What Actually Drives a Patent Deal
Vinay Sharma
Principal - Intellectual Asset Management
Iota Analytics
Seven principles that decide licensing and litigation outcomes
Every patent monetization conversation eventually reaches the same question: what is this portfolio actually worth, and how do we realize that value? The answer is never found in the patent grant alone. It is found in the interplay between the portfolio, the product, the evidence, and the economics of the technology stack. After years of building claim charts, negotiating licenses, and preparing litigation campaigns across US and European practice, seven principles keep proving themselves. They fall into three phases: set the strategy, prove the value, and execute right.
Phase One — Set the Strategy
Portfolio sets the playbook
There is no one-size licensing strategy. Portfolio size and technology domain define the playbook before the first term sheet is drafted. The most authoritative benchmark is the LES/LESI Royalty Rates & Deal Terms Survey, compiled from actual member deals rather than public filings. Its High Tech surveys report a median royalty rate that has held near 5% across the 2011–2017 rounds (average 5.66%), easing to a 4.75% median (4.82% average) in the 2021 round — with wide dispersion by sub-sector, aerospace technology averaging as high as 10.7%. The LES Life Sciences survey shows early-stage flat royalties clustering around 5%, rising materially through tiered structures tied to sales milestones and later-stage products. The takeaway is not any single number but the spread: the same patent count earns very different money in different domains, because domains differ in margin structure, design-around difficulty, and the binary nature of protection — a drug compound is either covered or it is not, while an electronic feature usually has substitutes.
Portfolio scale changes the strategy too, but not linearly. Studies of large portfolios show diminishing returns — each patent added increases the achievable rate by less than the one before it. A multi-thousand-family portfolio is monetized through programmatic, rate-based licensing; a dozen-patent medtech or chemical portfolio is monetized through targeted assertion against specific, high-value features. Confusing the two playbooks wastes both.
The product decides the play
Strategy equally depends on the accused or licensed product itself: what type of product it is, where it sits in the value chain, whether it is available in the market, and what compliance and regulatory context surrounds it. A component supplier and an end-device maker present entirely different royalty bases, different apportionment battles, and different negotiation leverage — value-chain position changes everything, whether the field is telecom, automotive, or drug-delivery devices. Market availability determines whether an injunction threat has teeth, and here the jurisdictions diverge sharply: injunctions in US courts pass through the eBay equitable factors, while German courts and now the Unified Patent Court treat the injunction as the ordinary consequence of infringement — which is precisely why value-chain and forum strategy are decided together. Regulatory regimes — FDA and EMA approval for pharmaceuticals, medical-device certification, automotive homologation — change both the cost of design-around and a defendant’s willingness to fight rather than settle.
Phase Two — Prove the Value
Evidence is the leverage
Both litigation and licensing rest on the same foundation: sufficient technical evidence, and the value of the infringing element in the product and the market. The instrument of that evidence is the element-by-element claim chart — every limitation mapped to a specific, pinpoint-cited feature of the accused product. In the US, the chart must survive Rule 11 scrutiny and local patent rules on infringement contentions; in Europe, the same mapping discipline underlies infringement arguments under Article 69 EPC and its Protocol on Interpretation, which govern how claims are construed in every EPC state and before the UPC. A chart that maps to the whole product instead of the specific feature, or that rests on conclusory assertions, collapses in a negotiation room just as surely as it does in Marshall, Munich, or Milan.
Evidence of infringement alone is not enough — the infringed element must matter. An airtight chart on a feature nobody uses, in a product with negligible sales, is a technically correct exercise with no commercial consequence. The strength of a campaign is the product of evidentiary quality and the economic weight of the feature it proves.
Technology base and technology share come first
Before any royalty rate is finalized, two numbers become critical: the aggregate royalty burden the relevant technology stack can bear, and your technology share within it. In the standards world, courts on both sides of the Atlantic already work this way: in TCL v. Ericsson the US court set an aggregate 4G burden in the range of 6–10% of handset price and apportioned it by patent share, and in Unwired Planet v. Huawei the UK courts — ultimately affirmed by the Supreme Court in 2020 — set global FRAND rates informed by aggregate-burden cross-checks. The CJEU’s Huawei v. ZTE framework governs how SEP holders and implementers must negotiate in Europe before an injunction is available.
The logic is not confined to SEPs. A biotech product may carry stacked royalties to the platform-technology owner, the compound patentee, and the delivery-device licensor — a licensee that would owe 20% in aggregate will not pay 8% to any single one of them. A university spin-out licensing a manufacturing process must price against the other process steps the licensee already pays for. Whoever ignores the total burden invites either a stacking objection or a court-set number.
Technology contribution value cannot be ignored
Apportionment is the law in both systems, not paperwork. In the US the principle runs from the Supreme Court’s Garretson v. Clark (1884) through the Federal Circuit’s modern line: Uniloc v. Microsoft struck down the once-popular "25% rule of thumb" as inadmissible precisely because it ignored the patent’s actual contribution, and LaserDynamics v. Quanta requires starting from the smallest salable patent-practicing unit rather than the entire product. In Europe, Article 13 of the Enforcement Directive (2004/48/EC) frames damages around the hypothetical royalty a reasonable licensee would have paid for the patented contribution — the German Lizenzanalogie (license analogy) method applies exactly that logic, with courts adjusting the rate to reflect what the patented teaching adds to the product. Different doctrinal paths, same question: what does your technology actually add? A monetization strategy that cannot answer it with evidence has already conceded the discount.
Make the Claim Chart Strong
Because the chart carries the evidentiary load, its quality deserves its own discipline. From systematic claim-chart review practice, the strongest charts share these traits:
One limitation per row, verbatim. Break each claim into discrete elements and quote the exact claim language — never paraphrase. Decide up front whether the preamble is limiting.
Map to the construed term, not plain English. Construction comes first: Phillips intrinsic-evidence hierarchy in the US; Article 69 EPC and the Protocol before the UPC and national courts.
Pinpoint every citation. Datasheet section and page, standard clause, figure, teardown photo, or source-code file and line range. An assertion without a source location is a gap, not evidence.
Label direct disclosure versus inference. Where the evidence expressly shows the feature, say so; where it is necessarily implied, state the inference chain and what discovery or testing will confirm it. Never dress inference as disclosure.
Keep literal and equivalents theories separate. Plead literal infringement first; run equivalents as a distinct theory — function-way-result with Festo estoppel limits in the US, the Actavis v. Eli Lilly questions in the UK, the Schneidmesser questions in Germany.
Respect claim mechanics. Establish the parent claim before dependents, show method steps in the claimed order, and check for divided-infringement gaps where multiple actors perform the steps.
Score by the weakest link. Under the all-elements rule the chart is only as strong as its weakest essential element — grade element by element, name the limiting element, and state the discovery path that closes each gap.
Keep evidence current and red-team it. Verify the cited documentation matches the accused product version, confront negative evidence honestly, and have someone play defense counsel before the chart leaves the building.
Phase Three — Execute Right
Technical and legal, one team
Close collaboration between the technical and legal teams is a must-have component, not a nice-to-have. The claim chart that survives a Markman hearing in the US, or a construction fight under Article 69 EPC before the UPC, is the one where the engineer’s understanding of how the product actually operates was tested against the lawyer’s understanding of how the claim will be construed — before the chart shipped. The same holds in licensing: the negotiator who can walk a counterpart’s engineers through the evidence, element by element, closes on better terms than one reading from a summary.
No case is perfect, but it has to be technically correct
Every chart has a weakest element, and the case is only as strong as that element. The discipline is not to pretend perfection — it is to find the weakest link, then close it through discovery, testing, or expert work, or price it into the settlement posture. What is never acceptable is shipping something technically wrong: an element mapped to a feature the product does not have, an inference dressed up as direct disclosure, a construed term charted in plain English. Imperfect and honest wins negotiations and trials; polished and wrong loses both, along with credibility for every future campaign.
The Bottom Line
Strong positions are built, not asserted. The portfolio and the product set the strategy; evidence, technology share, and contribution value prove the worth; and a unified technical-legal team executes with rigor. No case is perfect — but it has to be technically correct.
