“What is my patent worth?” It is a common question that IP portfolio managers and licensing executives hear from their clients.

As any IP professional knows, the answer is complicated and turns on numerous factors, including the scope of the claims and revenues they impact, the licensing history of the patent, and the rates that have been set in the industry. The U.S. Court of Appeals for the Federal Circuit and the district courts have further shaken up the valuation world over the last five years, issuing damages opinions that have excluded expert opinions for relying on non-comparable licenses or outdated licensing industry practices, or for failing to appropriately tie a patent claim to an accused product.

These litigation outcomes have an inevitable trickle-down effect that will impact current, ongoing licensing deals and valuation analyses.


Two lines of cases are reshaping patent damages law. The first has made two concepts—the “Entire Market Value Rule” (EMVR) and “Smallest Saleable Unit” (SSU)—part of the patent litigator’s lexicon.

The SSU concept was introduced by now-Chief Judge Randall Rader of the U.S. Court of Appeals for the Federal Circuit, sitting by designation in Cornell University v. Hewlett-Packard Co., 609 F.Supp.2d 279 (N.D.N.Y. 2009), and is a legal requirement that patent damages should be tied to revenues associated with the “smallest saleable unit” that practices a particular patent, rather than a larger product (with higher revenues) that may contain the SSU.

The EMVR requires that parties refrain from claiming the revenues of the larger product unless they can show that the patented feature or component is a major driver of the customer demand for that larger product. Federal Circuit case law in the 1980s and 1990s, including the en banc Federal Circuit in Rite-Hite Corp. v. Kelley Company Inc., 56 F.3d 1538, 1549 (Fed. Cir. 1995), articulated the EMVR concept in more or less its present form. More recent case law, such as the Federal Circuit’s Lucent Technologies Inc. v. Gateway Inc., 580 F.3d 1301 (Fed. Cir. 2009) opinion, has emphasized the EMVR is “a narrow exception” to the general rule that a patentee should not be able to claim royalties based on a larger product incorporating a patented component.

Together, the EMVR and SSU rules have led to more intense scrutiny on the measure of patent damages in cases where an asserted patent only covers part of an accused product.

The EMVR and SSU rules are a reaction to the proliferation of patents that are filed on components of increasingly complex devices, as well as the rise of patents that claim specific software features. A modern smartphone, for example, contains hardware components and software features that could potentially implicate the claims of thousands of different patents. These rules will substantially change the evidence that is required in litigation, to prove what the appropriate SSU is and whether the EMVR should apply, and will affect what past licenses are to be considered in litigation.

These rules are also at odds with at least some past licenses, since licensors often try to license patents against the products or services a licensee sells, rather than trying to divine what the SSU is for each patent that is subject to a license. If a licensed patent relates to a small part of a licensed product, then the parties could typically accommodate this by negotiating a low royalty rate for the patent. However, the SSU case law has suggested that ratcheting down a royalty rate against a larger product is no longer appropriate—the only relevant metric is the revenues of the SSU, unless the patent claim is so important that the EMVR applies.


A second trend, beginning with Lucent v. Gateway and ResQNet.com Inc. v. Lansa Inc., 594 F.3d 860 (Fed. Cir. 2010), requires courts to be stricter in making sure that past licenses truly are “comparable” to prove patent damages. Of course, the requirement that licenses need to be comparable to be relevant to a Georgia-Pacific analysis is not in itself new. However, case law post-ResQNet indicates that courts may generally be less lenient in admitting licenses that are not directly comparable.

This trend is a reaction to the fact that technology licenses are often for transactions that are more complex than the “hypothetical negotiation” for a license of an asserted patent in litigation.

Accordingly, courts may now be more likely to exclude past licenses that do not form an “apples to apples” comparison with a patentee’s assertion in litigation. For example, since ResQNet, courts have found licenses were not comparable because: They were portfolio licenses containing unasserted patents

They were procured to settle litigation

The licensor delivered non-patented technology, such as software, in addition to a patent license

They involved a lump sum for a small number of licensed products

They involved different technologies (e.g., wireless products were not comparable in a smartphone case)

They involved a lump sum in cases where the patentee sought a running royalty

These two trends—the increased emphasis on the SSU/EMVR and stricter attention to truly “comparable” licenses — have in some instances caused patent demands to be slashed dramatically or thrown out entirely by courts. License agreements that may have previously been admissible may no longer be, either because they do not cover an SSU or because they are no longer deemed “comparable.” Accordingly, licensing professionals going forward need to give additional thought as to how they structure their licensing efforts in order to maintain flexibility to claim maximum value in litigation.


The increased emphasis on the SSU means that licensors that wish their past licenses to be considered in litigation may start licensing against components, rather than end products. This will help undercut an opponent’s argument that the license should not be considered because the license is not tied to an SSU and therefore is inadmissible. Licensing parties may define the Licensed Product to be the smallest component that is covered by a patent claim, since the claim defines the scope of patent protection. And the parties may negotiate royalty rates and auditing protocols to track the SSU component.

Alternatively, if auditing a component’s revenues is not practical, the parties could eschew a royalty rate tied to product revenue in favor of a non-revenue-driven royalty measurement, such as a per-unit royalty, per-use royalty or other metric. Some district courts have found that royalty structures that do not vary with an end product’s revenues are not restricted by the Entire Market Value Rule.

To account for EMVR, a licensor may try to license its most significant patents through licenses containing a stipulation that the licensed patents cover a feature that drives customer demand. Therefore a royalty rate against the end-product revenues may be deemed appropriate.


Current law suggests that courts may exclude licenses that have a different scope (e.g., different patents or different technologies) or different economics (e.g., a lump sum versus a running royalty) than the reasonable royalty for asserted patents typically sought in litigation. This creates an opportunity for a licensor, to attempt to divide its efforts to negotiate licenses it wants to include in subsequent litigation from the licenses it wishes to exclude from future litigation damages analyses.

For example, a licensor that wishes to preserve the flexibility to argue for higher damages than the best royalty rate it can negotiate may consider trying cross-licensing deals, including other technology or patents in some licenses. Such additions may increase the chance that those licenses are found non-comparable, and either not considered in litigation or given less weight than more comparable licenses.

On the other hand, a licensor that wants to establish a predictable royalty rate for one or more patents together should consider licensing those patents together on standardized terms across licensees. If the licensor wishes to use a royalty structure other than a running royalty, such as a lump sum structure, it should consider including recitals that describe how the alternative royalty structure was calculated and what assumptions were made. Such recitals may be helpful to a litigation damages expert to “convert” the license into a litigation comparable.


Patent damages continue to a be a hot topic in district court and before the Federal Circuit, and we can anticipate that further refinements to the EMVR, the SSU, and comparability analyses will emerge in the years ahead. Licensing professionals would do well to keep abreast of these developments as they have had, and will continue to have, an acute impact on patent valuation.

Siddhartha Venkatesan is a partner in Orrick, Herrington & Sutcliffe’s Silicon Valley office, where he litigates high-stakes patent and trade secrets claims and also assists clients with IP licensing and counseling matters. He was recently named one of the Recorder’s 2013 “Lawyers on the Fast Track.” He can be reached at svenkatesan@orrick.com or 650-614-7456.

In Practice articles inform readers on developments in substantive law, practice issues or law firm management. Contact Greg Mitchell with submissions or questions at gmitchell@alm.com.