What a reference teaches should not change between forums.
• Statements about the accused product. These reach the Board through the patent owner's filings.

This is a real role, not a formality. In a case with fifteen filings across two forums over eighteen months, drift is the default and consistency requires someone whose job it is.

Step 15: Manage the information flow

From district court to the Board. Documents and testimony obtained in district court discovery may be used at the Board, subject to the protective orders. Cross-examination testimony from the district court can be powerful before the panel.

From the Board to district court. Board filings are public. Everything the petitioner says about claim scope is available to the patent owner in the infringement case, and vice versa.

Protective order interaction. Confirm that the district court protective order permits use of designated material in the Board proceeding, and that the Board's default protective order accommodates the designations. Address this early; retroactive fixes are awkward.

Step 16: Plan for each outcome

All claims cancelled. The district court case ends as to that patent. Confirm the certificate issues and address any remaining patents.

Some claims cancelled. The case narrows. Reassess damages, infringement, and settlement value on the surviving claims.

All claims upheld. Estoppel now applies. Confirm what invalidity defenses survive — system art, § 101, § 112, inequitable conduct — and rebuild the district court case around them. This scenario should have been modeled before filing.

Denied institution. No estoppel. The full invalidity case remains available in district court, and the denial says nothing binding about the merits — though the patent owner will cite it.

Settled before the final written decision. No estoppel. File the settlement agreement with the Board as required, and address the IPR expressly in the settlement documents; termination is not automatic after institution and the Board may proceed.

PART SIX — A WORKED PROCEEDING

Bellwood Instruments and Kestrel IP Holdings, on the facts from the companion article.

Month 0. Complaint filed in the Western District of Texas on two patents. Trial set for month twenty.

Month 1. Priyanka Osei-Lindqvist runs the framework. Two printed publications, one strong system prior art reference, moderate non-infringement. Estoppel would cost the publications and preserve the system art.

Month 1, week 3 — the real party analysis. Bellwood's supplier Thackery Components was served fourteen months ago and has an indemnity obligation. Priyanka's analysis concludes Thackery is likely a real party in interest. She names it, accepting estoppel against Thackery rather than risking the petition on a contested identification that Thryv would make unreviewable.

Months 1–3 — art development. A comprehensive search produces one additional publication. The expert is engaged in week four and drafts the declaration alongside the petition rather than after it.

Month 3 — petition filed. Two grounds, both obviousness, on three references. Every claim asserted in the district court is challenged. The petition:

• Identifies Bellwood and Thackery as real parties in interest, with the funding and control facts stated.
• Addresses § 325(d): one publication was in an IDS but never applied; the petition explains that the examiner never addressed the reference's disclosure of the disputed calibration timing.
• Offers a broad stipulation: Bellwood will not pursue in district court any ground it raised or reasonably could have raised.
• States the district court schedule and notes the petition's filing at month three.

Month 4 — stay motion filed in the district court, leading with the stipulation and the early petition.

Month 6 — preliminary response. Kestrel argues the time bar based on Thackery (unsuccessful, because Bellwood named it and the petition is within a year of Bellwood's service and of the properly identified real parties), discretionary denial based on the month-twenty trial date, and no motivation to combine.

Month 9 — institution granted on both grounds and all claims. On discretion, the Board credits the early filing and the broad stipulation.

Month 10 — stay granted. The court cites the stipulation, the fact that all asserted claims are challenged, and the early stage.

Months 10–18 — the proceeding. Kestrel's response develops the motivation-to-combine attack and submits evidence of commercial success. Bellwood cross-examines Kestrel's expert, who concedes that the primary reference does not disclose calibration during the measurement cycle — which is the point Bellwood needed and did not have from the papers.

Month 15. Kestrel moves to amend one patent's claims. It uses the preliminary guidance procedure; the Board's guidance identifies a written description problem; Kestrel revises.

Month 19 — oral hearing.

Month 21 — final written decision. All claims of the first patent cancelled. On the second patent, the original claims are cancelled and the substitute claims are entered — meaning Kestrel holds an amended patent with intervening rights that eliminate its past damages claim entirely.

Month 23 — settlement. Kestrel's remaining case is a going-forward royalty on amended claims, against which Bellwood retains its system prior art. The parties settle for a fraction of the original demand.

The three decisions that produced this:

Naming Thackery. It cost estoppel against a supplier that was not going to litigate anyway, and it removed an argument that would have been unreviewable if lost.

Filing at month three. It produced institution notwithstanding a month-twenty trial date, and it produced the stay.

The broad stipulation. It was the single most cited fact in both the institution decision and the stay order, and it cost Bellwood grounds it would have been estopped on anyway.

PART SEVEN — BUDGET, STAFFING, AND MISTAKES

Budget

Multi-petitioner filings materially reduce per-party cost, and joinder under § 315(c) lets later defendants join an instituted proceeding — with their own estoppel consequences.

Staffing

A lead with PTAB experience. The Board's practice — word limits, evidentiary rules, the shape of a persuasive chart, what panels credit — is learned by doing.

A technical expert engaged in the first month, not the last. The declaration is the petition's substance.

One person on real party in interest, treating it as a discrete workstream with a documented conclusion.

A consistency owner reviewing every filing in both forums.

District court counsel in the room for the stipulation decision, because it binds them.

Mistakes that decide these proceedings

Petitioner: filing at month eleven. It worsens discretion, worsens the stay motion, and is entirely self-inflicted.

Petitioner: a defective real party identification. Fatal, unreviewable, and avoidable with a week of diligence.

Petitioner: a conclusory motivation to combine. The most common merits failure.

Petitioner: too many grounds. After SAS, weak grounds drag the strong ones.

Petitioner: an expert brought in late. The declaration reads like it.

Petitioner: no stipulation offered. The cheapest available improvement to both discretion and the stay.

Petitioner: discussing system prior art in the petition. It is not usable at the Board and mentioning it complicates the estoppel argument later.

Patent owner: waiving the preliminary response. A third of petitions are denied and this is where it happens.

Patent owner: arguing the merits without arguing the bars and discretion. The procedural arguments are frequently stronger.

Patent owner: a motion to amend without modeling intervening rights. You can win the amendment and lose the damages.

Patent owner: no continuation on file. The best amendment is one made in a pending application before the IPR.

Both: inconsistent claim positions across forums. The most common self-inflicted wound.

Both: assuming last year's discretionary framework still applies. Check before every filing decision.

PART EIGHT — FREQUENTLY ASKED QUESTIONS

When should we file? As early as the art permits, and within one year of service on every real party and privy under § 315(b). Month three is materially better than month eleven for every downstream question.

Do we have to name our indemnitor? If it funds, directs, or controls the petition, or is a privy, yes. Err toward disclosure: the cost is estoppel against that entity; the risk of omission is a void petition that Thryv makes unreviewable.

Should we offer a stipulation? Almost always, and the broad form. It is the most effective response to discretionary denial and the strongest fact in a stay motion, and it costs grounds you would be estopped on anyway.

Will we get a stay? Depends on timing, institution, claim overlap, and whether the patent owner competes with you. File early, move promptly after institution, and lead with the stipulation.

Can we add a ground later? No. The petition defines the proceeding.

What discovery is available? Routine discovery — cited exhibits, cross-examination of declarants, and inconsistent information. Additional discovery only in the interests of justice, granted sparingly.

Should the patent owner move to amend? Only after modeling intervening rights, and preferably using the preliminary guidance procedure. A continuation filed before the IPR is usually the better tool.

What happens if we settle? Estoppel does not attach if you settle before the final written decision. File the settlement agreement with the Board; termination after institution is not automatic.

Can we appeal a loss? To the Federal Circuit under § 319, on de novo review of law and substantial evidence review of facts — a deferential standard for findings about what references teach. A petitioner without Article III injury may lack standing to appeal.

What is Director Review? Discretionary review of a Board decision by the Director, created to cure the Appointments Clause problem identified in Arthrex. Rarely granted; worth requesting where the panel departed from Board precedent.

PART NINE — THE OTHER VEHICLES

Before defaulting to an IPR, confirm it is the right proceeding.

Post-grant review — 35 U.S.C. § 321. Any ground of invalidity, including § 101 eligibility and § 112 written description and enablement, and including prior public use and on-sale art. Available only for patents with an effective filing date on or after March 16, 2013, and only within nine months of issuance.

When it fits: a competitor monitoring a rival's issuances, with a docketing system that flags relevant grants within the window. Why it is rare: by the time a patent is asserted, the window has closed. The caution: its estoppel is as broad as its scope.

Ex parte reexamination. Any person may request reexamination on patents and printed publications, anonymously if desired, at a fraction of an IPR's cost. A third-party requester who does not participate is not estopped.

When it fits: a defendant already estopped by a prior IPR; a party wanting a low-cost challenge without estoppel; or a situation where anonymity matters. The cost: no participation rights after the request — the proceeding runs between the examiner and the patent owner, and you cannot respond to the patent owner's arguments.

Supplemental examination. A patent owner tool: it permits the owner to have the Office consider information that might otherwise support an inequitable conduct defense, and — if the Office concludes the information does not raise a substantial new question, or if reexamination follows and concludes — it can immunize the patent from an unenforceability defense based on that information. Underused, and worth considering before asserting a patent with a prosecution problem.

Derivation proceedings — § 135. Available where an applicant claims an earlier applicant derived the invention from its inventor. Rare.

Reissue. A patent owner tool for correcting errors, including broadening within two years of issuance. Consider it alongside a motion to amend, and note that it carries its own intervening rights consequences.

The decision tree. Within nine months of issuance with a § 101 or § 112 argument, consider PGR. Otherwise, if the art is printed publications and the estoppel cost is acceptable, IPR. If you want a challenge without estoppel and can live without participation, ex parte reexamination. If your case is non-infringement, system art, or § 112, stay in district court.

PART TEN — WHERE TO GET HELP

PTAB counsel who files regularly. Word limits, chart conventions, what panels credit in a motivation-to-combine section, and how the current discretionary framework is being applied are learned by practice.

A prior art searcher, not a paralegal running keyword queries. The estoppel standard is defined by what a diligent search would find, so the search you commission also defines your exposure. Commission it properly once.

A technical expert engaged in the first month. The declaration is the petition, and an expert who has lived with the art for three months writes a different document than one engaged for three weeks.

District court counsel in the room for the stipulation and stay decisions. They are bound by both.

Appellate counsel before the final written decision, where the exposure is large. The Federal Circuit reviews factual findings for substantial evidence, which means the record built at the Board is largely the record on appeal.

And your own docketing system. Post-grant review windows close nine months after issuance and cannot be reopened. A company monitoring competitors' patents needs a process that flags relevant grants automatically, and most do not have one.

PART TEN-B — MULTI-DEFENDANT AND JOINDER PRACTICE

Where several defendants face the same patent, the economics change substantially.

Joint petitions

Several defendants filing one petition share the cost, present a single unified record, and avoid the redundancy that draws discretionary denial when multiple petitions target the same claims.

What to settle in a joint defense agreement before drafting:

• Who leads, and how decisions are made when the group disagrees.
• Cost allocation — by revenue, by exposure, or equally.
• Estoppel: it attaches to every named petitioner. A member with a strong system prior art defense may prefer not to join.
• What happens when one member settles: does it withdraw, and does the petition survive?
• Common interest privilege scope, and what is shared.
• Expert selection and control.
• Settlement authority — can one member settle without the others?

The disagreement that recurs. Members with different products have different non-infringement positions and therefore different appetites for estoppel. A defendant whose defense is non-infringement gains little from the IPR and bears the full estoppel cost. Let it decline rather than forcing consensus.

Joinder under § 315(c)

Section 315(c) permits the Director to join a party to an instituted proceeding.

Why a later defendant uses it: a defendant sued after the petition was filed — or one that was time-barred by § 315(b) — can in some circumstances join an existing proceeding and obtain the benefit of a decision it could not have obtained alone.

The mechanics: a joinder motion filed within one month of the institution decision in the proceeding to be joined, accompanied by its own petition. The joining party typically takes an "understudy" role — accepting the existing schedule and record, and not adding grounds.

The estoppel consequence. A joined party that reaches a final written decision is estopped like any other petitioner. A defendant joining primarily to obtain a stay in its own district court case should understand it is buying estoppel as well.

The practice has been contested, including whether a time-barred party may join and whether a party may join its own earlier petition to add grounds. Check the current state before relying on it.

PART ELEVEN — THE ONE-PAGE VERSION

Petitioner, month one: run the nine-question framework and write it down. Separate printed-publication art from system art. Do the real party in interest analysis as a discrete workstream and err toward disclosure. Engage the expert.

Petitioner, month two to three: draft a tight petition — few grounds, fully developed, with a motivation-to-combine section that does real work. Address discretion affirmatively. Offer the broad stipulation in the petition.

Petitioner, immediately after filing: move for a stay, or prepare to move the week institution issues. Lead with the stipulation.

Patent owner, within three months: file the preliminary response. Lead with the bars and discretion, not the merits — the procedural arguments are frequently stronger, and a third of petitions never get instituted.

Patent owner, after institution: attack the motivation to combine, cross-examine the declarant to establish what the references do not teach, and develop secondary considerations with a real nexus. Model intervening rights before moving to amend, and check whether a continuation is the better tool.

Both sides, throughout: one named person reviews every filing in both forums for claim-construction and expert consistency. This is the failure that costs cases.

Both sides, before every decision: check the current discretionary denial guidance. It has moved more than once.

And the framing to hold onto: an IPR is a forum choice with an irreversible price. The price is § 315(e) estoppel, the payoff is a preponderance standard before technical judges in eighteen months, and the variables that determine whether the trade is worth it are almost all improved by filing early.

Related documents

• Inter Partes Review Strategy: Discretionary Denial, Estoppel, and the Parallel District Court Case
• PTAB Post-Grant Checklist: A Practical Checklist
• PTAB Toolkit: Petitions, Patent Owner Responses, and Motions to Amend
• Patent Post-Grant Proceedings: IPR, PGR, and Reexamination at the PTAB
• Overcoming Obviousness Rejections: A Comprehensive Guide to Section 103 Analysis
• Preparing for a Markman Hearing: A Practical Guide
• Patent Eligibility Checklist: A Practical Checklist

This guide is general information, not legal advice, and does not create an attorney-client relationship.