{"id":39528,"date":"2026-09-01T10:14:00","date_gmt":"2026-09-01T14:14:00","guid":{"rendered":"https:\/\/www.drugpatentwatch.com\/blog\/?p=39528"},"modified":"2026-08-24T22:03:20","modified_gmt":"2026-08-25T02:03:20","slug":"the-free-patent-database-costing-your-team-thousands-an-hour-in-manual-cross-referencing","status":"publish","type":"post","link":"https:\/\/www.drugpatentwatch.com\/blog\/the-free-patent-database-costing-your-team-thousands-an-hour-in-manual-cross-referencing\/","title":{"rendered":"The Free Patent Database Costing Your Team Thousands an Hour in Manual Cross-Referencing"},"content":{"rendered":"\n
\"\"<\/figure>\n\n\n\n

The FDA Orange Book does not link to the USPTO’s patent database. The USPTO’s patent database does not link to PACER. PACER does not link to either. Three federal systems hold three pieces of the same fact pattern for any given drug patent, and none of them talk to the others.\u00a0<\/a><\/p>\n\n\n\n

That gap is not a minor inconvenience. It is a recurring line item. Every freedom-to-operate memo, every loss-of-exclusivity forecast, every Paragraph IV risk assessment starts the same way: a person opens the Orange Book, copies a patent number, pastes it into the USPTO’s Patent Public Search, reads the claims, checks the assignment record, then opens PACER to see whether anyone has sued over it yet. Multiply that by every patent on every drug in a portfolio, and the “free” tools stop being free. They just move the cost from a subscription fee to a headcount line.<\/p>\n\n\n\n

The Short Answer<\/strong><\/h2>\n\n\n\n

No single free government database contains a drug patent’s full legal status. The Orange Book lists patents but does not verify or link them to full patent text.To be listable, a patent’s claims must recite the active ingredient of the approved drug, and the Federal Circuit only settled that rule in December 2024<\/a>, after a five-year gap in which thousands of listings sat unchallenged. <\/a> USPTO’s Patent Public Search holds the full text but has no concept of “Orange Book listed.”Basic Search allows only two search terms joined by one Boolean operator<\/a>, which is why the USPTO’s own guidance tells searchers to plan on spending hours learning the search process, searching, and evaluating results<\/a>. PACER holds the litigation record but charges by the page and returns no results linked to either of the other two.A search that returns zero matches still costs money.<\/a> Reconciling the three by hand is a real, calculable labor cost, not a hypothetical one.<\/p>\n\n\n\n

Four “Free” Databases, Zero Integration<\/strong><\/h2>\n\n\n\n

Pharmaceutical patent intelligence work in the United States runs through at least four separate public systems, each maintained by a different federal body, each free to search, and none built to talk to the others.<\/p>\n\n\n\n

The Orange Book Tells You What’s Listed, Not What’s True<\/strong><\/h3>\n\n\n\n

The Orange Book, formally Approved Drug Products with Therapeutic Equivalence Evaluations<\/em>, is the FDA’s register of approved drugs and the patents their sponsors claim cover them. <\/a> A New Drug Application holder must submit patents that claim the drug substance, the drug product, or an approved method of use.Process patents, packaging patents, and patents on metabolites or intermediates are not eligible for listing<\/a> and, by regulation, should never appear there. <\/a><\/p>\n\n\n\n

The important detail is who checks that rule. FDA does not.The agency’s role in receiving patent submissions is described as ministerial, and if a sponsor submits a patent for listing, it goes into the Orange Book as a matter of course.<\/a> Nothing in the intake process confirms that a listed patent’s claims actually cover the approved drug. That verification only happens later, and usually only when someone with a financial incentive to challenge the listing does the work.<\/p>\n\n\n\n

USPTO Patent Public Search Tells You What’s Patented, Not What’s Listed<\/strong><\/h3>\n\n\n\n

Patent Public Search is the USPTO’s own full-text search platform, combining what used to be four separate legacy tools. <\/a> It covers granted patents back to 1790 and published applications back to 2001, <\/a> but full-text keyword search only works reliably for patents from 1970 forward; older patents are searchable mainly by number, issue date, and classification. <\/a> Nothing in the platform records whether a given patent is currently listed against a drug in the Orange Book, delisted, or was ever listed at all. That information exists only in the FDA’s separate system.<\/p>\n\n\n\n

The Purple Book Tells You Almost Nothing Unless Someone Has Already Sued<\/strong><\/h3>\n\n\n\n

Biologics are worse off. The Purple Book, FDA’s register for biological products, does not include patent information as a standing feature. <\/a> Patent disclosure for biosimilars runs through the BPCIA “patent dance,” a private negotiation between the reference product sponsor and a biosimilar applicant.A patent list only appears in the Purple Book after a biosimilar applicant has filed an application and the patent dance has begun. If no one has filed against a given biologic, the Purple Book carries no patent information for it beyond exclusivity dates.<\/a> A team researching a biologic with no current biosimilar filer is, for patent purposes, working from an empty page.<\/p>\n\n\n\n

PACER Tells You What’s Litigated, One Page at a Time<\/strong><\/h3>\n\n\n\n

Patent litigation dockets live in PACER, the federal judiciary’s electronic records system, priced at $0.10 per page as of 2026.The charge applies to every page a search returns, including a search that finds nothing.<\/a>A ten-page docket report costs a dollar; a five-page document costs fifty cents, capped at $3.00 per document.<\/a> PACER has no cross-district full-text index and no connection to either the Orange Book or USPTO’s patent records.Reviewers have described its search function as notoriously limited, with no faceted filtering and no way to search full text across districts at once.<\/a> Finding a Paragraph IV suit means already knowing, or guessing, which district and which case number to search.<\/p>\n\n\n\n

The Manual Cross-Reference Workflow, Step by Step<\/strong><\/h2>\n\n\n\n

Here is what actually happens when someone builds a patent profile for a single drug using only the free tools, in the order the work has to happen.<\/p>\n\n\n\n

Step One: Pull the Orange Book Listing<\/strong><\/h3>\n\n\n\n

The Orange Book can be searched by active ingredient, proprietary name, applicant, or application number. <\/a> Each result lists the patent number, patent type, and an expiration date. That date is the first trap. Orange Book expiration dates frequently exclude patent term extensions that have been applied for but not yet updated in the database, so the listed date has to be treated as preliminary, not final.<\/a><\/p>\n\n\n\n

Step Two: Translate the Patent Number Into a USPTO Record<\/strong><\/h3>\n\n\n\n

Every patent listed in the Orange Book exists as a separate public document at the USPTO, and confirming its actual issue date, priority date, claim scope, and assignment history means pulling the full text directly from Patent Public Search.<\/a> This is a manual, one-at-a-time lookup. Patent Public Search’s Quick Lookup accepts a single patent number formatted to a specific digit count,seven digits minimum for utility patents, with leading zeroes required to pad shorter numbers<\/a> \u2014 a small formatting rule that trips up a first-time or occasional user on every single query.<\/p>\n\n\n\n

Step Three: Recover What the Orange Book Left Out<\/strong><\/h3>\n\n\n\n

Because Orange Book dates are unreliable for patent term extension, the analyst has to independently check whether a PTE was filed and granted, since a PTE can add years to the term shown in the initial listing. <\/a> There is no field in either database that flags “PTE pending” versus “PTE applied to this date already.” That determination is manual, and it depends on cross-checking the patent’s own file history.<\/p>\n\n\n\n

Step Four: Check Litigation Status in PACER<\/strong><\/h3>\n\n\n\n

Finally, the analyst has to determine whether the patent has been the subject of a Paragraph IV suit, an IPR, or any other challenge. Because PACER has no link to either FDA or USPTO records, this means guessing the likely district, searching by party name or case number, and paying by the page for every docket pulled up.A parallel nationwide case locator search costs $30 per name searched, plus $0.10 per page for each document delivered.<\/a><\/p>\n\n\n\n

One Patent, Four Systems, Zero Automation<\/strong><\/h4>\n\n\n\n

That is the workflow for one patent. A drug with a modest patent estate can carry a dozen or more listings; a heavily protected biologic can carry over a hundred. <\/a> Every one of those patents goes through the same four-step, four-system, zero-automation process, by hand, before an analyst can say with confidence what actually protects the drug and until when.<\/p>\n\n\n\n

What the Orange Book Won’t Tell You That USPTO Search Will<\/strong><\/h2>\n\n\n\n

Claims Language Decides Listability, and the Orange Book Doesn’t Show Claims<\/strong><\/h3>\n\n\n\n

The Orange Book entry for a patent shows a patent number, an expiration date, and a type code. It does not show the claims. Whether a patent is properly listed turns entirely on claims language,specifically whether the claims recite the drug’s active pharmaceutical ingredient<\/a> \u2014 information that exists only in the full patent document at USPTO, not in the FDA’s listing.<\/p>\n\n\n\n

The Teva v. Amneal Test Came From Reading Claims, Not Listings<\/strong><\/h3>\n\n\n\n

In December 2024, the Federal Circuit resolved a years-long dispute over exactly this question. In Teva Branded Pharmaceutical Products R&D, Inc. v. Amneal Pharmaceuticals of New York, LLC, the court held that a patent must recite the active pharmaceutical ingredient of the approved drug in its claims to be listable in the Orange Book.<\/a> The Federal Circuit affirmed delisting of Teva’s ProAir HFA inhaler patents because those patents claimed the inhaler device and its components, not albuterol sulfate itself.<\/a> Nothing about that defect was visible from the Orange Book entry. It only became visible once someone pulled the claims from USPTO and read them against the statute.<\/p>\n\n\n\n

What USPTO Search Won’t Tell You That the Orange Book Will<\/strong><\/h2>\n\n\n\n

Two Search Boxes and a Boolean Operator<\/strong><\/h3>\n\n\n\n

USPTO’s Basic Search interface offers two text boxes and a single Boolean operator to join them.<\/a> Each input field accepts only a single word or number, with a maximum of two keywords usable in one search.<\/a> This is adequate for a quick lookup by known patent number. It is not adequate for discovering which patents cover a given drug’s formulation or method of use, since that requires either already knowing the patent numbers \u2014 meaning the Orange Book has to be consulted first \u2014 or running iterative keyword searches with no guarantee of completeness.<\/p>\n\n\n\n

Stopwords Silently Drop Search Terms<\/strong><\/h3>\n\n\n\n

Patent Public Search treats common words as stopwords that are not indexed in the body text of a document. When a stopword appears in a search query, the system searches it only in limited metadata fields, and returns no results if the term appears in the title, abstract, or claims text itself.<\/a> A searcher who does not already know which terms are silently excluded has no way to know their search missed relevant claims language.<\/p>\n\n\n\n

“Plan on Spending Hours” Is the Official Guidance<\/strong><\/h4>\n\n\n\n

The USPTO’s own seven-step search strategy tells users to run a keyword search, review front pages, conduct an in-depth review of the full documents, expand the search using classification codes, review cited references forward and backward, and then broaden the search to foreign patents and non-patent literature.The agency’s own guidance closes with: “Plan on spending hours learning the search process, searching, and evaluating results.”<\/a> That is USPTO describing its own free tool, for a single invention, before any Orange Book cross-reference is even attempted.<\/p>\n\n\n\n

The Blind Spot Both Databases Share<\/strong><\/h2>\n\n\n\n

9.4 Percent of Litigated Drug Patents Were Never Listed at All<\/strong><\/h3>\n\n\n\n

Even a perfect manual cross-reference between the Orange Book and USPTO misses a measurable share of the real patent landscape. Researchers who built the NBER Orange Book Dataset compared litigated pharmaceutical patents against Orange Book listings going back to 2003.Of 2,286 unique patents litigated in brand-generic disputes, 214 \u2014 9.4 percent \u2014 were never listed in the Orange Book in any edition through 2021.<\/a> A brand manufacturer is not limited to asserting only listed patents; it can and does assert manufacturing or ancillary patents outside the Hatch-Waxman framework entirely. <\/a> A search strategy built solely around Orange Book listings will never surface that share of the real risk, because the risk was never in the database to begin with.<\/p>\n\n\n\n

When the Databases Disagree: Real Cases Where a Listing Became a Lawsuit<\/strong><\/h2>\n\n\n\n

The FDA’s ministerial, non-verifying approach to Orange Book listings has produced a specific, recurring category of litigation: cases where a listing sat unchallenged for years because nobody with the time to read the underlying claims happened to do so before a generic filer did.<\/p>\n\n\n\n

In re Lantus: An Injector Pen Patent That Wasn’t a Drug Patent<\/strong><\/h3>\n\n\n\n

In In re Lantus Direct Purchaser Antitrust Litigation, the First Circuit found that a device patent covering Sanofi’s SoloSTAR injector pen drive mechanism had been improperly listed, because the patent did not claim insulin glargine or the Lantus SoloSTAR product itself.<\/a> The defect was structural, not factual \u2014 the patent covered a mechanical component, not the drug \u2014 and it took a federal antitrust suit to establish that formally.<\/p>\n\n\n\n

UFCW v. Takeda: Combination Patents Face the Same Test<\/strong><\/h3>\n\n\n\n

United Food & Commercial Workers Local 1776 v. Takeda Pharmaceutical Co., decided by the Second Circuit in 2021, applied the same listability scrutiny to Actos combination patents.<\/a> Both cases required a court, not either database, to resolve whether the listing matched the statute.<\/p>\n\n\n\n

Teva v. Amneal: Five Patents, One Bright-Line Rule<\/strong><\/h3>\n\n\n\n

The Federal Circuit’s Teva v. Amneal decision covered five ProAir HFA inhaler patents and established that a patent must claim at least the active ingredient \u2014 the primary mode of action of the drug \u2014 to be listable.<\/a> Teva’s petition for en banc rehearing was denied in March 2025, and the five patents had to be delisted.<\/a> That ruling did not exist when most currently listed device patents were originally submitted, which is why its retroactive effect on the broader Orange Book has been so disruptive.<\/p>\n\n\n\n

The FTC’s Three-Round Campaign Shows How Slowly Errors Get Corrected<\/strong><\/h3>\n\n\n\n

The clearest evidence that nobody is systematically checking Orange Book listings against underlying patent claims is the FTC’s own multi-year campaign to force the issue.<\/p>\n\n\n\n

Three Rounds of Warning Letters<\/strong><\/h4>\n\n\n\n

In September 2023, the FTC issued a policy statement putting drug manufacturers on notice that improper Orange Book listings could be scrutinized as an unfair method of competition under Section 5 of the FTC Act.<\/a> On November 7, 2023, the FTC sent its first round of warning letters to ten manufacturers, challenging more than 100 patents covering asthma inhalers, epinephrine autoinjectors, and an ophthalmic drop.<\/a> A second round followed on April 30, 2024, targeting more than 300 additional patents across twenty branded products.<\/a> A third round on May 21, 2025 challenged more than 200 patents across seventeen products that remained listed despite the earlier warnings.<\/a><\/p>\n\n\n\n

Most Challenged Listings Simply Stayed Listed<\/strong><\/h4>\n\n\n\n

Of the ten companies that received the first round of letters, four elected to delist most or all of the challenged patents.<\/a> By August 2024, independent legal analysis found that the overwhelming majority of patents the FTC had declared improperly listed remained listed and in force, with only three companies \u2014 GlaxoSmithKline, Glaxo Group, and Kal\u00e9o \u2014 having voluntarily delisted.<\/a> The FTC’s own account credits its first two rounds with the delisting of patents across 22 different brand-name products<\/a> \u2014 a real number, but a fraction of the 400-plus patents originally challenged. Every one of those still-listed patents sits in the Orange Book today exactly as it did before the FTC wrote its letter, waiting for a generic filer’s legal team to independently pull the claims and decide whether to litigate the point.<\/p>\n\n\n\n

Even the Patent Office Doesn’t Have Time to Search Properly<\/strong><\/h2>\n\n\n\n

Nineteen Hours, Most of It Spent Searching<\/strong><\/h3>\n\n\n\n

The upstream reason all of this manual verification is necessary is that the patents entering these databases were not exhaustively vetted to begin with. Research by Michael Frakes and Melissa Wasserman, published through the National Bureau of Economic Research, found that a USPTO patent examiner spends on average only nineteen hours reviewing an application \u2014 a figure that covers reading the application, searching prior art, comparing prior art to the claims, writing a rejection, and responding to the applicant’s arguments.<\/a><\/p>\n\n\n\n

\n

Prior art searches alone consume nearly half of a patent examiner’s working hours \u2014 46.3 percent of total activity, according to a machine-readable analysis of 560,603 USPTO examiner search records from 2010. <\/a><\/p>\n<\/blockquote>\n\n\n\n

What Nineteen Hours Means for Everyone Downstream<\/strong><\/h3>\n\n\n\n

Frakes and Wasserman’s broader body of research found that doubling examiner time allocations was associated with a 44 percent reduction in the rate at which an issued patent later became the subject of federal patent litigation.<\/a> That correlation matters directly to the cross-referencing problem: a meaningful share of the patents an analyst is manually verifying against Orange Book listings were themselves examined under significant time pressure, using the same fragmented public search tools the analyst is now using to check the examiner’s work after the fact.<\/p>\n\n\n\n

The Fifth Free Database: PACER, Priced by the Page<\/strong><\/h2>\n\n\n\n

Ten Cents a Page, No Cross-District Search<\/strong><\/h3>\n\n\n\n

PACER is free in the sense that no subscription is required, but it is not free to use. The $0.10-per-page charge applies to every search result page and every document viewed, including searches that return no matches at all.<\/a> The PACER Case Locator, the tool used to find which district a case was filed in when the district is unknown, itself costs $30 per name searched.<\/a> A search strategy that has to guess at jurisdiction before it can even confirm whether litigation exists is, by construction, going to generate charges for dead ends.<\/p>\n\n\n\n

The Price Is About to Go Up<\/strong><\/h3>\n\n\n\n

In June 2026, the Judicial Conference’s Executive Committee approved raising the PACER fee from $0.10 to $0.12 per page, effective January 1, 2027, to help fund an estimated $700 million to $800 million modernization of the judiciary’s case management system.<\/a> PACER currently generates about $147 million a year in fee revenue.<\/a> It is the first fee increase since 2012, when the rate rose from $0.08 to $0.10 per page.<\/a> The increase is described as temporary and set for a five-year period, but the underlying search limitations \u2014 no full-text index across districts, no connection to Orange Book or USPTO data \u2014 are not part of the fee change.<\/p>\n\n\n\n

The Time Cost, Calculated<\/strong><\/h2>\n\n\n\n

What a Cross-Reference Analyst Actually Costs Per Hour<\/strong><\/h3>\n\n\n\n

The people doing this work are usually paralegals, IP analysts, or junior associates. The Bureau of Labor Statistics reports a median annual wage of $61,010 for paralegals and legal assistants as of May 2024 \u2014 approximately $29.33 an hour.<\/a> Specialization pushes that figure higher. Salary.com’s May 2026 data puts the average Patent Paralegal salary at $96,533 per year, or $46 an hour, with a typical range of $38 to $56 an hour.<\/a> A blended team \u2014 a mix of paralegal-level and more senior IP-analyst time \u2014 realistically costs an employer somewhere between $30 and $55 an hour in direct labor, before overhead.<\/p>\n\n\n\n

A Labeled Illustrative Model<\/strong><\/h3>\n\n\n\n

The following is an original calculation built from the wage data above and a stated, clearly flagged assumption about time-per-patent. It is not an independently reported industry statistic.<\/p>\n\n\n\n

The Assumptions<\/strong><\/h4>\n\n\n\n

Assume a mid-size branded drug carries 15 Orange Book-listed patents \u2014 a realistic mid-range figure for a drug with more than a single compound patent. Assume each patent requires, at minimum, one Orange Book lookup, one USPTO Quick Lookup and claims read, one PTE\/PTA verification check, and one PACER litigation-status check \u2014 the four-step workflow described above. Based on the documented mechanics of each system (single-number Quick Lookup, two-term Boolean search, per-page PACER billing), a conservative estimate is 25 to 40 minutes of hands-on analyst time per patent to complete all four steps once, before accounting for any follow-up, discrepancy, or a PTE that requires additional file-history research.<\/p>\n\n\n\n

The Arithmetic<\/strong><\/h4>\n\n\n\n

Fifteen patents at 25 to 40 minutes each is 6.25 to 10 hours of analyst time for a single drug’s initial patent profile. At a blended rate of $30 to $46 an hour, that is roughly $190 to $460 in direct labor cost to build one drug’s patent picture \u2014 once. A portfolio team tracking loss-of-exclusivity timing across even a modest 20-drug portfolio, refreshed quarterly as new patents are listed, delisted, or litigated, is repeating some version of that exercise dozens of times a year. This is a labor-cost floor, not a ceiling \u2014 it excludes review time, senior sign-off, and the cost of an error that reaches a board deck or a licensing negotiation.<\/p>\n\n\n\n

DrugPatentWatch, which aggregates and cross-links Orange Book, Purple Book, and patent data into a single searchable system, has separately published a range of four to twelve weeks and $50,000 to $200,000 for a comprehensive external freedom-to-operate search conducted by outside counsel or a specialist firm <\/a> \u2014 a figure that reflects a much deeper analysis than the single-drug patent profile modeled above, but one built on the same underlying cross-referencing problem.<\/p>\n\n\n\n

A Taxonomy of Cross-Referencing Failure Modes<\/strong><\/h2>\n\n\n\n

Four distinct ways the manual process breaks down recur across the cases and mechanics described above.<\/p>\n\n\n\n

Temporal Mismatch<\/strong><\/h3>\n\n\n\n

The Orange Book date shown at the time of lookup is not necessarily the final date. Term extensions applied for but not yet reflected in the database leave the analyst with a preliminary figure that looks final.<\/a> A profile built on the listed date alone will understate the patent’s real life.<\/p>\n\n\n\n

Blind-Spot Omission<\/strong><\/h3>\n\n\n\n

Patents that were never submitted for Orange Book listing \u2014 9.4 percent of litigated pharmaceutical patents in the NBER sample \u2014 never enter the workflow at all, because the workflow starts from the Orange Book.<\/a> A search strategy anchored to one database inherits that database’s blind spots.<\/p>\n\n\n\n

Format Fragmentation<\/strong><\/h3>\n\n\n\n

Each system has its own input syntax: patent numbers padded to specific digit counts in USPTO’s Quick Lookup, a two-keyword limit in Basic Search<\/a>, and page-based billing in PACER that varies depending on whether a document is HTML or PDF<\/a>. Time spent reformatting a query for the next system is time not spent evaluating substance.<\/p>\n\n\n\n

Verification Gap<\/strong><\/h3>\n\n\n\n

The Orange Book’s listing criteria exist in statute and regulation, but FDA does not independently verify that a submitted patent’s claims meet those criteria before listing it.<\/a> The gap between “listed” and “properly listed” is exactly where In re Lantus<\/a>, UFCW v. Takeda<\/a>, and Teva v. Amneal<\/a> all landed, and it is a gap no automated feed of either database closes \u2014 it requires someone to actually read the claims against the statute.<\/p>\n\n\n\n

What This Means for Freedom-to-Operate Teams<\/strong><\/h2>\n\n\n\n

An FTO analysis has to be exhaustive by definition \u2014 the entire point is to find every patent that could block a product, not just the ones already flagged. Every hour spent on database mechanics \u2014 reformatting a patent number, guessing a PACER district, rereading Orange Book dates that turn out to be preliminary \u2014 is an hour not spent on substantive claim analysis, which is the part of the job that actually requires legal judgment. The four-to-twelve-week, $50,000-to-$200,000 range typically quoted for outside FTO work reflects, in large part, the cost of this reconciliation labor, not just the legal analysis layered on top of it.<\/a><\/p>\n\n\n\n

What This Means for Competitive Intelligence and Business Development<\/strong><\/h2>\n\n\n\n

Loss-of-exclusivity timing feeds directly into revenue forecasting, licensing valuations, and M&A diligence. A patent profile built on an unverified Orange Book date, without an independent USPTO check for term extension status, risks feeding a forecast model a date that is wrong in the brand’s favor \u2014 overstating remaining exclusivity \u2014 or wrong in the generic filer’s favor if a term extension was missed. Either error compounds every time the same unverified figure gets copied into the next quarter’s deck.<\/p>\n\n\n\n

Methodology<\/strong><\/h2>\n\n\n\n

This analysis draws on primary regulatory and legal sources: FDA guidance and regulations governing Orange Book listing criteria, USPTO’s own published search guidance and FAQs for Patent Public Search, PACER’s official fee schedule and FAQ pages, the Administrative Office of the U.S. Courts’ 2026 fee-increase announcement, the NBER Orange Book Dataset user’s guide, the Frakes and Wasserman examiner-time research published through NBER, Bureau of Labor Statistics wage data, and primary case law and FTC enforcement documents covering Orange Book listability disputes through mid-2026. The illustrative time-and-cost model in this article is an original calculation clearly built from stated assumptions about per-patent research time; it is not drawn from a third-party time-and-motion study, and readers building their own estimates should adjust the assumptions to their own team’s actual workflow. Case selection favored decisions with a clear, documented listability defect (device or combination patents found not to claim the active ingredient) over disputes still working through appeal as of this writing.<\/p>\n\n\n\n

Key Takeaways<\/strong><\/h2>\n\n\n\n