The Georgia-Pacific Factors

Existing licenses for the same patent are among the most probative evidence — if the patentee has licensed this exact patent at 2%, that rate anchors the hypothetical negotiation. Courts scrutinize whether prior licenses were truly comparable: litigation-settlement licenses are discounted because they reflect litigation risk, not market value. Lump-sum licenses must be converted to effective rates with care.

Evidence: Existing license agreements for the patent in suit; royalty reports; negotiation history.

What the accused infringer pays to license similar technology shows what it considers a fair market rate for technology of this kind. The Federal Circuit requires technological and economic comparability — an expert cannot rely on licenses for unrelated technology or radically different deal structures (Lucent Technologies v. Gateway, Fed. Cir. 2009, vacated a damages award built on non-comparable licenses).

Evidence: Defendant's license portfolio for similar technology; industry license databases.

Industry norms for royalty rates in the relevant technology field. Historically experts invoked a '25% rule of thumb' (25% of expected profits to the licensor) — the Federal Circuit categorically rejected the 25% rule as arbitrary in Uniloc USA v. Microsoft (Fed. Cir. 2011). Industry-specific evidence is still admissible if tied to the facts of the case.

Evidence: Industry licensing surveys; expert testimony grounded in field-specific transactions.

An exclusive license commands a higher rate than a non-exclusive one; an unrestricted license commands more than one limited by field of use or territory. The hypothetical license is typically assumed to be non-exclusive and limited to the accused products, so rates from broader real-world licenses must be adjusted downward.

Evidence: The structure of comparable licenses; scope of the defendant's accused use.

A patentee that refuses to license at all (to preserve exclusivity) or licenses only at premium rates can argue the hypothetical rate should be high — it would only have surrendered exclusivity for a substantial payment. A patentee with a broad, low-rate licensing program faces the opposite inference.

Evidence: Licensing program documents; testimony about refusal to license competitors.

Direct competitors negotiate higher rates than non-competing parties — a license to a head-to-head competitor sacrifices market share, not just exclusivity. If the parties are inventor-promoter, supplier-customer, or operate in different markets, the rate trends lower.

Evidence: Market overlap analysis; competitive sales data; head-to-head bidding history.

If practicing the patent drives sales of the licensee's other (unpatented) products and services — accessories, consumables, service contracts — the royalty rises to reflect that derivative value. Convoyed sales must have a functional relationship to the patented product, not mere bundling convenience.

Evidence: Attach-rate data; sales of complementary products tied to the patented feature.

A highly profitable, commercially successful product supports a higher rate — there is more profit to share in the hypothetical negotiation. Courts examine profit margins on the accused products and whether success is attributable to the patented feature versus marketing, brand, or unpatented features.

Evidence: Profit-and-loss statements for accused products; market share trends.

The royalty reflects the increment of value over the next-best non-infringing alternative. If a design-around is cheap and nearly as good, the rational licensee would pay little; if the patent has no acceptable substitute, the rate climbs. Non-infringing alternatives are one of the most heavily litigated damages issues.

Evidence: Technical comparison with prior art; cost and performance of design-arounds.

The character of the patented invention as embodied in the licensor's own commercial products, and the benefits to those who actually use it. Overlaps with factor 9 — both probe how much real-world value the claimed invention delivers.

Evidence: Product documentation; customer surveys on feature value.

How much the infringer actually used the invention and what that use was worth. A feature that drives the purchasing decision justifies a higher rate than one that is rarely activated. Usage data and customer-demand evidence are central.

Evidence: Feature-usage telemetry; consumer survey evidence; marketing emphasis on the patented feature.

A license running the patent's full remaining term is worth more than a short-term deal. Remaining patent life at the date of first infringement frames the hypothetical license term.

Evidence: Patent expiration date; period of accused infringement.

The most consequential factor in modern practice. Damages must be apportioned to the incremental value of the patented invention — not the value contributed by manufacturing, marketing, unpatented features, or the standardization of the technology. For multi-component products, the royalty base should generally be the smallest salable patent-practicing unit (SSPPU) rather than the entire product, unless the patented feature drives consumer demand for the whole product (the entire market value rule's narrow exception — LaserDynamics v. Quanta, Fed. Cir. 2012).

Evidence: Component pricing; regression or conjoint analysis isolating the feature's value.

Damages experts synthesize the other factors into a rate opinion. Expert methodology is policed under Daubert — the Federal Circuit has repeatedly excluded opinions resting on non-comparable licenses, the rejected 25% rule, or unapportioned entire-market-value bases.

Evidence: Expert reports; underlying data and methodology.

The capstone factor: the amount a willing licensor and willing licensee would have agreed to in a hypothetical negotiation on the eve of first infringement, both assuming the patent is valid and infringed. This legal fiction sets the analytical frame for all other factors — the negotiation is deemed to occur at the date infringement began, with both parties acting reasonably and voluntarily.

Evidence: All of the above, synthesized as of the date of first infringement.

What are the Georgia-Pacific factors?

The Georgia-Pacific factors are 15 considerations, first catalogued in Georgia-Pacific Corp. v. United States Plywood Corp., 318 F. Supp. 1116 (S.D.N.Y. 1970), that courts and damages experts use to determine a reasonable royalty for patent infringement damages under 35 U.S.C. § 284. Section 284 guarantees a successful patent plaintiff 'damages adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer.' The factors structure the 'hypothetical negotiation' analysis: what royalty would a willing patent owner and a willing licensee have agreed to in a negotiation taking place just before the infringement began, with both parties assuming the patent is valid and infringed? The 15 factors fall into rough groups: market evidence (factors 1, 2, 12 — existing royalties for the patent, rates the infringer pays for comparable patents, customary rates in the industry); licensor's position (factors 3, 4, 5 — license scope, the patentee's licensing policy, whether the parties are competitors); value to the infringer (factors 6, 8, 9, 10, 11 — convoyed sales, profitability and commercial success, advantages over alternatives, nature and benefits of the invention, extent of use); and structural factors (factors 7, 13, 14, 15 — patent duration, apportionment of profit to the invention, expert testimony, and the hypothetical negotiation construct itself). Not every factor applies in every case; experts weigh the relevant ones and explain how each adjusts the rate up or down.

What is the hypothetical negotiation in patent damages?

The hypothetical negotiation (Georgia-Pacific factor 15) is the legal construct at the center of reasonable royalty analysis. It asks: what license terms would the patent owner and the infringer have voluntarily agreed to in a negotiation occurring on the eve of first infringement? Key assumptions that make it 'hypothetical': (1) Timing — the negotiation is deemed to occur just before the infringement began, so the parties negotiate with the information available at that time (though courts sometimes permit later 'book of wisdom' evidence about actual post-infringement results to inform the value); (2) Validity and infringement assumed — both parties negotiate assuming the patent is valid, enforceable, and infringed by the contemplated use, which removes the discount real-world licenses carry for litigation risk; (3) Willing parties — both the licensor and licensee are assumed willing to reach a deal, even if the real patentee would never have licensed a competitor. The output of the hypothetical negotiation can be a running royalty rate applied to a royalty base, or a lump sum. Because the construct assumes validity and willingness, royalty rates from the hypothetical negotiation often exceed rates in real-world licenses for the same technology — real licenses price in the risk that the patent might be invalidated or found not infringed.

What is apportionment and why does it dominate modern damages law?

Apportionment (reflected in Georgia-Pacific factor 13) is the requirement that patent damages compensate only for the value attributable to the patented invention — not the value of unpatented features, manufacturing efficiency, brand, or marketing. It dominates modern reasonable-royalty litigation, especially for multi-component products like smartphones, where a single device may implement thousands of patents. Key rules developed by the Federal Circuit: (1) Smallest salable patent-practicing unit (SSPPU): the royalty base should generally be the smallest salable unit that practices the patent — a component or subassembly — rather than the entire end product (LaserDynamics, Inc. v. Quanta Computer, Inc., Fed. Cir. 2012). (2) Entire market value rule (EMVR): the entire product can serve as the royalty base only if the patented feature drives consumer demand for the entire product — a demanding standard rarely satisfied for multi-feature products. (3) Even within the SSPPU, further apportionment may be required if the smallest salable unit itself contains substantial unpatented functionality (VirnetX, Inc. v. Cisco Systems, Inc., Fed. Cir. 2014). (4) Rate-base interplay: an expert cannot inflate the base and rely on a 'low rate' to compensate — using the entire market value as the base can skew the jury even with a small rate (Uniloc USA, Inc. v. Microsoft Corp., Fed. Cir. 2011). Practical methods experts use to apportion: comparable licenses that embody built-in apportionment; conjoint or survey analysis isolating consumer willingness to pay for the patented feature; regression analysis; cost or performance comparisons with non-infringing alternatives.

What happened to the 25% rule of thumb in patent damages?

The '25% rule of thumb' was a once-common shortcut under which damages experts started from the assumption that a licensee would pay roughly 25% of its expected profits from the patented product to the licensor, then adjusted up or down using the Georgia-Pacific factors. The Federal Circuit categorically rejected the rule in Uniloc USA, Inc. v. Microsoft Corp., 632 F.3d 1292 (Fed. Cir. 2011), holding that the 25% rule is 'a fundamentally flawed tool' that is inadmissible under Daubert because it fails to tie the royalty rate to the facts of the specific case — the same 25% starting point would apply to a groundbreaking pharmaceutical and a minor mechanical improvement alike. After Uniloc, damages experts must ground their rate opinions in case-specific evidence: licenses that are technologically and economically comparable to the hypothetical license; apportionment analysis isolating the patented invention's contribution; profitability and demand evidence specific to the accused products. Uniloc is part of a broader Federal Circuit tightening of damages methodology that includes LaserDynamics (SSPPU requirement), ResQNet.com v. Lansa (comparability scrutiny for licenses), and VirnetX v. Cisco (rejecting the Nash Bargaining Solution 50/50 split as a starting point for the same reason — an arbitrary rule not tied to the facts).

How do the Georgia-Pacific factors differ from lost profits damages?

Reasonable royalty (framed by the Georgia-Pacific factors) and lost profits are the two main measures of patent damages under 35 U.S.C. § 284, and they compensate different injuries. Lost profits compensate the patentee for sales it would have made but for the infringement — the profits the patent owner actually lost when the infringer took its customers. To recover lost profits, the patentee typically must prove the four Panduit factors (Panduit Corp. v. Stahlin Bros. Fibre Works, 6th Cir. 1978): (1) demand for the patented product; (2) absence of acceptable non-infringing substitutes; (3) manufacturing and marketing capacity to exploit the demand; and (4) the amount of profit it would have made. Lost profits are available only to patentees who sell competing products — a non-practicing entity or a university licensor cannot lose sales and therefore cannot recover lost profits. Reasonable royalty, by contrast, is the statutory floor available to every successful patent plaintiff regardless of whether it practices the patent. It compensates for the unauthorized use of the invention itself, measured by the hypothetical negotiation. The two measures can be combined: a patentee may recover lost profits on the infringing sales it can prove it would have captured, plus a reasonable royalty on the remaining infringing sales. Reasonable royalty awards are also the basis for ongoing royalties when courts deny injunctions after eBay v. MercExchange (2006).