Document type: Guide Practice area: Intellectual Property — Patents Jurisdiction: United States and international Last reviewed: 5 September 2026

Who this is for

This guide is for the person who owns the patent portfolio inside a company — a chief IP counsel, a general counsel who inherited the function, a CFO who has started asking what the line item buys, or an outside lawyer being asked to bring order to a portfolio that grew without a plan.

It assumes the portfolio spans more than one country, which changes the problem from a legal one into an operational one. A single-country portfolio can be run out of a good outside firm's docketing system and the client's attention. A multi-country portfolio cannot. It has too many dates, too many currencies, too many counterparties, and too many decisions whose consequences arrive four years after the decision.

Our example is Cordwain Robotics, a 900-person company outside Pittsburgh that builds autonomous material-handling systems for warehouses and ports. Nadia Oyelaran-Pruitt joined as Chief IP Counsel fourteen months ago. She inherited 61 patent families across eleven jurisdictions, two outside firms with overlapping mandates, an annuity service nobody could produce a contract for, and a portfolio that had absorbed roughly $2.1 million a year for six years without anyone being able to say what it was for. Her CFO, Desmond Achterberg, had begun the conversation with a sentence she has heard from every CFO since: "Explain to me what we get for this."

The steps below are the ones she actually took, in order.

Step 1 — Build the asset register

Before anything else, find out what you own. This sounds like a formality. It is the step that most often produces surprises, and it takes six to ten weeks.

Collect the sources. Outside firm dockets from every firm that has ever handled a case. The annuity service's records. The USPTO's assignment database and Patent Center. Equivalent national office records for each jurisdiction. Prior in-house spreadsheets, including the abandoned ones. Acquisition documents and IP schedules. Employment and consulting agreements with assignment provisions. Joint development agreements, which frequently contain ownership terms nobody has looked at since signature.

Build one row per case, not one per family, with family grouping as an attribute. For each case record: family identifier; title; jurisdiction; application and publication numbers; patent number if granted; filing date; priority date; status; the responsible outside firm and foreign associate; the responsible annuity payer; next action and next action date; annual cost; recorded assignee; and the product or program it relates to.

That last field is the one that matters and the one nobody has. A patent that cannot be mapped to a product, a program, a licensing position, or a defensive purpose is a patent nobody can justify. Oyelaran-Pruitt could map 38 of 61 families on the first pass. The remaining 23 became the agenda.

Reconcile ownership independently. Confirm that recorded assignments exist and are recorded in every jurisdiction that requires recordation. Chain-of-title gaps are common — an inventor who left before signing, an entity that was renamed, an acquisition where assignments were made to a subsidiary that has since been merged away. These are cheap to fix while everyone is findable and expensive to fix during a transaction. Cordwain found four families with a missing inventor assignment from a 2019 acquisition and two granted patents recorded to an entity that no longer existed.

Deliverable: a single register, owned in-house, that is the authoritative record. Not the outside firm's docket, which covers only that firm's cases. Not the annuity service's list, which covers only cases it has been instructed on. Yours.

Step 2 — Set the jurisdiction strategy before you set anything else

Every later decision is downstream of one question: in which countries does a patent do this company any good?

Answer it once, deliberately, at the portfolio level, and then apply it case by case. The alternative — deciding country by country under deadline pressure, case by case, with a different person deciding each time — produces the portfolio Cordwain had: eleven jurisdictions, no pattern, and several countries represented by exactly one family that someone once felt strongly about.

Build the analysis on four inputs.

Revenue. Actual and forecast, by country, for the products the patents cover. Get this from finance, in writing. Cordwain's revenue was 54% United States, 21% Germany and the Netherlands, 11% Japan, 6% South Korea, and the remainder scattered.

Manufacturing. Where are the products made — yours and your competitors'? A patent in the country of manufacture stops the product before it ships. Cordwain's principal competitor manufactures in China and assembles in Poland. That single fact justified a Chinese filing program the company had never had.

Enforceability. Litigation cost, timelines, injunction availability, and the practical experience of companies in your industry. This changes; refresh it every few years rather than relying on what was true when you learned it.

Non-enforcement value. Customer procurement requirements, government tender qualifications, investor and acquirer diligence, licensing leverage, and the deterrent value of a visible portfolio. These are real and they are frequently the actual reason for a filing. Say so explicitly rather than pretending the case is about enforcement.

Produce a tiered list.

  • Tier 1 — file everywhere: the jurisdictions that get every family. For Cordwain: United States, EPO, China.
  • Tier 2 — file on the significant cases: Japan, South Korea.
  • Tier 3 — file by exception, with a written reason: everything else.

Then write it down and circulate it. A one-page jurisdiction strategy signed off by the CFO and the head of engineering converts every future filing conversation from a debate into an application of policy. Achterberg approved it in twenty minutes, which is what happens when a budget request arrives with a rationale attached.

Step 3 — Build invention capture that engineers will actually use

A portfolio strategy is worthless if the inventions never reach the lawyers. Most capture processes fail for the same reason: they ask busy engineers to fill out a long form for an uncertain benefit, and engineers rationally decline.

Make the form short. One page. What is the problem, what is the solution, what is different from what existed, who contributed, when was it first described or shown outside the company, and is it in a product. If your form is longer than one page, you are optimizing for the patent attorney's convenience at the cost of getting any submissions at all.

Put the trigger where the work happens. A field in the engineering ticket system. A standing agenda item at design reviews. A quarterly thirty-minute session with each team where a patent attorney asks what they have been working on — which produces more disclosures than any form, because engineers describe inventions well in conversation and badly in writing.

Close the loop fast. Every disclosure gets a decision within thirty days: file, hold, or decline, with a reason. An engineer who submits a disclosure and hears nothing for eight months never submits another one. Cordwain's disclosure rate tripled in a year on the strength of this single change.

Recognize contribution. A modest award at filing and at grant, publicly announced. It is not the money; it is that the company noticed.

Capture the disclosure date. The one-page form's most important field is the question about public disclosure, because absolute novelty regimes abroad make a pre-filing disclosure fatal and the engineer is the only person who knows about the conference abstract submitted last month. Ask the question in the form, and ask it again in conversation.

Step 4 — Set the filing policy and the decision gates

Decide in advance what gets filed and who decides. A written policy with three gates.

Gate 1 — file a priority application? Criteria: is it in a product or roadmap; is it detectable in a competitor's product; is it defensible; would we notice if someone else used it; is the alternative trade secret protection. Decided by IP counsel with input from the business unit. Bias toward filing something cheap and early, because the priority date is unrecoverable.

Gate 2 — file the PCT or Paris applications at twelve months? This is a real decision with real cost, and it is the gate most companies skip by default. Criteria: has the business case held up; has the technology changed; what did the prior art search show; is the product shipping. Decided by IP counsel plus the business unit leader.

Gate 3 — enter the national phase at thirty months, and where? The largest single spend in the family's life. Criteria: the jurisdiction strategy, plus case-specific factors. Decided by IP counsel plus the business unit leader plus, above a threshold, the CFO.

Set an internal deadline sixty days before every statutory deadline. Not as a suggestion — as the operative date in the system. The statutory date is where the right dies; the internal date is where the decision gets made. Cordwain had missed one twelve-month Paris deadline in 2023 on a family the business later described as strategically important. That is the failure this rule prevents.

Write down what happens when a gate produces "no." A declined disclosure should get a defensive publication decision, or a trade secret designation with the corresponding handling requirements, or an explicit note that neither applies. "No" should not mean "nothing happens," because nothing happening means an invention with no protection and no record of the decision.

Step 5 — Run the priority year properly

The twelve months between the priority filing and the Paris deadline under 35 U.S.C. § 119 are the most productive period in a family's life, and most companies waste them.

Thicken the disclosure. File a second, and if warranted a third, priority application as the invention develops. Ranges, alternative embodiments, additional configurations, the things the engineers learned in months three through nine. Each subsequent filing adds support at its own date, and the eventual PCT can claim priority from all of them. This is the cheapest scope you will ever buy.

Write for foreign offices from the start. Explicit ranges rather than single values. Explicit statements that features may be combined. Explicit alternatives. The European added-matter rule means you cannot create basis later; describe it now, including combinations you do not currently want. This is a drafting instruction to give your outside firm in writing, because a firm drafting for United States practice alone will not do it by default.

Run a real prior art search if you did not before filing. Month six is a good time. It is cheap relative to the national phase spend it may prevent.

Confirm the foreign filing license. For any invention made in the United States, verify the license appears on the filing receipt before any foreign filing. Under 35 U.S.C. § 184 and 35 U.S.C. § 185, filing abroad without one can invalidate the United States patent. The rules are at 37 C.F.R. Part 5. This takes ninety seconds and prevents an unfixable defect.

Revisit the business case at month nine. Products get cancelled. Markets change. A family that made sense in month zero sometimes should not be carried into the PCT, and month nine is when you can still make that decision cheaply.

Step 6 — The PCT phase and what to do with it

Most companies file a PCT application under 35 U.S.C. § 361 and then ignore it for eighteen months. That is a mistake, because two things happen in that window that should drive decisions.

Choose the International Searching Authority deliberately. United States applicants often have a choice. Different authorities differ in fee, search quality, literature coverage, and how their work is treated downstream. Where European rights matter most, an ISA whose search the EPO will adopt can eliminate a round of European prosecution. Ask your foreign associates which authority they prefer for your technology, and record the answer as policy rather than re-deciding each time.

Read the International Search Report and Written Opinion when it arrives. Around month sixteen to eighteen you receive what amounts to a free examination-quality prior art assessment, eighteen months before the national phase money is spent. Treat it as a decision document. Three outcomes:

  • Clean opinion: proceed with confidence; consider accelerated examination downstream.
  • Fixable objections: consider amendments under Article 19 or a Chapter II demand, and adjust national phase expectations.
  • Serious prior art: this is the moment to narrow, restructure, or abandon — before spending six figures.

Cordwain abandoned two families on the strength of the ISR in Oyelaran-Pruitt's first year. That decision alone saved roughly $180,000 in national phase and prosecution cost, and it is the single easiest saving available to a company that is not currently reading these reports.

Decide about the Chapter II demand on the merits. Worth it when the Written Opinion is negative and a well-supported amendment could turn it, because a positive International Preliminary Report on Patentability smooths downstream prosecution and can qualify the case for acceleration. Usually not worth it when the opinion is already positive or you plan only one or two national phases.

Publication happens at eighteen months under the PCT and, for the United States application, 35 U.S.C. § 122. Tell the business. Competitors read publications, and a product launch and a patent publication that surprise each other cause avoidable problems.

Step 7 — National phase entry: the pruning conversation

At thirty months from priority, the family enters the national or regional phase under 35 U.S.C. § 371 or its foreign equivalents, and the cost per family multiplies. This is the conversation the jurisdiction strategy exists to make easy.

Present each family with a one-page decision memo: what the invention is, what product it covers, what the ISR said, what the recommended jurisdictions are under the strategy, what the deviation is if any and why, the entry cost, and the ten-year total cost of ownership.

That last number is the one that changes behavior. Entry cost is $6,000 per country; nobody blinks. Ten-year cost of ownership including prosecution and annuities is $60,000 to $90,000 per country; that gets attention. Present the number people are actually approving.

Consider the bypass continuation for the United States. Rather than entering the national stage under § 371, a United States applicant may file a continuation under 35 U.S.C. § 120 claiming benefit of the PCT. The bypass permits freer preliminary amendment and lets the claims be restructured for United States practice. Ask your firm which it uses and why; the default answer of "we always do it this way" is worth examining once.

Budget translation as a legal cost. In Japan, China, and Korea, translation is the dominant entry expense and the translated text is generally the operative text. A translation error is a claim scope error. Pay for a technically qualified translator and have the foreign associate review against the original — a step Cordwain had never funded, and which caught a mistranslation in its second year that would have limited a claim to a single actuator where the original said "at least one."

Say no in writing. For each jurisdiction not entered, record the decision and the reason. Three years later, when someone asks why there is no Brazilian patent, the answer should be in the file rather than reconstructed.

Step 8 — Manage the foreign associate network

Foreign prosecution is done by foreign lawyers, and the quality of that network is the quality of the portfolio.

Consolidate. Cordwain had seven associates in Europe alone, inherited from three different United States firms over a decade. Consolidating to one primary associate per jurisdiction produced better rates, more consistent claim strategy, and a relationship in which the associate actually knew the technology.

Instruct in substance, not in form. The common failure is a United States firm forwarding an office action to a foreign associate with "please advise" and forwarding the response back to the client with "associate recommends." Nobody is making a strategic decision. Give associates the business context: what the product is, what claim scope actually matters, what a competitor is doing, and what you will accept. An associate who understands that the manifold geometry is the commercially important feature will fight for it differently than one who is optimizing for allowance.

Set claim strategy centrally. Claim sets should not diverge across jurisdictions by accident. Deliberate divergence — a narrower European set that clears a prior art reference, a Chinese utility model in parallel — is fine and often correct. Accidental divergence, where each associate amends independently and nobody holds the whole picture, produces a portfolio whose coverage nobody can describe.

Review invoices. Foreign associate billing conventions vary and are not always transparent. Ask for fee estimates in advance for defined tasks, and compare actuals annually. This is not distrust; it is the same discipline applied to any vendor.

Meet them. A twenty-minute video call once a year with each primary associate, in which the technology and the business are explained, changes the quality of the work more than any process change.

Step 9 — Budget and annuities

Build a five-year rolling forecast, not an annual budget. Patent costs are lumpy and back-loaded, and an annual budget hides the wave of annuities arriving in years eight through twelve.

Model four components separately: new filings (a function of your disclosure rate and filing policy); PCT and national phase entry (a function of families filed two and a half years ago — this is forecastable and almost nobody forecasts it); prosecution (a function of pending cases); and annuities (a function of everything you have ever filed and not abandoned).

Annuities deserve specific management. In most jurisdictions they begin before grant, continue for the patent's life, escalate in later years, and abandon the patent if missed. Three rules:

  1. One responsible party per case, recorded in the register. Not "the firm" and "the service" — a named entity.
  2. An annual reconciliation in which the register, the outside firm's docket, and the annuity service's list are compared case by case, and every discrepancy is resolved. This is a day of work and it is the single highest-value day in the calendar.
  3. Annuity payment instructions flow from the portfolio review, not the other way around. The default of paying everything forever is how portfolios become expensive without becoming valuable.

Show the CFO the forecast, not the invoice. Achterberg's original objection was not the amount; it was the unpredictability. A five-year forecast with a stated rationale converted the patent budget from a mystery into a plan, and the plan survived a cost-reduction cycle that cut other line items.

Step 10 — Run an annual portfolio review with real authority

Once a year, every family gets looked at by someone with the authority to abandon it.

The review question is not "is this a valid patent." It is: would we acquire this asset today, at today's remaining cost of ownership? Most portfolios contain a meaningful percentage of cases where the honest answer is no, and the reason they persist is that nobody has ever been asked.

Prepare a one-page sheet per family: the invention in two sentences; the product or program; the jurisdictions and status; annual and remaining cost; whether a competitor practices it; whether it appears in any licensing or enforcement position; and the recommendation.

Score against four uses. A family is worth carrying if it does at least one of these:

  • Exclusivity. It covers something in your product that a competitor would otherwise copy, in a country where that matters.
  • Leverage. It reads on a competitor's product and is useful in a cross-license, a licensing discussion, or a counterclaim.
  • Deterrence. Its existence and visibility shapes competitor behavior.
  • Transaction value. It materially affects diligence in a financing, acquisition, or major customer relationship.

A family serving none of the four is a candidate for abandonment, jurisdiction reduction, or sale.

Prune with a scalpel, not by jurisdiction sweep. The right move is frequently not "abandon the family" but "abandon it in five of eight countries." Keep the Tier 1 jurisdictions and drop the ones that were filed by inertia.

Consider the alternatives to abandonment. Sale to a broker or an operating company. Donation, where the economics work. Dedication to the public, which occasionally has business value. Or a decision to let it lapse quietly, which is usually the answer.

Cordwain's first review abandoned 14 of 61 families entirely, reduced jurisdictions on 9 more, and identified 6 that had been undervalued and deserved continuation filings. Net annual saving: about $340,000. Net effect on protection of the actual product line: none. Achterberg's reaction was not gratitude for the saving; it was relief that someone could now explain the portfolio in a sentence.

Step 11 — Actually use the portfolio

Portfolios that are only maintained decay into cost centers. Portfolios that are used generate returns, and using one is a set of deliberate practices.

Keep continuations pending on the important families. A pending continuation is an option to write claims that read on what a competitor actually built, informed by three more years of market information. In the United States this is the most underused tool in patent practice. It costs relatively little to keep one child pending in a strategically important family, and it converts a static patent into a live instrument. Foreign systems offer divisionals with their own deadlines and rules — coordinate the strategy across the family rather than treating the United States continuation practice as the whole picture.

Map the portfolio to competitor products. Once a year, take the two or three most significant competitors and ask which of your claims read on what they sell. This is the analysis that turns a portfolio into leverage, and it is the analysis that produces a licensing conversation rather than an enforcement action. Do it before you need it.

Prepare the diligence package in advance. Every financing, acquisition, and major customer relationship generates an IP diligence request. A company that can produce a clean asset register, recorded assignments, a jurisdiction rationale, and a maintained annuity record answers in a week. A company that cannot spends six weeks and creates the impression that its IP function is disorganized — which colors everything else in the transaction.

Decide what enforcement means before you need to enforce. Enforcement is a business decision with a legal component. Know, in advance, which families you would assert, in which forums, against whom, and what it would cost. A portfolio held for enforcement that has never been assessed for assertability is a portfolio held on faith.

Publish defensively where filing is not warranted. A technical disclosure published in a recognized venue creates prior art against competitors at trivial cost. For the improvements that do not justify a filing, this is the right answer, and it converts a "no" at Gate 1 into an action rather than a shrug.

Step 12 — Look outward: freedom to operate

A patent portfolio protects what you make. It does not tell you whether you are free to make it. These are different analyses and confusing them is a recurring source of trouble.

Run a freedom-to-operate analysis at product-definition time, not at launch. The cost of designing around a blocking claim is measured in engineering hours before the design freezes and in millions after. For Cordwain, the trigger is the hardware design review at which the mechanical architecture is fixed.

Scope it honestly. A full FTO across all major jurisdictions for a complex product is expensive and never complete. The useful version is targeted: the specific functional areas where the design is novel, the specific competitors most likely to hold blocking rights, and the specific jurisdictions where you manufacture and sell.

Handle the results carefully. Written analyses of third-party patents raise privilege and willfulness considerations that vary by jurisdiction and that should be managed with counsel from the outset. This is a place to get the process right before the first search runs.

Watch competitor filings. A standing watch on the principal competitors' published applications in your technology classes is inexpensive and produces two things: early warning of a blocking position while their claims are still being prosecuted, and intelligence about where they are investing. In Europe, it also produces the input for the decision that matters most: whether to file an opposition within nine months of grant, which is dramatically cheaper than national revocation actions and can eliminate the patent everywhere at once.

Step 13 — Integrate an acquired portfolio

Acquisitions are where portfolios acquire their worst problems, because the acquired cases arrive with dates that nobody in the acquiring company knows about.

Within thirty days of closing: obtain the complete docket from the target's firms; identify every deadline in the next twelve months; confirm who is currently responsible for each; and confirm annuity instructions are in force. Deadlines fall in the gap between the target's disengaged firm and the acquirer's not-yet-engaged firm more often than anywhere else in this practice.

Within ninety days: record assignments in every jurisdiction that requires recordation. Chain of title from an acquisition is the most common defect found in the next transaction's diligence, and it is much easier to fix now.

Within one hundred eighty days: run the acquired families through your own jurisdiction strategy and portfolio review criteria. The target filed on its own logic, which may have been a fundraising logic rather than a protection logic. Expect to prune substantially.

Watch for the specific traps. Joint development agreements with ownership or license terms that survive the acquisition. Government funding with march-in or license-back obligations. Inventor assignments never executed. Encumbrances from prior financings. Licenses granted to customers that constrain your enforcement options. Each of these should have been found in diligence; each is regularly found afterward.

Step 14 — Survive a firm transition

Changing outside counsel, or changing annuity providers, is the second-most common way established portfolios lose cases.

Do not transfer by spreadsheet. Transfer case by case, with an acknowledgment from the receiving firm that it has assumed responsibility for each specific case and each specific next deadline.

Keep both firms on the docket during the overlap. Sixty to ninety days of redundant docketing costs very little and covers the gap where a deadline falls between systems.

Reconcile after. Thirty days after the transition completes, run the full reconciliation against your own register. Find the cases nobody picked up before a deadline finds them for you.

Own the register in-house. The single structural protection against transition loss is that the authoritative record lives with the company, not with any firm. Every firm relationship ends eventually; the portfolio does not.

Step 15 — Measure the right things

Do not measure patent count. It is the metric that most reliably corrupts the practice, because it rewards filing on inventions that do not merit filing and maintaining cases that should lapse.

Measure instead:

  • Coverage: what percentage of current product revenue is covered by at least one granted claim in the countries where that revenue is earned. This is the number that answers "what does the portfolio protect."
  • Disclosure rate and cycle time: disclosures per engineer per year, and days from disclosure to decision. These measure whether the capture process works.
  • Cost per protected revenue dollar, trended. This is the CFO's metric and it is a fair one.
  • Portfolio utilization: how many families appear in a licensing position, an enforcement position, a diligence package, or a competitor-mapping analysis. Families appearing in none are candidates for the next review.
  • Deadline integrity: missed deadlines, near misses, and reconciliation discrepancies. Target zero, measure honestly, and treat a near miss as a process failure rather than an individual one.

Step 16 — Handle the special cases the policy does not cover

Every portfolio contains families that do not fit the standard workflow. Decide how to handle them in advance rather than improvising under deadline.

Joint inventions and joint ownership. Where an invention arises from a collaboration, ownership follows inventorship unless an agreement says otherwise, and joint ownership in the United States is an unpleasant default: absent agreement, each co-owner may exploit and license the whole patent without accounting to the other, and all co-owners are generally necessary parties to an enforcement action. Foreign rules differ, sometimes sharply — several jurisdictions require co-owner consent to license. The fix is contractual and it belongs in the collaboration agreement, before the invention exists: who owns what, who prosecutes, who pays, who may license, and who may enforce.

Inventions made abroad. An invention made in Germany by a German employee is subject to German employee-invention law, which imposes notification obligations, compensation entitlements, and deadlines that have no United States analogue and that are not waivable by contract. Companies with foreign engineering teams routinely discover this late. Get local employment advice for every country where you have inventors, once, and build the resulting obligations into the capture process.

Government funding. Federally funded inventions carry disclosure obligations, election-of-title deadlines, government license rights, and United States manufacturing preferences. Missing an election deadline can forfeit title. If any research is federally funded, the compliance calendar is part of the patent calendar.

Standards participation. If your engineers participate in standards bodies, the organization's IPR policy may commit your patents to licensing on particular terms. Know which employees participate in which bodies, and know what the applicable policy commits you to, before an engineer contributes a proposal that reads on your core claims.

Open source. Contributions under licenses containing patent grants, and use of code under such licenses, can affect assertability. This is an engineering-practice question with patent consequences, and the two functions frequently do not talk to each other.

Step 17 — Decide when a patent is the wrong tool

Some of the best portfolio management decisions are decisions not to have a patent.

Trade secret for undetectable processes. A manufacturing method practiced behind a factory door, whose use cannot be determined from the resulting product, is often better held as a secret. A patent publishes the method to the world at eighteen months under 35 U.S.C. § 122 and buys an enforcement right you cannot exercise because you cannot detect infringement. The test is detectability. Many sophisticated manufacturers patent the apparatus and hold the process.

Trade secret protection has its own requirements — reasonable measures to maintain secrecy, which means access controls, agreements, and documentation — and it fails against independent development and reverse engineering. It is not free, and treating it as the default for anything not filed is how companies end up with neither protection.

Speed-based advantage. Where the product cycle is eighteen months and the patent takes four years, the patent may protect a generation nobody sells anymore. Defensive publication plus execution speed is sometimes the honest strategy.

Design rights for appearance. Where the value is in how the product looks rather than how it works, design protection is faster and cheaper, and the Hague system provides centralized international filing. The Paris priority period for designs is six months, not twelve — a difference that has cost more than one company its foreign design rights.

Write these alternatives into the Gate 1 criteria so that "decline to file" is a routed decision rather than a dead end.

Where Cordwain ended up

Fourteen months in, Cordwain Robotics has 47 active families rather than 61, in six jurisdictions rather than eleven, at an annual run rate of about $1.7 million rather than $2.1 million — with a filing program in China it never had, a continuation strategy on its six most important families, and a five-year forecast that survived a budget cycle.

Wenjun Sato-Klein, who runs engineering, described the change in the way engineers do: "I used to hear about patents when someone needed a signature. Now I hear about them when we're deciding what to build." That is the actual objective. A patent portfolio is not a collection of documents; it is a set of decisions about where a company is willing to be copied and where it is not.

Oyelaran-Pruitt keeps one page on the wall behind her desk. It has the jurisdiction tiers, the three gates, the five-year forecast, and the January reconciliation date. Everything else in this guide is implementation of that page.

Related documents

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