The Federal Circuit recently heard oral arguments in the much-anticipated en banc review of EcoFactor, Inc. v. Google LLC, a case focusing on how courts evaluate expert testimony on patent damages. And, in particular, when a court should step-in to bar expert testimony that does not have sufficient factual basis for its conclusions. The arguments focused on whether EcoFactor’s damages expert improperly derived royalty rates from license agreements that contained performative non-binding “whereas” clauses stating EcoFactor’s belief about rates, while the operative provisions specified lump-sum payments. It seems clear to me that these provisions were added as elements of the patentee’s smaller license agreements in order to later be used in larger cases, such as the one against Google.
Judge Alan Albright (W.D.Tex.) admitted testimony from EcoFactor’s damages expert, David Kennedy, who derived a per-unit royalty rate from three prior EcoFactor settlement agreements to calculate a $20 million damages award against Google for infringing EcoFactor’s smart thermostat patent, U.S. Patent 8,738,327.
Kennedy’s testimony relied heavily on “whereas” clauses in prior settlement agreements stating EcoFactor’s “belief” that the lump-sum payments were based on a particular per-unit rate. However, the operative provisions of two agreements explicitly stated that the payments “are not based upon sales and do not reflect or constitute a royalty.” Despite these contradictions, Kennedy testified that these licenses reflected an agreed-upon per-unit rate that should apply to Google. Notably, Google chose not to present its own damages expert at trial.
The Federal Circuit initially sided with the patentee — upholding the verdict. However, Google persisted with an en banc petition that was granted in 2024. Oral arguments were held on March 13, 2025 before 10 of the 12 Federal Circuit judges. Notably missing was Juge Newman who was excluded from participating by her fellow judges. Professor Greg Dolin had filed a brief in the case arguing that it could not move forward en banc without her participation, but the court has seemingly ignored that contention.
Google’s Arguments
Ginger Anders of Munger, Tolles & Olson represented Google, opening with a clear framing of their position: The district court failed to fulfill its gatekeeping role under Rule 702 by allowing expert testimony that lacked sufficient factual basis. Notably, Anders did not suggest that the court needed to overrule or modify any of its precedent, but rather simply hold that the district court failed to apply the clearly stated standards:
Anders: The District Court abused its discretion in this case by holding that the reliability of Kennedy’s testimony, assigning the ex-royalty rate to the three lump sum licenses, was a matter for the jury to decide for itself. Rule 702 has long required the District Court to ensure reliability before admitting challenged expert testimony. Here, Kennedy’s testimony was clearly unreliable.
This unreliability argument centered on Kennedy’s statement that licensees had agreed to a specific royalty rate when they had not.
Anders: My friend on the other side said that this kind of unilateral clause, just asserting a rate in a license, is rare. It will not be rare in the future if this expert testimony is upheld.
Patentees could enter into small, lump-some licenses. They could insert unilateral assertions about what the rate they think should be. Licenses would have little incentive to protest in many cases, and they could use expert testimony to transform those unilateral assertions into authoritative expert testimony that the licensees actually agreed to this rate.
EcoFactor’s Arguments
Brian Ledahl of Russ August & Kabat defended the district court’s decision to admit Kennedy’s testimony, focusing on three main counterarguments:
Ledahl: I think broadly speaking, Google’s argument rests on what we consider to be three critical mischaracterizations of the issues in this case.
First, Ledahl argued that Google’s challenge was really about the weight of evidence, not admissibility. He contended that Rule 702 requires only that an expert’s opinion be based on sufficient facts—if true—not that those facts be undisputed. Here, the expert opinion is based upon a disputed fact about the licensing rate — something that should go to the jury.
Second, Ledahl emphasized that Kennedy didn’t solely rely on the rate in the licenses as his conclusion, but rather used it as one input in a broader analysis of the appropriate royalty rate.
And finally, Third, Ledahl pointed to supporting evidence beyond the licenses, including contemporaneous email correspondence between EcoFactor and Johnson Controls (one of the licensees) that showed negotiations around the rate. (Although that email had been excluded by the court).
The Judges’ Questions and Concerns
1. Threshold for “Sufficient Facts” Under Rule 702
Judge Hughes?: What if there are times where what you are relying on is just too, too slender of a reed to be something that anybody could regard as a sufficient fact? If they were just to isolate down on the Johnson license and we just had this one bare recital with nothing else to go on, I guess the question was why is that alone enough to be deemed a sufficient fact under these circumstances?
In my view, this question hits at the heart of the case: What constitutes “sufficient facts or data” under Rule 702(b)?
2. Distinguishing Weight from Admissibility
Judge Hughes?: We have to try to understand whether the expert’s basis for his opinion was based on sufficient facts. That’s step one before we get to the weight of the ultimate question of what could the jury decide was the actual rate. We have to figure out whether the expert’s allowed in the front end to even though crying on these contested questions. Therefore, 702 has some work to do.
My prediction for what its worth: The en banc court will likely vacate and remand without a sweeping ruling that Google is seeking. The basic problem is that the patentee’s expert Kennedy overstated facts about the royalty rate – indicating that it was “agreed upon” without providing a basis for that statement. It is OK to rely upon disputed facts in expert reports, but there must be a basis for such reliance. The court should have considered whether the royalty-rate assumption used Kennedy was at least reasonable, sufficiently supported, and not speculative. But, Judge Albright did not offer any reasoning for admitting the testimony.
One judge asked during oral arguments: “Should we send this back down and have the District Court review this in the first instance as far as providing the explanation for the rulings that he did do?” This sort of vacate-and-remand outcome is what I expect.
Is Google Simply Asking for More Efficient Infringement? | Patently-O