The Federal Circuit Continues to Evolve Its Daubert Gatekeeping Framework: Willis and Exafer

Need to know

• Two recent Federal Circuit decisions address questions the court left open in EcoFactor.
• In Willis, the court established that a district court’s gatekeeping assessment may be based on the full expert report, not just trial testimony.
• In Exafer, the court held that a royalty base anchored to an unaccused product can survive Rule 702 scrutiny if the expert can establish a causal connection grounded in record evidence.

---

The Federal Circuit’s en banc decision in EcoFactor, Inc. v. Google LLC, 137 F.4th 1333 (Fed. Cir. 2025), clarified the admissibility standard for patent damages expert testimony under Federal Rule of Evidence 702. The court held that where “the relevant evidence is contrary to a critical fact upon which the expert relied, the district court fails to fulfill its responsibility as gatekeeper by allowing the expert to testify at trial.” Id. at 1346. Two recent Federal Circuit decisions — Willis Elec. Co. v. Polygroup Ltd., 166 F.4th 1363, 1373 (Fed. Cir. 2026), and Exafer Ltd. v. Microsoft Corp., No. 2024-2296, — F.4th — (Fed. Cir. Mar. 6, 2026) — are now developing the practical contours of EcoFactor’s gatekeeping mandate. Together, they address two questions left open by EcoFactor: (1) What record the district court may consult to assess reliability and (2) whether a damages expert may use a royalty base comprising non-accused product sales.

Willis clarifies the reliability standard post-trial

In Willis, the Federal Circuit affirmed the district court’s denial of a motion for a new trial on damages following a jury verdict. 166 F.4th at 1367. The accused infringer, Polygroup, challenged admissibility on multiple grounds, including the argument that the plaintiff’s damages expert failed to adequately explain her methodology to the jury. Id. at 1377.

In rejecting Polygroup’s argument, the Federal Circuit clarified an important aspect of the gatekeeping inquiry. Specifically, the court explained that the district court must “assess[] the reliability of the expert’s methodology on the basis of all the evidence before the court” at Daubert. Id. At the time of the motion for a new trial in Willis, that record “included both [the plaintiff’s damages expert’s] testimony at trial and her expert report.” Id. The court explicitly contrasted this Daubert framework with a sufficiency-of-the-evidence challenge, which “would, in contrast, be based only upon the evidence presented at trial.” Id. at 1382 n.8.

Willis concretely illustrates the practical significance of this distinction. The plaintiff’s expert presented an income-based apportionment damages model that relied on the plaintiff’s profit margin for its own allegedly patent-practicing products. Id. at 1387. Polygroup argued that the plaintiff’s damages expert had failed to exclude from her analysis certain of Polygroup’s products that did not practice the asserted claims. Id. at 1387-88. The Federal Circuit noted that the plaintiff’s damages expert report expressly addressed this point in a footnote, with the expert explaining that she had excluded several models from her analysis because they did not practice the asserted claims. Id. at 1378 (quoting J.A. 4486 n.9). Relying on the expert report, the court held that testimony based on correct underlying data is not rendered inadmissible “merely because the expert could have described her characterization to the jury more precisely.” Id.

Exafer and the causal connection standard

Where Willis clarifies the sources a district court may consult when performing its gatekeeping function, Exafer addresses a different question: Whether a damages expert’s royalty base may be anchored to an unaccused product. The district court held that it may not, reading Enplas Display Device Corp. v. Seoul Semiconductor Co., 909 F.3d 398 (Fed. Cir. 2018), as categorically barring the use of unaccused products in a patent damages royalty base. Exafer, slip op. at 1.

The Federal Circuit reversed, holding that Enplas “does not prescribe the per se rule” that the district court applied and for which defendant Microsoft advocated on appeal. Id. at 7. Instead, the court explained, “[a] reasonable royalty is not necessarily unreliable under Rule 702 because it uses a royalty base associated with an unaccused product.” Id. The touchstone the court identified is whether there is a demonstrated “causal connection” between the accused features and the value of the unaccused product. Id. at 6. In Enplas, the unaccused products had “no causal connection” to the accused products, and therefore the unaccused products’ inclusion in the royalty base caused it to improperly comprise activities unrelated to the alleged infringement. Id. at 6 (citing 909 F.3d at 409-12).

In contrast to Enplas, the Federal Circuit found a “causal connection” in Exafer. The Exafer asserted patents “recite[d] systems and methods of optimizing network traffic paths.” Id. at 7-8. The plaintiff’s technical expert opined that the accused features reduced central processing unit (CPU) usage in Microsoft’s servers, thereby freeing the CPUs to host additional virtual machines (VMs). Id. at 6. The plaintiff’s damages expert then used the technical expert’s opinions “to value the Accused Features at the time of the hypothetical negotiation in terms of the additional VMs Microsoft would be able to host on its servers.” Id. The Federal Circuit held that this methodology is “tethered to the patented invention” and does not “expand Exafer’s patent monopoly to unpatented technology.” Id. at 8.

Practical takeaways

• The expert report is the admissibility record. Under Willis, the district court’s gatekeeping assessment draws on the full expert report, not just trial testimony. Analytical choices, data filtering decisions, and methodology disclosures made in the report — even if not fully explained to the jury — are part of the record the court considers in its gatekeeping role. Experts should therefore draft their reports not only with an eye toward jury persuasion, but with a Daubert hearing in mind as well.
• Daubert challenges must engage the full report, not just trial testimony. Because the admissibility record is broader than the trial record, a challenge limited to what the expert said on the stand may miss the methodological disclosure that saves the opinion under the Willis framework.
• A non-standard royalty base requires a technical foundation, not just a damages theory. Exafer confirms that a royalty base anchored to an unaccused product can, in some cases, survive Rule 702 scrutiny — but the expert must at least establish a causal connection grounded in record evidence. The damages model is only as strong as the technical causation opinion supporting it. If the technical foundation can be narrowed or rebutted, the royalty base may not survive.
• EcoFactor, Willis, and Exafer define a workable spectrum. Expert testimony is excludable where its factual premises are directly contradicted by the evidence on which the expert relied. Daubert challenges based on methodology disputes, however, are less likely to be upheld on appeal. Mapping a Daubert challenge to the right point on this spectrum will maximize its likely effectiveness.

For more information about the Federal Circuit’s evolving Daubert jurisprudence, please see “En Banc Federal Circuit Grants Google a New Trial in EcoFactor Case” and “Federal Circuit Signals Tension in Expert Testimony Jurisprudence: EcoFactor and Barry.”